Acknowledgments

An earlier version of this paper was presented at the 2001 Annual Meeting of the American Sociological Association, 18–21 August, Anaheim, California. I am grateful to D. J. Hatfield, Michael Kennedy, Ming-cheng Lo, Peter Zarrow, an anonymous reviewer and the editors of CSSH for their helpful and encouraging comments. I also wish to extend my gratitude to David Akin for his assistance in preparing the final manuscript. The study was in part supported by the ROC National Science Council in Taiwan (NSC-89-2412-H-002-028). All Chinese names and characters are romanized in pinyin, unless there is a common usage that has gained wide popularity (e.g., Taipei, Chiang Kai-shek). To follow the convention in Taiwan, however, a dash is inserted between the second and the third characters of the given name where applicable, although it is acknowledged that this practice does not conform to the pinyin rule developed by the People's Republic of China. As will be made clear, problems concerning romanization are themselves a contested terrain which this paper intends to investigate.

In his groundbreaking Nationalism Reframed, Rogers Brubaker challenges conventional understandings of nations and nationalism by advancing a distinctive, if not innovative, approach to the subject. Drawing on recent theoretical developments that problematize the realist ontology implicitly assumed in previous literatures, Brubaker calls for an institutionalist approach to the study of nations and nationalism. As he points out, nation and nationhood can be better understood not as substance but as institutionalized form, not as collectivity but as practical category, and not as entity but as contingent event (1996:18). He then employs this approach to analyze the breakdown of the Soviet Union. According to Brubaker, nationhood and nationality were institutionalized in the Soviet Union in two different modes: political-territorial and the ethno-personal. While the incongruence between these two modes led to tensions and contradictions within Soviet society, the dual legacy of such an institutionalization, manifesting itself as unintended consequences, eventually shaped the disintegration of the Soviet Union and continues to structure nationalist politics in the successor states today.

Illuminating as it is, Brubaker's analysis opens up a new ground for research that is yet to be fruitfully explored. More importantly, I contend that the analytical power of such an approach has not been fully realized even in his own work, as there emerges from it a new set of research questions that can be further pursued. For instance, what are the embodied forms of those institutions that constitute a specific nation? To whom, in what manner, and in what contexts do these institutions perform their functions? What are the consequences if such institutions are challenged, undermined, or even fail? To answer these questions, this paper extends analytical strengths of the institutionalist approach by examining recent nationalist politics in Taiwan. The study is, then, not simply another application of the institutionalist approach to a specific case; rather, the case itself may shed new light on our understanding of nations and nationalism by contributing to ongoing theoretical development. Simply put, the major argument holds that the existence of a nation is hinged upon an ensemble of intersecting institutions that can be classified into civic-territorial and ethno-cultural types; both types of institutions, moreover, have to articulate on both domestic and international levels. If either type of institution fails to articulate at either level, the existence of the nation will be jeopardized, and an identity crisis may ensue. This was precisely what happened in Taiwan during the 1990s with the escalation of nationalist politics. Previously, I have shown how political institutions in Taiwan—officially (yet rarely) known as the “Republic of China” (ROC)—failed, and led to chaotic national identifications (2002b). In this paper, I shall instead focus on the crises of ROC's cultural institutions, which eventually contributed to an escalation of identity crises in Taiwan.

Before I set forth my analysis, a brief historical sketch will be helpful to those who are not familiar with Taiwanese history. Prior to the year of 1889, when it was made a province of the Chinese Empire by the Qing Dynasty, different parts Taiwan had been occupied by the Dutch, the Spanish, the Americans, and the Japanese, some of whom established administrative offices for purposes of short-term rule. In 1895, Taiwan was ceded to Japan by the Qing Dynasty and was colonized for the next fifty years. While China was undergoing a modern nation-building process during the Republican period (1911–1945), Taiwan was becoming “Japanized” under the Japanese colonialism.¹

Against this backdrop, the election of DPP-nominated Chen Shui-bian as President of the ROC in 2000 marked a watershed in Taiwan's history. Not only did it end the fifty-five-year rule of the KMT on the island, but it also shifted the momentum of ROC nation-building. Given the rupture brought up by this event, my analysis in this paper will be divided into two parts: the first will focus on how cultural institutions of the ROC nation were built up by the KMT state during the period from 1949 to 2000, while the second will investigate the post-2000 situation under DPP rule. This analytical strategy is necessary for two reasons. First, as institutional analysis emphasizes path dependence of historical development, Chen's new government inherited enormous institutional legacies from the KMT state. Suffice it to say that most of the signifying institutions of the ROC—its national title, national anthem, and national flag, each of which the proponents of the Independence Movement had long vowed to do away with—were kept intact after Chen's inauguration. This indicates the analytical power of the institutionalist approach—many institutional crises and predicaments that the KMT created are still haunting the DPP government. Moreover, since the central argument of this paper holds that the recent identity crisis in Taiwan has deep historical roots in ROC institutions, one can hardly comprehend the situation after 2000 without understanding how institutions of the ROC came into existence in the first place. The main body of my analysis, therefore, will focus on the situation before 2000, and most references to the ROC government refer to the KMT state. The situation after 2000 under DPP rule will be discussed in the second part of the paper, in which I shall show how institutions of the ROC have both enabled and constrained DPP's pursuit of an independent nationhood.

I shall first sketch a brief analytical framework of the institutionalist approach to nations and nationalism by integrating Brubaker's original formulation with theories of new institutionalism developed by John Meyer and others. I will show that there exists an institutional repertoire that presumes that every nation should have a distinctive culture to actualize its existence in the symbolic field. Next, I shall focus on two main institutional sites of national culture—namely, materialized objects and language—to show how cultural institutions of the ROC were challenged in international settings, to the effect that their failures eventually fueled the identity crisis in Taiwan. I shall then examine the post-2000 situation by analyzing how institutional legacies of the ROC have been shaping Taiwan's nationalist politics to date. The concluding section will explore the theoretical implications of this case, and highlight the analytical strengths of the institutionalist approach in the light of Pierre Bourdieu's notion of “rite of institution.”

having a culture: an “imperative” for the nation-state?

In his original formulation, Brubaker distinguishes between two aspects of institutionalization of nationhood and nationality: one concerning the territorial organization of political administration, the other the classification of persons (1996:30). To take this a step further, I argue that such a distinction of institutions corresponds roughly to two “ideal types” of the nation-state: the civic-territorial model, and the ethno-cultural model (Smith 1991:82; Brubaker 1992). While such a distinction might seem banal today, it nonetheless has important implications for our understanding of the institutions of the nation. To be sure, no nation has, in reality, been founded solely on either of the two models. For those nations built on the ethno-cultural model, there are nonetheless institutions that define these nations in civic-territorial terms. Conversely, nations based on the civic-territorial model more or less contain ethno-cultural elements in defining their nationhood. The nation, regardless of which model it is built upon, can be analyzed as an ensemble of intersecting institutions that can be classified into two types: civic-territorial and ethno-cultural. The former concerns political organization of the nation regarding such matters as sovereignty, territoriality, and citizenship; the latter involves (re)presentation of the nation in the symbolic field, including national culture, national history, schemes of classifying people, and the like. The weaving together of these two types of institutions actualizes the existence of a specific nation. Furthermore, on the worldwide level, these two types furnish the grids of classification schemes on political/territorial and cultural/cognitive maps. The two are highly correlated, but neither can be reduced to the logic of the other.

According to the property of relativity that we learn from the theory of new institutionalism, whether a social pattern or practice can be seen as an institution depends on the context of our analysis (Jepperson 1991). In the context of nationhood and nationality, the property of relativity brings us to what is known as “institutionalist theory of world polity” developed by John Meyer and his colleagues (Meyer et al. 1997; Thomas et al. 1987). Institutions of nationhood and nationality have to perform in two relative contexts: at the domestic/national level, and at the global/international level. Brubaker in his study dwells primarily on the domestic/national level; in contrast, this study will emphasize the global/international level, with a focus on ethno-cultural institutions. To be sure, there exists an international repertoire, or “world culture,” as Meyer and his colleagues put it, that helps to organize actors and agents in world society. Regarding national culture, Löfgren (1989:21–23) has usefully distinguished three analytical levels: an international cultural grammar of nationhood, a specific national lexicon, and a dialect vocabulary. While most previous studies of national culture deal mainly with the latter two, here I will concentrate instead on national culture on the first level—the international cultural grammar of nationhood, or, to phrase it in the vocabulary of institutionalism, the world-level institutional scripts for national culture.

Indeed, that every nation has its own “culture” is widely held to be commonsense, taken for granted as a “natural fact.” An “invented tradition” of the modern era, the very notion of “national culture” is itself a consequence of globalization conceived broadly (Hobsbawm 1983; Fox 1990; Wallerstein 1990). Not only has it become axiomatic that every nation must have its own standardized, homogeneous culture (Gellner 1983), but national culture itself is considered a manifestation of the existence of the nation: “We are a nation because we have a culture” (Handler 1988:153). Even in cases where the nation is founded primarily on a civic-territorial model, there is nonetheless a tendency to construct a reified national culture as a collective representation for the community. For those “new nations” that follow in the footsteps of their Western precursors, the construction of national culture is part of the “quest for modernity” (Foster 1991:237). Following Western (including anthropological) definitions of culture, nations and peoples around the world have been endeavoring to maintain a unique cultural identity by importing “Western technical routines to manage their objectified cultures” and by promoting “their ‘self-image’ internationally in an effort to woo the economically crucial tourist trade.” In short, “everyone wants to put his own culture in his own museums” (Handler 1987:139).

The international grammar of national culture is crystallized in the discourses of international institutions such as the United Nations Educational, Scientific, and Cultural Organization (UNESCO). In their report at the World Conference on Cultural Policies in 1970, members of UNESCO assert, on the one hand, that cultural differences are acknowledged and tolerated from a pluralist stance of universal humanity, and, on the other, that one cannot speak of cultural autonomy without reaffirming the fundamental concepts of national sovereignty and territorial independence (Tomlinson 1991:70–75). In other words, cultural differences between nation-states are taken for granted and necessary, as the sovereignty of nation-states is extended from the political to the cultural field under the veil of guarding cultural autonomy.

The sovereignty of nation-states in the cultural field manifests itself in those institutional measures through which national culture is created. As Wallerstein (1991) argues, national culture emerges as a result, whether intended or not, of the allocation of resources by the state. Since each states allocates its resources to different sectors within society, it becomes “natural” that, in the long run, each nation-state will have its own distinctive “national culture,” even if such a distinctive national culture has never existed before. As a result, national culture now serves as an institutionalized paradigm worldwide for collective representation, and as the grid of classification schemes on the international level.

It is under such international scripts of national culture that the ROC nation on Taiwan comes up against a crisis in the symbolic field. The ethno-cultural model of Chinese nationalism calls for the institutionalization and standardization of a national culture. Up to the 1990s, the ROC on Taiwan claimed, both domestically and internationally, to be the “true heir” and guardian of traditional Chinese Culture. This was particularly so during the 1960s and 1970s when the CCP launched the devastating Cultural Revolution in China. The KMT took this as an opportunity to promote itself as the guardian of Chinese culture, and hence the genuine heir of the Chinese nation. The now widely acknowledged fact that the KMT state has imposed a Chinese Culture on Taiwanese society, therewith creating a Chinese identity, has been studied in previous literature.²

making the nation visible: cultural objects and national heritage

Interrogating the elusive concept of “culture” from a world-system perspective, Wallerstein (1990:33) raises the question: “What is the evidence that any given group [such as a nation] has a ‘culture?’” He contends that we can, at most, argue for a statistically significant relationship between group membership and certain behaviors or objects, but it is obvious that surely not all presumed members of such groups act similarly or have the same traits (Wallerstein 1990:33–34). In other words, the idea that every nation has its own distinctive, homogeneous culture is no more than a myth.

While Wallerstein's critique of national culture appears quite revealing, things look different if we apply an institutional perspective to the same question: “what the evidence is that a nation has a culture.” As pointed out above, a nation is best understood neither as a substantive entity nor an aggregate of individuals, but rather as institutionalized political and cultural forms. It is through institutions that a nation is constructed as a “real entity.” By the same token, it is through institutions that a “culture” is formulated as a real object that an individual nation owns. In other words, “possession of a heritage, of culture, is considered a crucial proof of national existence” (Handler 1988:142). Hence, responding to Wallerstein's question from an institutional perspective, the most obvious evidence that a nation does have a culture lies in those visible/audible practices and materialized objects that, while intended to promote the idea/ideal of the nation, have been organized through, and woven into, cultural institutions such as museums, galleries, schools, ceremonies, rituals, landscapes, historic sites, public displays and performances, mass media, and so on. In what follows, I shall examine the politics of national culture within the contexts of these cultural institutions.

“cultural assets”: cultural capital for the nation

In the official English publications by the Council for Cultural Affairs of the ROC, the Chinese term “wenhua zichan,” which is much like the more conventional notions of “cultural patrimony” or “cultural heritage” in English, is literally translated into “cultural assets.” Whether a mistranslation or an intentional coinage, this formalized term keenly reflects what ROC cultural officials have in mind when they speak of culture: culture, after all, is regarded as a kind of “asset” endowed with values and productivity.

In contrast to the accounting concept of “liabilities,” “assets” imply valuable properties necessary for running a business, for production and reproduction. “Cultural assets” in its literal sense implies a kind of cultural capital—to borrow from Bourdieu's (1984) famous concept³

In the Chinese context, culture is above all considered an “asset” for the nation, since the so-called “five millennia of civilization” have left the nation abundant traditions and legacies that the Nation can rightly boast about. “The Old Nation of Culture” (wenhua guguo) or “the Big Nation of Culture” (wenhua daguo) is rhetoric that Chinese nationalists often employ, not only to characterize the Chinese nation, but also to assert the nation's value as topping other nations. In making the Cultural Assets Preservation Law in the 1980s, a legislator expressed his self-reflections on the “global trend of preserving cultural assets”:

Behind this statement lie two features of constructing national culture. First, there was a process of mimicry or emulation: We (the ROC) need to have a culture because every nation has one. A nation by definition, as it were, must have an objectified culture, and such a culture was most readily found in such institutions as museums. While this mimetic process is reminiscent of Anderson's (1991) argument that the nation as a “modular form” of imagined communities was “pirated” worldwide in non-Western societies, the process itself is also an embodiment of institutional isomorphism of the modern nation-state that John Meyer and his colleagues have pointed out (Meyer et al. 1997). Second, cultural assets in the Chinese context were above all nationalistic, since these assets were found endangered, missing, incomplete, or pillaged due to Western invasions. It was considered the responsibility of the nation not only to preserve but also to rescue or redeem these cultural assets.

The ROC's official definitions classify cultural assets into five major categories: (1) antiquities, (2) historic sites, (3) national arts, (4) folkways and related materials, and (5) natural-cultural landscapes (Ziran wenhua jingguan [Council for Cultural Affairs] 1995:67). Among these, the first category—antiquities—best characterizes the ROC's “possession” of Chinese culture, and the National Palace Museum has been the most important institution for the preservation and display of historic relics and antiquities as “national treasures.”

The possession of historic relics and antiquities has a highly symbolic meaning peculiar to the Chinese context. In the Chinese tradition, relics imply or even equate orthodoxy. Those who possess the relics and antiquities from the preceding dynasty can claim themselves to be its orthodox successors. As a former president of the National Palace Museum recounted, the takeover of past relics and antiquities had throughout Chinese history been one of the most important tasks whenever there was a regime change; therefore, “the first thing we had to take care of after [the KMT] seized power was the expropriation of relics and antiquities” (quoted in Yang 1994:164). The antiquities and relics in the National Palace Museum were formerly stored in various palaces of the preceding Qing Dynasty located in Beijing, Shenyang, and Chengde in mainland China. In 1948, when the KMT was gradually losing the civil war to the CCP, Chiang Kai-shek gave orders to transport this Chinese cultural patrimony to Taiwan, and the shipment of the relics was given a higher priority than the retreat of the military and officials (Yang 1994).

The transportation of these relics to Taiwan, characterized as “theft” by both Taiwan Independence activists and the PRC because the act was done half secretly, has proven critical. Prior to 1949, there had been few Chinese relics on Taiwan, considering that it was a periphery incorporated as late as the seventeenth century by the last dynasty of the Chinese empire. Under such circumstances, the possession of these relics and antiquities is crucial to the ROC's claim to Chinese nation, since these material objects are the most visible and immediate manifestations of Chinese culture. The National Palace Museum, where most of these relics and antiquities are stored and displayed, has for this reason become one of the most important cultural institutions of the ROC. With a present accumulation of 640,000 items, it is recognized to have one of the world's best collections of Chinese art, from ancient bronze urns to scroll paintings and snuff bottles. While it was considered a “must-see” for foreign visitors to Taiwan, in later years the ROC government also sought to promote these “cultural assets” on the international stage. An unprecedented U.S. tour of 452 of the Palace Museum's finest pieces took place from March 1996 through April 1997. The exhibition, entitled “Splendors of Imperial China,” made stops at the Metropolitan Museum of Art in New York, The Art Institute of Chicago, The Asian Art Museum of San Francisco, and The National Gallery of Art in Washington, D.C., attracting some 900,000 viewers in total (Government Information Office 1998:351). From October 1998 through January 1999 a similar exhibition titled “Mémoire d'Empire” debuted in Paris and toured Europe.

While these tours and exhibitions were celebrated as successes of “cultural exchange,” these cultural events have been tinted by political hues. Since Taiwan's diplomatic situation was deteriorating in the 1990s, these efforts of so-called “cultural exchange”—characterized as “cultural diplomacy”—were explicit attempts by the ROC to surmount its international isolation. For Taiwanese society, these activities also provided psychological compensation for its long exclusion from international communities, and each received extensive coverage in the Taiwanese media. Commentators and reporters hailed these events in nationalistic tones, praising them for returning Taiwan to the spotlight of the international stage by “grabbing a million eyes” worldwide (Minsheng Daily, 22 Oct. 1998; also Fu 1997). However, the nationalistic implications of these events also generated controversies, both internal and external, regarding the ownership of culture.

who owns what?: contested property rights and the endangered ownership of culture

Before the touring exhibition “Splendors of Imperial China” set out on its extraordinary journey to the United States, it had sparked hot debates in Taiwan. Artists, intellectuals, legislators, and Taiwan Independence activists competed to criticize the ROC government on various scores, some of them mutually contradictory. Some argued that the National Palace Museum had harmed Taiwan's national dignity by making too many concessions to American exhibit organizers; others insisted that certain pieces were too precious to be sent abroad; while still others contended the artifacts did not represent Taiwanese culture and thus the exhibition amounted to misrepresentation. Some defended the exhibition as worthwhile, since it would help Taiwan counter its international isolation.

These acrimonious debates were highly charged with nationalist sentiments, but the most interesting among them involved the ownership of the artifacts. Some intellectuals and politicians worried that it was too risky to send these “national treasures” overseas, since the PRC might seize the opportunity to pressure the United States to “return” them to China (Independence Morning Post, 12 Jan. 1996). On the other side, some Taiwan Independence activists maintained that, because these so-called “treasures” were originally “stolen” and taken to Taiwan by Chiang Kai-shek and his exile KMT regime, it was now time to return them to the PRC to show that Taiwanese culture has nothing to do with Chinese culture (China Times, 13 Jan. 1996).

The implicit common ground between these two views is a shared questioning of Taiwan's problematic possession of these cultural objects. Who, after all, is the “proper” owner of such a culture? Both camps take up one common assumption, implicitly or explicitly: these “treasures” putatively belong to the PRC, the now-recognized China. But why is that so?

As is the case with individual possessions, once national culture has been reified and institutionalized as a real entity, problems concerning “ownership” and “property rights” emerge. A nation must have a culture, and the ownership of such a culture has to follow the principle of property rights. Presumably the subject that owns such a culture is the collective “nation,” that is, the ensemble of its people and territory. However, two problematic scenarios may occur if there is a change in either people or territory. One is that cultural objects of a nation may have been inappropriately acquired due to flows of people, as is the case of the “national treasures” in the National Palace Museum. The other is that the emblematic tokens of a certain national culture may not be in the possession of the nation due to territorial constraints. This can cause another set of problems.

The mystical bonds that tie the nation to its culture exist not only in artifacts, but also in Mother Nature. As with the case of religion, objects or spectacles in the natural world can be endowed with sacred aura or a holy status, thereby becoming totemic or emblematic of a national culture. The Yangtze River serves as a good example. Entwined with historical events and endowed with rich literary legacies, the Yangtze—the longest river in China and the third longest in the world—has been held as a natural spectacle that symbolizes the long-standing history of Chinese civilization. In 1995, the ROC Government Information Office funded the publication of a book by environmentalist Han Han entitled Women Zhiyou Yitiao Changjiang (We have only one Yangtze River). This immediately brought criticisms from pro-independence cultural elites. In a public symposium they asked: Who are “we”? Since the writing project was funded by the government, “we” putatively referred to the people under the governance of that authority. However, “we” (namely, the collective of the ROC) do not actually have the Yangtze River, but the PRC does. These cultural elites asserted that the publication of the book not only reflected the “schizophrenia” of the ROC government, but also entailed “a violation of property rights” since the discourse was based on “improper ownership” (Independence Morning Post, 12 July 1993).

Indeed, the problem of ownership due to territorial constraints has also restricted the stock of those ROC's cultural assets under the category of historic sites. None of the famous, albeit stereotyped, historic sites known as “essential” to Chinese history—for example, the Great Wall, or the Emperor's palaces or tombs—is “owned” by the ROC. On the contrary, what the ROC “has” in the way of historic sites simply reflects the embarrassing fact of its shallow historical depth and cultural hybridity. Sites in Taiwan, apart from archeological ones,⁴

In fact, the preservation of such sites has always been contested, since many policy-makers saw these imperialist and colonial legacies as “national shames” (guochi) that ought to be erased rather than preserved. To most mainlanders who bear the historical memory of fighting the Japanese during the Second World War, Japan is their biggest common enemy and almost exclusively the defining “Other” of the Chinese nation. It is an unbearable irony for them to see that Japanese imprints on this island—for instance, old generic buildings designed by Japanese architects—should be preserved as the “cultural assets” of the ROC.⁵

Indeed, the insufficiency of “Chineseness” on Taiwan is perceived by both insiders and outsiders. In addition to the Yangtze River, the Yellow River and the Great Wall are among the “national emblems” of Chinese culture that have been made popular in global media and are beloved by Western audiences and tourists (Waldron 1993; Kim and Dittmer 1993). However, none of these emblems is possessed by the ROC. Taiwan's problematic “Chineseness” can be gleaned from the general opinions foreign visitors hold about the country. According to an official survey conducted by the ROC Tourist Bureau in 1990, the top reason that foreign tourists gave for visiting Taiwan was to see Chinese culture. Ironically, the biggest disappointment that these tourists said they felt after leaving the island also concerned Chinese culture. In other words, foreign visitors were attracted to Taiwan in order to see Chinese culture, but in the end were disappointed to find that there was little of it to see. It was suggested that, to foreign visitors, Taiwan simply did not appear “Chinese” enough, nor was the island's “Chineseness” convincingly authentic. Even a reporter writing about the survey had to painfully admit that, except for the National Palace Museum, there was not much “Chinese culture” on the island (United Daily News, 19 Oct. 1990). This irony should not surprise us. If we combine this with other aspects of cultural institutions with which a Chinese nation is supposed to be defined, we find that Taiwan's problematic “Chineseness” is manifested not only in its deficient stock of cultural objects and national heritage, but also in its incompetence to define authenticity and orthodoxy. This can be best illustrated through the case of language.

language: visible speech and the struggles for orthodoxy

The role of language in building a nation simply cannot be overemphasized. Viewed by nationalists as a major carrier of culture, language is widely considered essential to the modern definition of nationality, as well as to the popular perception of it (Hobsbawm 1990: 59). It is one of the most significant diacritics that immediately distinguishes social groups from one another. Even a slight nuance or variation in language, be it in oral or written forms, can suffice to make people aware of their different origins or belongings in terms of region, race/ethnicity, or nationality. While the role of philologists, grammarians, and lexicographers in standardizing a national language is emphasized in the construction of homogeneous linguistic communities (Anderson 1991), language itself has been found to be a central mechanism through which states make nations (Laitin et al. 1994).

In China, language has been a quintessentially nationalist concern, not only in the sense discussed above, but also in another way that is somewhat idiosyncratic to the case. As early as the turn of the century, when China was threatened by Western powers, modernist intellectuals in China reflected on the backwardness of their traditional culture and found language responsible for having impeded the progress of their nation. For one thing, the Chinese script, widely regarded as perhaps the only major ideographic system still in use in the world, was incompatible with Western Latin alphabets, and this was considered by modernist intellectuals to be a manifestation of the “backwardness” of Chinese civilization. In this way the writing system was blamed for having deterred the Chinese people's absorption of new, Western, modern knowledge and technologies that would have made China a stronger nation.⁶

As Hobsbawm perceptively observed, “[l]anguages multiply with states; not the other way round” (1990:63). This is precisely the case for the Chinese language. Since there have been two states since 1949, there are arguably two distinct Chinese national languages. Although both the PRC and the ROC adopted Mandarin as their official language, they have followed very different paths toward language institutionalization, so much so that the two Mandarins now differ significantly from each other in their written scripts, systems of romanization, and lexicons and pronunciation:

(1) Written scripts: As noted, the ultimate goal of Chinese language reform was to make it simple to write, preferably in Latin alphabets. In 1956 the CCP launched a first stage of simplifying Chinese characters, and a second stage followed in 1977, with the ultimate goal of replacing Chinese characters with Latin alphabets. Although the second stage was officially abandoned in 1986, simplified characters implemented during the first stage had successfully taken root in the PRC. In contrast, on Taiwan a quite different process occurred, as the KMT state insisted on the use of traditional characters for ideological reasons. As a consequence, there now exist two Chinese writing systems that some consider “non-reciprocally intelligible”⁸

(2) Romanization: This refers to methods of converting words or characters written in non-Roman scripts into Latin alphabetical scripts. For Chinese there have been numerous romanization systems. The situation is further complicated by dialectal variations because romanization transcribes the sounds rather than meanings of Chinese characters. For Mandarin alone, more than one hundred different schemes of romanization are known, some developed by Westerners, some by Chinese (Lo and Miller 1991:221). Among these, the Wade-Giles system, named after two Englishmen, was the system most widely adopted in Western scholarship and media.

As I have said, the PRC has been strongly motivated toward language reform, initially with the intention of transforming the Chinese script into orthographic transcription in Latin alphabets. In 1958, the PRC implemented the Scheme for Chinese Phonetic Alphabets (Hanyu pinyin fang'an, henceforth pinyin) as the official standard for romanizing Chinese characters. While the PRC simultaneously banned the Wade-Giles and other systems in China, pinyin has gradually become the international standard for romanizing Chinese characters. On the other hand, there has been no equally institutionalized romanization system in Taiwan—in addition to the Wade-Giles system, many other variants are used. In January 1986, in response to the worldwide spread of pinyin, the ROC's Ministry of Education announced the Second Form of Mandarin Phonetic Symbols (Zhuyin dier shi, hereafter MPS II). Although this system is meant to be an official standard of transliteration, it was never implemented in Taiwanese society, let alone known to the international community.

(3) Lexicon and pronunciation: It has been observed that there exist remarkable differences between two Mandarins regarding pronunciation, diction, and idiomatic expressions (Bosco 1994:395). To begin with, the pronunciations of some Chinese characters are standardized differently. For instance, shei (who) in Taiwan is standardized as shui in the PRC, and zhanshi (temporarily) as zanshi. Moreover, Taiwan's Mandarin, greatly influenced by the phonology of the Taiwanese dialect (Holo), tends to lose palatalization in retroflex initials, making it sound distinctive from the Beijing standard version of Mandarin. In addition, Taiwan's Mandarin has incorporated many lexical units from Holo, some of which can be traced back to Japanese as they were coined during the colonial period. The incorporation of Taiwanese Holo into Mandarin was furthered by the rise of popular culture in the late 1980s and became widespread via booming cultural industries. On television and in newspapers, one can find numerous Taiwanese terms being formalized to become an integral part of Mandarin. These terms and expressions are rather unintelligible to Mandarin speakers (or even Holo speakers) outside of Taiwan.¹⁰

These three major linguistic disparities had immediate effects on Taiwanese nationalist politics. Pro-unification elites and state officials expressed concern that these disparities in language would become impediments to any future unification of China and Taiwan.¹¹

The Chinese language standardized by the PRC has gradually won overall dominance over that of the ROC. To begin with, the total number of users of simplified characters is overwhelmingly greater than those who use traditional characters. At one time some might have argued that simplified characters were only used in the confined territory of the PRC, but the situation has changed, as Chinese characters are becoming ever more visible globally with increasing flows of people, commodities, and culture from and through the PRC. The politics of Chinese language has now overflowed territorial boundaries into transnational arenas. There have been debates within overseas Chinese communities regarding whether their descendants in North America should learn Chinese in simplified or traditional characters (United Daily News, 1 Dec. 1996). On 1 July 1992, the PRC's official mouthpiece, People's Daily (Renmin Ribao), changed its overseas edition from traditional characters to simplified ones, claiming that “the condition of using simplified characters [overseas] has basically matured, as simplified characters have become an accepted fact among the absolute majority of Chinese people in the world.” In response, the ROC government made extra efforts to countercheck what one ROC high official called the “globalization of simplified characters.”¹²

Regarding romanization, the situation is even worse, since Taiwan is facing more institutional pressures from outside. Internationally and in terms of institutions, the pinyin developed by the PRC has not only gained wide acceptance in the West but has also become a prescribed international standard. In 1977, the United Nations Conference on the Standardization of Geographical Names resolved to adopt pinyin as the standard for Chinese geographical names. In June 1979, the United Nations Secretary-General further notified its member states to use pinyin in all official documents, including pacts, agreements, and legal papers.¹³

In Taiwan, by contrast, as of 2000 there had been no effectively implemented standard for romanization. The international standard of pinyin was never formally adopted in Taiwan, nor was the KMT state motivated to institutionalize a consistent standard of its own. In the 1990s, with the rapid globalization of Taiwanese society, romanization became a more pressing problem since increasing interaction with the outside world necessitated an institutionalized standard for effective communication. The romanization of street names best illustrates this. In the past, signposts in Taipei were shown only in Chinese, but with the growth of the population of foreigners who did not read Chinese, it was found necessary to supply romanized street names. This raised a thorny problem: which romanization system would be used? Since romanization of street names was meant to guide foreigners, it would be more sensible to use the pinyin that had been adopted internationally. However, seeing that pinyin was developed by the rival PRC state, to adopt such a system implied compliance, at least symbolically, to the authority of the PRC, which, in turn, was considered harmful to the “national dignity” and even the state sovereignty of Taiwan. However, since the ROC's official system of MPSII was barely known to the general public in Taiwan, let alone to the outside world, to adopt MPSII would confuse those foreigners already familiar with pinyin. The Taipei City Government, then run by the opposition DPP, finally decided to use an eclectic “Tongyong (literally “general”) Pinyin System,” purposefully invented by a social psychologist at Academia Sinica. Although it was claimed that this system was compatible with pinyin while preserving Taiwan's national dignity, the policy suffered harsh criticisms when it was announced.¹⁵

The Chinese language institutionalized by the PRC in terms of simplified characters and pinyin has gained further popularity through the implementation of the Chinese Proficiency Test, also known as HSK, an official standardized test to evaluate the Chinese proficiency of non-native Chinese speakers. The global dominance of PRC's pinyin system has also weakened ROC's authority in defining and introducing Chinese culture on the international stage. As most geographical and historical names of China are now understood by foreigners in pinyin (notable examples being the shifts from “Peking” to “Beijing,” and the last dynasty “Ch'ing” to “Qing”), ROC's resistance to the adoption of pinyin made its presentations of Chinese history or interpretations of Chinese culture in Western languages somewhat obsolete or unusual. Causing confusions from time to time, this peculiar situation has drawn the attention of the Western media (see Washington Post, 6 Nov. 1998).

We can further explore the interplay between the state and culture at this conjuncture. Following Ferdinand de Saussure (1959), the relationships between phonetic signs (the signifiers) and their pronounced sounds (the signified) are arbitrary. In the case of romanization of Chinese characters, in which an extrinsic signification system is to be linked to an existing language, the state appears to be the best “arbitrator” to institutionalize such a relationship. One of the many tasks assumed by the state in its construction of a homogeneous (at least putatively) nation is to define an authoritative paradigm for a standardized culture. Language being a major carrier of culture, it is important for the state to “undertake the work of normalizing the products of the linguistic habitus” (Bourdieu 1991:48). The dilemma here is obvious: if the ROC officials adopted pinyin, they would deprive themselves of the authoritative role of defining the cultural paradigm by succumbing to the PRC's authority. On the other hand, if they refused to adopt such a widely accepted system, romanized Chinese in Taiwan would be less comprehensible to outsiders. The result would entail yet another form of international isolation—in addition to the political one—which has been afflicting Taiwan for decades.

The discrepancies in the writing systems have had profound effects on the imaginations of the nation. To put it in DeFrancis' words, the written script should be seen as “visible speech,” since “the primary defining feature of writing is as the representations of speech” (1989:248). Thus, even though the two Mandarins are orally communicable, written texts and printed words have made them visibly different. In other words, through the visible speech of different writing and romanization systems, Taiwan's Mandarin and China's Mandarin speak in different tongues, as it were, in the world of print. Here, Anderson's famous thesis on print-capitalism is turned on its head. According to Anderson, print-capitalism helped to conjure up a nation via three major mechanisms: by creating a sense of belonging among common readers, by giving a new fixity to language, and by creating a language-of-power endowed with administrative capacities (Anderson 1991:44–45). However, these same mechanisms have helped to differentiate rather than unify the people separated by the Taiwan Strait. To some Taiwanese, the visible differences in the script engender immediate alienation from, and sometimes antipathy toward, mainland China. The fixity of written scripts further reifies and reinforces the senses of difference between the ROC and the PRC. And finally, the two languages-of-power, signifying two sovereignties embodied in different scripts, imply that there are two nations instead of one.

Some have used the analogy of American English and British English to characterize the differences between “Taiwanese Mandarin” and “Chinese Mandarin.” This analogy, however, is somewhat misleading in that it fails to capture the major imparity between two Mandarins. For one thing, there is a significantly asymmetric “center-periphery” relationship that is not normally palpable in the English analogy. For another, since the written forms, diction, pronunciations, and romanization systems are all significantly different, some Taiwan Independence supporters contend that the national languages used in Taiwan and China are now two different languages. In an attempt to debunk the myth of “orthodox Chinese,” Qiu Gui-fen, a literature professor, contends from a post-colonial perspective that: “In the eyes of those China-centric scholars, Chinese [language] outside the mainland has to be inauthentic . . . Mandarin in Taiwan is literally Taiwanese Mandarin, a clumsy inferior Chinese. . . . [But] postcolonial discourses . . . debunk the myth that ‘Mandarin is the correct Chinese.’ Mandarin popularized in Taiwan has in fact combined the Taiwanese experience, loaded with Taiwan's history of being colonized [by both Japan and the KMT]; it is now a language of Taiwan's own, rather different from the commonly so-called “orthodox Chinese.” (Qiu 1992:155–56, my italics).

Qiu's postcolonial argument is echoed by other scholars who draw on the trendy discourse of postmodernity. In an attempt to defend Taiwan's cultural autonomy, they affirm these linguistic differences and refuse to use the “orthodox Chinese” standardized by the PRC (Liao 1998). As we can see, the Chinese language as developed in Taiwan is considered by many—particularly those in the PRC—as “unorthodox” or “inauthentic” for the Chinese nation. Meanwhile, “Taiwanese Mandarin,” already institutionalized by the KMT state within Taiwanese society, has taken an ironic turn to serve as new ground for building an alternative “Taiwanese nation.”¹⁶

the scene after 2000: external constraints and institutional legacies

External Constraints on Internal Politics

As one might expect, the institutional crisis of the ROC was exacerbated in the 1990s as Taiwan established increasing interactions with the outside world, while the legitimacy of the KMT regime was undermined during the democratization process (Wang 2000). The KMT under Lee Teng-hui's leadership launched a series of reforms to save its legitimacy crisis, but only to set the stage for future schism and the eventual breakup of the party.¹⁷

Some may contend that Chen's statement of “five no's” amounts to little more than paying lip service so as not to provoke the PRC. Be that as it may, it is ironic that such a hard-line Taiwan Independence proponent should have to openly deny his will to declare independence once he acquired the power to do so. Indeed, we have seen that there are many institutional legacies left by the KMT that Chen, now that he is the ROC president, cannot do without. This is where we can observe how external and institutional constraints exercise profound impacts on internal politics. For one example, under military threat from the PRC, Chen was cautious not to provoke his watching neighbor, and he was well aware that his regime had to gain backup from allies such as the United States. For another, Chen's source of power stems from neither a revolution nor a coup, but rather from an institutional procedure stipulated in the ROC Constitution, namely, a democratic direct election for the presidency. Had Chen overturned these institutional rules, including the Constitution on which the ROC was founded, he would have undermined the very foundation of his own legitimacy.¹⁹

To return to my opening points, “having a culture” is regarded as an “imperative,” as it were, to the nation-state. This imperative, previously assumed by the ruling KMT, has now passed to the newly elected DPP government. Indeed, in their blueprint for a new nation, the DPP has long vowed to construct a new “national culture” distinguishable both from the (KMT-founded) ROC and from the PRC. But when presented with the opportunity to realize this goal, the DPP found themselves caught in a paradoxical dilemma. Here we see how institutions both enable and constrain actors. The civic-territorial institutions of the ROC—including its Constitution, armed forces, administrative bureaucracy, and bounded territories—have enabled the DPP to claim that Taiwan is already an independent country. However, the legacies left by the KMT regime—ranging from the national title, the national flag, to the ROC's cultural institutions—have deterred the DPP from making the immediate claim that Taiwan under its rule is a new nation that is completely separable from Chinese culture and history. As Brubaker has shown in his study, institutional setups of the Soviet Union continue to shape nationalist politics in successor states; the same has held very true in Taiwan since the regime turnover.

Moreover, we must note that, by winning less than 40 percent of the vote, Chen did not gain a majority mandate. As Chen's victory resulted from the schism of the KMT, the DPP government under his leadership has been in effect a minority rule.²⁰

cultural politics under dpp rule: the multicultural turn and de-sinicization

When the DPP was still an opposition party, many Taiwan Independence proponents maintained that the so-called “treasures” in the National Palace Museum should be returned to the PRC because they neither belonged to Taiwan nor represented Taiwanese culture. However, since the DPP became the ruling party the situation has been reversed. The DPP, no longer questioning the legitimacy of owning these “national treasures,” utilizes the collections to help it break Taiwan's international isolation, whereas the pan-KMT camp,²¹

On the policy side, the DPP has changed their strategies as well. Few, if any, continue to insist that these treasures should be returned to the PRC; instead, it is maintained that Chinese culture is also part of Taiwanese culture, and that the treasures in the National Palace Museum represent part of Taiwanese history—a historical imprint of the KMT's rule on the island. However, it is also acknowledged that the collection in the National Palace Museum cannot represent the entirety and variety of Taiwanese culture. Tu Cheng-sheng, the new Director, has set an explicit goal that the museum will endeavor to enlarge its collection to include cultural artifacts of both aboriginal and Taiwanese history. A renowned historian of ancient China, Tu is well aware of the role that the National Palace Museum has played in the past politics of orthodoxy. Keen to avoid incurring controversies, he now takes a multicultural turn, emphasizing a “universal beauty” that ought to transcend all national, ethnic, and cultural boundaries.

In this way, instead of selling itself as “the representative of Chinese culture,” the ROC under DPP rule is now steering toward a model of multiculturalism in which the relationships between Chinese culture and Taiwanese culture are reversed. In the old KMT model, so-called “Taiwanese culture” is at best a local variation of Chinese culture. In the new DPP model, however, “Chinese culture” is merely part of the more encompassing, inclusive Taiwanese culture. Under the new mosaic model of multiculturalism, aboriginal cultures, along with the once disgraced imprints of Japanese colonialism—both of which were repressed and destroyed under the KMT's project of Chinese culture—are now preserved and promoted to a “national” status to represent Taiwanese culture.

A number of changes are underway as well regarding language policies. In the traditional discourse of Taiwanese nationalism, Mandarin was considered a language of “the colonizer” (i.e., the KMT rulers) that should be resisted by any and all means. Instead, the Taiwanese dialect Holo should become the “national language” of Taiwan. However, since the DPP assumed power, ironically, such radical appeals have to a great extent vanished. In order to maintain its legitimacy as the ROC leader, the DPP could not take such a radical step as suddenly abandoning Mandarin. Instead, subtler changes are being proposed and made. For instance, the “Language Equality Law,” which no longer privileges Mandarin as the only official language, promotes Holo, Hakka, and other aboriginal languages to an equal status with Mandarin in an attempt to counter its hegemony.

The overall tendency of the DPP to dilute the “Chineseness” of cultural landscapes in Taiwan has been characterized as “de-sinicization” (quzhongguohua, literally ‘removing Chineseness’). At this conjuncture, globalization and internationalization are invoked to increase the DPP's leverage regarding de-sinicization. For example, the central government proposed that, in order to help “internationalize” Taiwanese people and society, English should become an official language in Taiwan. It is now implemented policy that all students in the elementary schools must take English courses beginning from the third grade. In addition, the controversial Tongyong Pinyin, an ad hoc invention to allow distinction from the Hanyu Pinyin developed by the PRC, has now been adopted as the official standard of transliteration.²²

discordances from within and without: the politics of cultural orthodoxy, continued

Although the DPP is eager to cast aside the politics of orthodoxy that the KMT was unsuccessfully striving for, such an attempt, ironically, has caused loud discordant voices both inside and outside of Taiwan. The intention of “de-sinicization,” above all, has caused great anxieties and harsh criticisms from the major media and cultural elites in Taiwan. During the First Lady's “cultural journey” to Europe, a major newspaper published a commentary titled “the Son of Taiwan and the Sons of Heaven in China,” asserting that “We do not think ‘China’ is necessarily a liability to Taiwan. At least it is our asset in terms of culture. Except for aborigines, all people in Taiwan have their cultural roots in China.” The author goes on to argue: “We have preserved beautiful traditional Chinese characters, with which we write down joys and sorrows of this land. . . . Taiwan has inherited Chinese culture; this is a simple fact that can neither be denied nor distorted. . . . Many fundamentalists [of Taiwan independence nationalism] have been advocating “de-sinicization,” denying the Chinese elements in Taiwanese national culture. This not only betrays the truth, but also sees our own past as an enemy. (China Times, 17 July 2003)

The notion that “Chinese culture is an asset” reemerges in these debates, while the politics for cultural orthodoxy persists in the cultural field. A climax was reached in a recent debate initiated by popular writer Long Ying-tai. In June 2003, she wrote a series of newspaper columns that launched a general attack on the DPP's cultural policies. Lamenting the decline of Chinese culture on the island, her protest expressed a mix of anger and bitterness:

Influential and provocative, Long's articles set off a firestorm of debate in the media and on the Internet. Some criticized her for being nostalgic of the old authoritarianism of the KMT, while some others praised her for “courageously” speaking out what they felt in an undisguised manner: “Chinese culture is the most precious asset that Taiwan could have in the international competition. We can't even wait to grab it as our own; how could one want to throw it away?!” (Long 2003). As indicated, even though the ruling party has changed and the momentum of nation-building has shifted, Chinese culture is nonetheless regarded by some cultural elites as “the most precious asset” with which Taiwan can gain a competitive edge in the international race.

Across the Strait, the reactions from the PRC have been mixed and intriguing. During the KMT's rule, the Chinese culture constructed on Taiwan was dismissed by the PRC as “peripheral” or “unorthodox,” unable to represent the “authentic” “new” China. The use of “complex characters” and the old-style Wade-Giles system, were regarded as altogether “reactionary” residuals retained only by the corrupt KMT, which should be renounced and banned by all means.²³

This situation illustrates the predicament in which Taiwan has been caught: when the KMT strived for orthodoxy through the construction of the ROC, it proved unsuccessful. When the DPP tries to steer away from this politics of orthodoxy, it faces even stronger opposition from both inside and outside. Even if the DPP endeavors to construct a new Taiwanese culture as distinguishable from Chinese culture, their efforts at de-sinicization are nonetheless subject to external constraints (of the PRC) and institutional legacies (of the ROC) left by the KMT.

conclusion: how and why do institutions matter?

The preceding analysis is not meant to preach the necessity of a reified national culture, nor do I mean to suggest that there is such a thing called “Chinese Culture,” with a capital C, as essentialized by Chinese people and stereotyped by foreigners. On the contrary, I have tried to point out that, precisely due to its partial failures to meet institutional expectations—on both domestic and international levels—that it would standardize an essentialized national culture, the ROC was during the KMT rule losing its grip on its claim of being the Chinese nation. In other words, there was an institutional repertoire for national culture, but the ROC could not quite play it out. Gellner once characterized nationalism as “the striving to make culture and polity congruent, to endow a culture with its own political roof, and not more than one roof at that” (Gellner 1983:43). An important implication lies behind this insight: once established and recognized as a “nation” in the political field, the state has also monopolized the symbolic capital of the nation in the cultural field (Bourdieu 1998:35-63). By winning the battle in the political field, the PRC was able to extend its victory to the cultural field as well. The ethno-cultural institutions of the ROC will face a series of crises as long as the PRC preempts the meanings of “China” and “Chinese” in the symbolic sphere. These crises, in turn, nourish the emerging identity crisis in contemporary Taiwan. Even though the KMT stepped down in 2000, the DPP government could hardly be expected to turn the tables. The intended new Taiwanese nation did not come into existence with the DPP's coming to power, while the intentions of de-sinicization have caused harsh criticisms and objections from both cultural elites and the RPC.

Thus viewed, the Taiwanese case adds an intriguing facet to the current discussions of identity politics in general, and Chinese identity in particular. Taiwan has often been juxtaposed with Hong Kong and Singapore, characterized together as “Chinese societies outside China.”²⁴

Such a situation can greatly enhance our theoretical understandings of nationhood, nationality, and national identity. This returns us to the institutionalist approach discussed at the beginning of this essay. To fully comprehend its theoretical leverage, we may turn to Pierre Bourdieu's notion of “rite of institution.” As Bourdieu puts it, institutions have a performative magic that can create arbitrary boundaries and differences ex nihilo. Thus, “to institute . . . is to consecrate, that is, to sanction and sanctify a particular state of things, an established order, in exactly the same way that a constitution does in the legal and political sense of the term.” The institution of a (national) identity, therefore, “is the imposition of a name, i.e., of a social essence. . . . It is to signify to someone what he is and how he should conduct himself as a consequence. In this case, the indicative is an imperative. . . . To institute, to give a social definition, an identity, is also to impose boundaries” (Bourdieu 1991:119–20, my italics).

In this light, the theoretical implication of the above analysis are twofold. First, it demonstrates the analytical strengths and fruitfulness of the institutionalist approach to the study of nations and nationalism. Combining Brubaker's original formulation with John Meyer's institutionalist theory of world polity, I have carried further the institutionalist approach to nations and nationalism by pointing out that there are two types of institutions of nationhood—civic-territorial and ethno-cultural—that have to perform on both domestic and international levels. If either type cannot perform its “social magic” well, then the existence of the nation will be jeopardized, and an identity crisis may ensue. The institutionalist approach also helps us to uncover enduring effects as well as unintended consequences in nationalist politics, as has been shown in both Brubaker's analysis of the Soviet Union and this analysis of Taiwan. The unintended consequences in the Soviet case were manifested in elites' deeply institutionalized sense of political ownership and entitlement that continues to shape nationalist politics in successor states. Similarly, in Taiwan, there has been an deeply institutionalized sense of ownership and entitlement that cannot be fulfilled at the international level, while the KMT state has arguably institutionalized a nation that was not originally intended. The institutional shell of the ROC has enabled the DPP to claim that Taiwan is already an independent country, although, in effect, it has also constrained DPP's efforts to build a Taiwanese nation to a great extent.

Second, this study also highlights the importance of international or “external” factors in understanding the dynamics of nations and nationalism. Once again, we can obtain further insights from Bourdieu's formulation: “The act of institution is thus an act of communication, but of a particular kind: it signifies to someone what his identity is, but in a way that both expresses it to him and imposes it on him by expressing it in front of everyone and thus informing him in an authoritative manner of what he is and what he must be” (1991:121, my italics).

In other words, identity is at once both self-recognition and recognition by others. The failures of the ROC's institutions have led to an identity crisis among its nationals in that, in front of others, there is no institutionalized way for them to signify who they are in a definite, “authoritative” manner. Commenting on the chaotic quality of identifications in Taiwan, where neither “Chinese” nor “Taiwanese” have gained institutional support, a journalist wrote:

To put this in Bourdieu's terms, the reason why Taiwanese people are “not knowing how to be themselves” lies in the failures of ROC's institutions that cannot perform the social magic in front of outsiders to signify “who they are” in an affirmative manner.

To some, Taiwan may appear to be a strange nation if the view of outsiders carries so much weight regarding its self-definition. In this respect, Taiwan is indeed a “strange” nation in that many of its characteristics, including its ambiguous international status, appear incomprehensible to outsiders. And this is precisely where the case of Taiwan may contribute to our understanding of nations and nationalism from an institutionalist perspective. “Strange” as it may seem, Taiwan can serve as a “pathological” case, in Garfinkel's ethnomethodological sense. By “breaching” the mundane reasoning of our commonsensical world, it can help us to illuminate how things work under “normal” conditions.²⁶

As I have pointed out, most literature on Taiwan's nationalist politics tends to focus on the “internal politics” in the domestic domain, and to ignore external factors. It is one of my major intentions here to highlight the significance of external dimensions, but of course I am not suggesting that internal politics is unimportant or irrelevant. Rather, I have attempted to show the profound and enduring effects that external factors have had upon internal politics. By examining how ROC institutions have been built, and how the ROC has gradually lost its “performative magic” in the international arena, I have demonstrated how and why external factors matter, and how they have both enabled and constrained internal politics.

As Meyer and his colleagues (Meyer et al. 1997) have pointed out, the construction of a nation-state, particularly for those “late comers” in world history, is exogenously driven by a larger institutional structure that lies outside the society under investigation. However, such “exogenous variables” are often dismissed or ignored in sociological studies and in theorization of nations and nationalism. In classic studies such as Gellner's (1983), Hroch's (1985), and Smith's (1991), “international” or “external” factors play merely marginal roles. Even in Brubaker's institutionalist account of the Soviet Union, institutions are dealt with only at a domestic, national level. As our case has demonstrated, institutions that define the nation/nation-state need to perform their “social magic” not only on the internal, domestic level, but also the external, international level. To fully grasp the dynamics of nationalism and national identity formation, it is essential to incorporate both levels into our analyses.²⁷