Taiwan in the Chinese Imagination, 17th–19th Centuries
By Emma Jinhua Teng
Winter 1697: After four days and nights aboard a junk crossing the treacherous waters of the Taiwan Strait, Chinese traveler Yu Yonghe excitedly spotted the peaks of Taiwan’s mountains on the horizon.  In sight at last was the frontier island that he had longed to see since the Chinese conquest of Taiwan fourteen years earlier. As Yu wrote in his travel diary:
Taiwan lies far beyond the Eastern Ocean and has never, since the dawn of Creation, sent tribute to China. Now we have made . . . Taiwan the ninth prefecture of Fujian. By nature I am addicted to distant travel and I am fearless of obstacles and danger. Ever since Taiwan was put on the map, I have said that I would not be satisfied until I could see the place for myself. 
Yu Yonghe’s wish came true at last in 1697, when he volunteered for an expedition to Taiwan to obtain sulfur, a vital strategic item used to manufacture gunpowder. Friends and associates warned him against the voyage: the Taiwan Strait was perilous, filled with obstacles such as the notorious “Black Water Ditch,” which had capsized countless junks; Taiwan itself was a dangerous place, a mountainous jungle inhabited by “savages” and rife with deadly tropical diseases. Travelers told stories of shipwrecked sailors cannibalized by the islanders and of headhunting raids across the Taiwan Strait. Taiwan had also gained infamy as a “pirates’ lair.” Above all, the island was known as a stronghold for the Ming loyalist forces of Koxinga,  who had waged a war of resistance against the new Manchu Qing dynasty (1644–1911), and whose defeat by Qing forces in 1683 resulted in Taiwan’s becoming an imperial possession for the first time in Chinese history. It was this feat that sparked Yu Yonghe’s desire to travel to the island.
Despite the risks the journey presented, Yu was intrigued by the notion of seeing the empire’s newest frontier. Before the Qing conquest, few Chinese literati had traveled to this “savage island.” This voyage was Yu’s chance for adventure, his opportunity to go beyond the old boundaries of China and explore uncharted terrain.
Yu’s enthusiasm for the Taiwan frontier stands in sharp contrast to the disdain expressed by many of his contemporaries, who regarded the acquisition of this new territory as a waste of imperial resources. As one critic declared, “Taiwan is merely a ball of mud beyond the seas, unworthy of development by China. It is full of naked and tattooed savages, who are not worth defending. It is a daily waste of imperial money for no benefit.”  Such objections reflected the prevailing Chinese perception of Taiwan as a barren wilderness, an insignificant parcel of land beyond the pale of civilization. So deeply ingrained was this notion that the court had proposed in 1683 to abandon the newly conquered island after repatriating Ming loyalist troops to the mainland. Admiral Shi Lang, who had led the capture of Taiwan, vigorously protested this decision. In a memorial submitted to the emperor in February 1684, Shi argued for the importance of annexing Taiwan on both strategic and economic grounds.
I have personally traveled through Taiwan and seen firsthand the fertility of its wild lands and the abundance of its natural resources. Both mulberry and field crops can be cultivated; fish and salt spout forth from the sea; the mountains are filled with dense forests of tall trees and thick bamboo; there are sulfur, rattan, sugarcane, deerskins, and all that is needed for daily living. Nothing is lacking. . . . This is truly a bountifully fertile piece of land and a strategic territory. 
Shi Lang could speak with authority because he, unlike the emperor’s other advisers, had traveled to Taiwan and observed local conditions with his own eyes. Opponents of annexation knew little about Taiwan beyond the cliché of the island as a “miasmal wilderness,” for there was a dearth of information about the island in the Chinese histories and geographical records. With his personal knowledge of the island, Shi was uniquely empowered to speak as an expert on this subject. The emperor was sufficiently persuaded by Shi’s eyewitness account of the island and its riches to convene a meeting to debate the issue of annexing this territory. Shi’s faction eventually carried the day, and in the spring of 1684, Taiwan was officially incorporated into the Qing empire. Over two hundred years later, in 1887, the court would grant the island full status as a province of China.
In Taiwan’s Imagined Geography: Chinese Colonial Travel Writing and Pictures, 1683–1895 (2004), I examine the place of travel writing, pictures, and maps in Taiwan’s transformation from a “savage island” located “beyond the seas” (haiwai) into a “Chinese province,” an integral part of the Chinese empire. Qing travelers like Shi Lang and Yu Yonghe played a crucial role in the production of geographic knowledge about this newest addition to the Qing domain. Their writings helped to demonstrate that far from being a “ball of mud” inhabited by “naked and tattooed savages,” Taiwan was endowed with land worth cultivating and populated by natives deserving of inclusion as subjects of the empire.
The Qing incorporation of this island involved not only a reconsideration of Taiwan’s place in imperial geography but also a reconceptualization of the Chinese domain itself. The Ming conviction that Taiwan was not part of this domain was rooted in the traditional conception of China as a territory bounded by natural geographic features, such as mountains, rivers, the desert, and the sea.  Since Taiwan was separated from the Chinese mainland by the Taiwan Strait, it was, ergo, outside China. The Qing expansion into territory “beyond the seas” entailed a shift from the established conception of China to a new spatial image of an empire that transgressed the traditional boundaries.
The annexation of Taiwan was only one incident in the much larger phenomenon of Qing expansionism, a phenomenon that scholars have recently begun to treat as an example of imperialism, comparable to European imperialisms.  Following the conquest of China proper, the Manchu rulers of the Qing dynasty pursued numerous military campaigns on China’s frontiers. These campaigns were driven largely by the Qing need to consolidate the empire and eliminate potential military rivals, including the Ming loyalist regime in Taiwan and the Mongols and Russians on the Central Asian frontiers.  A gradual process that spanned approximately a century, Qing expansionism was also motivated in part by economic interests and by population pressures in China proper, which generated a demand for new arable lands. Having annexed Taiwan in 1684, the Qing turned its attention to Central Asia, “pacifying” the Mongols and bringing eastern Turkestan and Lhasa, the capital of Tibet, under Qing rule. The Qing further expanded its control in south and southwest China, subjecting various non-Chinese peoples of this region to Qing domination. At its height, in the eighteenth century, Qing influence extended into Korea, Vietnam, Laos, Thailand, Burma, and Nepal, all of which came under the suzerainty of the empire.
Figure 1. Expansion of the Qing empire (cartographic design by Patrick Florance and Martin Gamache/AMG)
By 1760, the Qing had achieved the incredible feat of doubling the size of the empire’s territory (see Fig. 1), bringing various non-Chinese frontier peoples under its rule. The impact of Qing expansionism was thus tremendous, as the Qing not only redefined the territorial boundaries of China but also refashioned China as a multiethnic realm—shifting the traditional border between Chinese (Hua) and barbarian (yi). In doing so, the Qing created an image of “China” that differed vastly from that of the Ming.
In order to promote this new conception of the Chinese empire, the court commissioned a number of major projects to depict the expanded imperial domain. Among them were the Kangxi-Jesuit atlas (1717), a comprehensive survey of the empire; The Comprehensive Gazetteer of the Great Qing Realm (Da Qing yitong zhi; ca. 1746), a compendium of geographic information about the empire; The Qing Imperial Tribute Illustrations (Huang Qing zhigong tu; ca. 1769), an illustrated catalogue of the peoples of the empire and other “tributaries” (see Figs. 2–4); and the Imperial Glossary of the Five Dynastic Scripts (Wuti Qingwen jian), an encyclopedic, multilingual glossary of the five major languages of the empire. These texts at once served to define the extent of the empire and to articulate the vision of a geographically diverse and multiethnic imperial realm.
Figure 2. Picture of Raw Savages of Zhanghua County, Taiwan from
Qing Imperial Tribute Illustrations (ca. 1751)
Figure 3. Picture of Raw Savages of Danshui, Taiwan from Qing
Imperial Tribute Illustrations (ca. 1751)
Figure 4. Picture of Cooked Savages of Taiwan County, Taiwan from
Qing Imperial Tribute Illustrations (ca. 1751)
I argue that travelers’ representations of frontier regions such as Taiwan played an important role in the creation of the new imagined geography of the expanded Qing empire. Frontier travel writing emerged as a vital genre during the Qing, as Chinese literati, military men, and merchants traveled to the frontiers in unprecedented numbers. Not only did frontier travelers compose written accounts of their journeys, but a good number of them produced various kinds of tu (maps, pictures, illustrations) as visual records of their observations (see Fig. 5). These tu included pictorial maps, ethnographic images, drawings of flora and fauna, architectural renderings, and pictures commemorating battles and other events on the frontier. (Henceforth, I will refer to these various tu as “topographical pictures.”) As the empire expanded, travelers’ accounts and topographical pictures became an important source of geographic knowledge about the newly acquired lands, knowledge that was crucial for strategic and administrative purposes. Travel writing and pictures also served an important ideological function. In representing the distant lands and the ethnically diverse peoples of the frontiers to audiences in China proper, these works transformed places once considered non-Chinese into familiar parts of the imperial realm and thereby helped to naturalize Qing expansionism through the production of a re-imagined imperial geography.
Figure 5. Chen Lunjiong, “General Map of the Four Seas” from Record of Things Seen and Heard at Sea (ca. 1723–30)
The legacy of Qing imperialism for modern China has been profound: because the People’s Republic of China (PRC) now claims sovereignty over virtually all the territory acquired by the last dynasty, the impact of Qing expansionism continues to be felt by the people of Tibet, Xinjiang, Taiwan, and other former frontier regions. Separatist (“splittist” in PRC jargon) movements in all these areas have met with staunch opposition from the Chinese state, which considers such lands inseparable parts of China’s sacred territory. Hence, the PRC claims Taiwan—which was a Japanese colony between 1895 and 1945 and which has been ruled by a separate (and recently democratic) government as the Republic of China (ROC) since 1949—as “sovereign territory” that must be returned to the Chinese motherland with due speed. Ironically, the “territorial integrity” that Chinese nationalists seek to defend is based on a territorial image of “China” created by an invading Manchu dynasty, and not the older Ming image.
Of the former Qing frontiers, Taiwan is of particular interest because the question of the island’s sovereignty in the postwar era remains unresolved and hotly contested: Is Taiwan de facto a “sovereign state,” or is it, in the words of the U.S. media, a “renegade province” of China?  Taiwan’s relationship to the PRC and the question whether Taiwan might officially declare independence were leading issues in the 2000 presidential race in Taiwan, and remain hot-button topics. In an attempt to influence the outcome of that election, the PRC issued a thinly veiled threat of force: “To safeguard China’s sovereignty and territorial integrity and realize the reunification of the two sides of the straits, the Chinese government has the right to resort to any necessary means.”  The “Taiwan issue” (involving arms sales to Taiwan) is the prickliest thorn in U.S.-China relations and has the potential to bring the two powers into armed conflict.  The geopolitical importance of Taiwan combined with Taiwan’s emergence since 1987 as a “Chinese democracy” has contributed to the growth of Taiwan Studies as an important new field in Asia and the United States.
In examining the process by which Taiwan was incorporated into the imagined geography of the Qing empire, Taiwan’s Imagined Geography helps to explain how an island that was terra incognita for the better part of Chinese history came to be regarded as an integral part of China’s “sovereign territory.” My work views Taiwan-China relations as a product of a particular history—that of Qing expansionism—rather than as a matter of vague “ancestral ties.”  By elucidating the nature of this historical relationship, I seek to add to our understanding of current political events in the region.
Expanding Colonial Discourse Studies
Although the primary focus of Taiwan’s Imagined Geography is the Qing construction of Taiwan’s imagined geography, in writing this book I also hoped to challenge prevailing preconceptions of “the colonizer” and “the colonized” by examining a non-Western imperial power. The presumption that colonizers were European and the colonized non-European is deeply entrenched both inside and outside the academy. The very notion of studying “Chinese colonialism” thus seems alien to many. On more than one occasion, I have been asked: “What do you mean by ‘Chinese colonial travel writing’? Do you mean European colonial travel writing about China?” The idea that “imperialism” is essentially a Western phenomenon has also been reinforced by scholars of modern China’s “postcoloniality,” who have tended to focus on China’s historical experiences with Western imperialism while ignoring China’s own history as an imperialist power.  This is due in no small part to the PRC’s ardent denials that the Chinese were ever anything but victims of imperialism; hence official PRC discourse refers to Qing expansionism as “national unification,” and talk of “Chinese imperialism” is heresy.  I seek to remedy this situation by asserting that China’s postcoloniality must also be understood in terms of the legacy of Qing expansionism.
Expanding colonial discourse studies to include imperial China is no easy task, for one immediately runs into terminological difficulties. Scholars (both Western and Chinese) frequently argue that terms such as “imperialism” or “colonialism” cannot be applied to China on the grounds that Qing expansionism does not fit the model of European imperialism. Of course, the Chinese had an empire, just as Rome had an empire, it is often argued, but an empire is not the same as “imperialism.” Yet the notion of “European imperialism” is itself problematic. The scholarship of the past several decades has shown that there is no single model of European imperialism: British imperialism differed from French and German imperialisms; nineteenth-century imperialism differed from the earlier conquests of the New World and the mercantile colonialism of the seventeenth and eighteenth centuries. And then there is the matter of American “neo-imperialism” or “neo-colonialism.” Theorists have debated whether imperialism is best understood primarily as a political system (as in late nineteenth-century England) or as an economic system (as by early twentieth-century critics). Definitions of “imperialism” range from Lenin’s (1916) restrictive “monopoly stage of capitalism” to Michael Doyle’s broadly inclusive “imperialism is simply the process or policy of establishing or maintaining an empire.”  Thus, there is no universal agreement, even among Europeanists, on the precise definition of “imperialism.”
Theorists have also been concerned with distinguishing “colonialism” from “imperialism,” although the two terms are often used interchangeably. Again, there is no consensus on this score, with some theorists taking “colonialism” as a subset of “imperialism” (e.g., Benita Parry) and others taking “colonialism” as the more general term and “imperialism” as the particular, late nineteenth-century European phenomenon (Nicholas Thomas).  Because of this ambiguity, there will necessarily be some slippage in discussing theoretical approaches to “imperialism” and “colonialism” here. Like “imperialism,” “colonialism” is a complex and multivalent term that refers to a variety of historical and regional experiences, ranging from the “settler colonialism” of Australia to the “internal colonialism” of the American ghetto.  Since “colonialism” derives from the Latin colere (to cultivate), the distinction that I find most useful is Edward Said’s: “‘Colonialism,’ which is almost always a consequence of imperialism, is the implanting of settlements on distant territory.”  This distinction notwithstanding, following convention I employ the term “colonial discourse,” rather than “imperial discourse,” to describe the complex of signs and practices within which the Qing empire was known. As David Spurr writes: “In speaking of the discourse of colonialism, the distinction [between colonialism and imperialism] tends to collapse, since the basic principles of this discourse . . . also constitute the discourse of imperialism.” 
My use of the term “Qing imperialism” therefore rests on the premise that, to quote Raymond Williams, “imperialism, like any word which refers to fundamental social and political conflicts, cannot be reduced, semantically, to a single proper meaning. Its important historical and contemporary variations of meaning point to real processes which have to be studied in their own terms” (italics added).  To this end, I use the word “imperialism” to refer to the Qing conquest of vast tracts of non-Chinese lands through military force, their rule of these distant lands from an imperial center, and their incorporation of significant numbers of ethnically distinct, non-Chinese peoples as subjects of the empire. Since an important aspect of Qing imperialism was the implanting of Han Chinese settlements on distant frontier territories, I consider Qing expansionism at once an imperial and colonial phenomenon.
More fundamentally, I use the term “imperialism” to denote the set of practices, policies, and ideologies through which the Qing empire was fashioned and maintained. In this I follow Edward Said’s definition: “‘Imperialism’ means the practice, the theory, and the attitudes of a dominating metropolitan center ruling a distant territory.”  I find Said’s definition useful, since it is neither so restrictive that it inhibits cross-cultural comparison nor so general that it loses theoretical rigor. Moreover, it speaks directly to my concerns, for my work does not deal primarily with imperialism as an economic or political system, but rather with imperialism as a set of attitudes and power relations: precisely those aspects of imperialism that seem to be the most intransigent and that some argue have outlasted formal colonial rule. In shifting the focus to the cultural and ideological dimensions of imperialism/colonialism, I draw on Nicholas Thomas’s argument: Colonialism is not best understood primarily as a political or economic relationship that is legitimized or justified through ideologies of racism or progress. Rather, colonialism has always, equally importantly and deeply, been a cultural process; its discoveries and trespasses are imagined and energized through signs, metaphors and narratives; even what would seem its purest moments of profit and violence have been mediated and enframed by structures of meaning. 
I would argue that it is in imperialism/colonialism as a cultural process that we can begin to find the common ground on which “European imperialism(s)” and “Qing imperialism” can be discussed. This is not to deny the historical specificity of late nineteenth-century European imperialism or of Qing imperialism. Nor is it a plea for a return to general, universal theories of imperialism. Rather, it is an attempt to extend the ground on which particular, historical, and localized accounts of imperialisms and colonialisms can be delineated. It is an attempt to initiate a dialogue where there has been none.
My intent in reclaiming the use of the word “imperialism” for China studies is not to suggest the equivalence of Qing and European imperialisms; although recent scholarship in the China field has shown that, despite the manifest differences, Qing and early modern European imperialisms bear enough similarities in terms of institutions and processes to merit comparison.  What I want to do instead is to suggest a problematic: Why is it that Qing expansionism, which involved territorial conquest, political control, economic exploitation, and cultural hegemony, cannot be considered “imperialism”? What does it mean to call the Qing an empire without “imperialism”? What difference would it make if we were to see Qing expansionism as imperialism and not simply as an imperium? I address this problematic further in the Epilogue to Taiwan’s Imagined Geography and suggest that it informs our understanding of China-Taiwan relations today.
Setting Qing expansionism within the broader framework of colonial studies rather than confining it within the perimeter of “area studies” allows us to see China in the context of global historical processes, rather than as a unique and timeless civilization unto itself.  An intriguing statement made by a nineteenth-century travel writer demonstrates that there were Chinese literati who perceived the Qing imperial project in global terms. This traveler, Ding Shaoyi, compared his own observations of Qing Taiwan to Guileo Aleni’s description of the European colonization of North America:
The savagery of the native barbarians of the newly opened frontiers of North America is no different from that of the savages of Taiwan. In the past, they were extremely ferocious, yet Europeans have managed to guide them with their senseless, confused religion and have finally changed the native customs. So it is a real injustice to say that the raw savages of Taiwan have absolutely no human morals despite their human appearance and that they cannot be civilized with our Kingly Governance (wangzheng)! 
Ding addressed his comments to contemporaries skeptical about the possibility of civilizing the “savages” of Taiwan. Since the Europeans had succeeded in North America, he concluded, surely the Chinese, with their superior civilization, would prevail in Taiwan. Although recognizing differences in specific beliefs and institutions, Ding perceived fundamental parallels between European and Chinese expansionism and their civilizing enterprises. Historian Laura Hostetler has also demonstrated that the Qing emperors were keenly aware of imperialism as a global phenomenon and perceived themselves as players in an international game of territorial expansionism. 
The notion of “Chinese imperialism” is rendered particularly complex by the fact that the Qing was itself a conquest dynasty. The majority, Han Chinese population of China proper was subjugated to the Manchu ruling class. Yet the Qing adopted many of the fundamental political, economic, and cultural institutions of the Chinese imperial system, becoming somewhat “sinicized” in the process. At the same time, ethnic Han Chinese participated actively in the military and political life of the Qing, becoming part of the ruling class and perhaps somewhat “Manchuized” in the process.  This ethnic complexity only intensified with Qing expansionism. Whereas the Qing army, with its multiethnic troops, conquered the frontier territories, the colonists who settled these frontiers were nearly exclusively Han Chinese. 
What we find in frontier regions such as Taiwan, then, is no simple dichotomy of colonizer/colonized, but a multilevel hierarchy of colonial officials (both Manchu and Han), Han Chinese settlers, and indigenous peoples. Each group had its own interests—sometimes these interests competed with those of other groups; sometimes they intersected. Qing administrators did not necessarily view Han Chinese settlers as natural allies on the frontiers. Indeed, given the island’s history as a base for Ming loyalists, Qing officials regarded the settlers in Taiwan with suspicion. During the initial years of Qing expansionism, in particular, the court often adopted policies that favored indigenous peoples over Han Chinese colonists, in large measure to prevent costly ethnic unrest.  The study of Qing imperialism therefore seems particularly apt at a time when scholars of colonial studies are arguing for the need to rethink older models of a strict colonizer/colonized dichotomy and to consider the complexity of ethnic interactions in colonial contexts. 
The analysis of frontier travel writing and topographical pictures promises to enhance our understanding of the Qing frontier experience greatly. Historians have produced excellent studies of the political, economic, and military administration of the Qing frontiers, but relatively little has been written on cultural representations of the frontiers.  Like European imperialisms, Qing imperialism was a complex and dynamic convergence of strategic, economic, political, cultural, and ideological interests. Studies of Western imperialism have emphasized the vital role of colonial discourse in sustaining empires and producing colonial subjects.  I would argue that if we wish to understand the Qing formation of a geographically and ethnically diverse empire, scholars of China must similarly attend to the role of colonial discourse in this process and the cultural dimensions of the frontier experience. Travel literature and topographical pictures, both of which constitute forms of colonial discourse, are valuable resources for such a project: these texts express and articulate ideologies of imperialism even as they engender ideas about the frontiers.
The study of colonial discourse has largely been inspired by Edward Said’s Orientalism (1978). Although widely critiqued as reductionist or one-sided in the decades since its publication, this provocative work nonetheless remains a foundational text in the field. Said described “Orientalism” as a complex set of ideas and images through which European culture defined the East as “other” to Western civilization. Chinese representations of the frontiers as exotic, uncivilized, and barbarous bear fundamental similarities to European Orientalism. If European culture, as Said argued, derived its sense of identity and strength by setting itself off against the Orient, Chinese civilization gained its sense of identity as “the Middle Kingdom” (Zhongguo) in opposition to the “barbarians of the four directions” (siyi). Both traditions attempted to establish their own civilization as the normative ideal and to project “over there” qualities and traits (lasciviousness and indolence, for example) that they sought to repress in their own societies. Painted in the broadest strokes, European Orientalism and Chinese discourse on barbarians can be regarded as comparable. Indeed, the similarities are striking and point to the existence of shared, cross-cultural modes of constructing foreign “others.”
In my work on Qing travel literature and topographical pictures, I consider Chinese representations of the frontiers as a form of discourse roughly equivalent to Orientalism yet shaped by the particular conditions of the Qing imperial enterprise.  Much as European Orientalism has outlasted European colonialism, its Chinese counterpart has outlasted the particular institutions and circumstances of Qing imperialism and lives on in contemporary representations of ethnic “minorities” in the modern Chinese nation-state. This phenomenon has been described by anthropologist Dru Gladney as “oriental orientalism” and by anthropologist Louisa Schein as “internal orientalism.”  Expanding the scope of colonial studies allows us to view China not simply as the object of Orientalist discourse or as a mimic of Western Orientalism but also as the producer of its own brand of exoticist discourse.
One aspect of my research has been to ask what the concepts of “race” and “ethnicity” meant in the Qing context.  Were Qing representations of frontier peoples culturalist or racialist? Or a combination thereof? Like “imperialism” and “colonialism,” “race” and “ethnicity” are two terms that bear comparative treatment across cultural contexts. These two terms are perhaps the more difficult pair, for they are inadequately defined (even in contemporary American usage) and, moreover, have long and complicated histories tied to the histories of imperialism and conquest. Anyone trying to define these terms quickly enters a swamp in which concepts of race, ethnicity, culture, nation, and tribe are inextricably tangled. “Race” and “ethnicity” are, furthermore, loaded terms in English; “race,” in particular, immediately brings to mind racism and eugenics. Using the word “race,” even in reference to a historical idea or construct, is often mistaken as a signal that one believes in race as an objective fact—which I certainly do not. 
Without entering too deeply into this terminological morass, it is safe to assert that both race and ethnicity essentially refer to the categorization of peoples based on some notion of difference; how this difference has been defined has varied historically and culturally (witness the latest U.S. Census Bureau attempts to redefine its categories of race and ethnicity).  As such, my analysis focuses on the ways in which Qing writers conceptualized human difference and categorized groups of people within the empire. 
In frontier travel writing, Chinese literati were concerned with a range of human differences—physical, cultural, linguistic, intellectual, and moral, as well as in human nature (xing). For Qing authors, the relative significance of these types of differences was a matter of debate. In my analysis of Chinese sources on Taiwan, I identify two discourses concerning human difference: what I call a “racialist discourse” and an “ethnical discourse.” I roughly define the first as a discourse that focuses on physical differences and innate differences in human nature. The racialist discourse further constructs difference as categorical and absolute, along the lines of the distinction between humans and animals. In contrast, ethnical discourse focuses on cultural differences and constructs difference as a matter of degree within certain human universals.
In labeling these discourses “racialist” and “ethnical,” I imply certain parallels with Western discourses that place “race” on the side of nature and “ethnicity” on the side of nurture, in the “nature” versus “nurture” debate on human difference. However, in drawing these parallels, I by no means suggest that Qing concepts of race and ethnicity are precisely equivalent to their Western counterparts (of any place or period).  Why, then, use a loaded word like “race” at all? Why not stay with a seemingly more neutral term like “ethnicity”? Like Pamela Kyle Crossley, I argue that “unless one has resort to the term ‘race,’ ‘race’ and ‘ethnicity’ would remain mingled with each other, the process of differentiation forever muddled by the notion that ‘ethnic group’ is just a better word for what was once called ‘race.’”  I therefore use the word “race” primarily to distinguish what I call “racialist discourse” from “ethnical discourse,” rather than to highlight the similarities between Western and Chinese thinking on race, as does Frank Dikötter in his Discourse of Race in Modern China.
As a form of colonial discourse, travel texts provide crucial data for the analysis of Qing constructions of race and ethnicity, for they allow us to see in greater detail Qing views on frontier peoples beyond official pronouncements on ethnic policy. In my analysis of Qing accounts of Taiwan, I demonstrate how the Qing ideology of empire, which sought to accommodate different ethnic groups and to suppress the distinction between Chinese (Hua) and barbarian (yi), came into conflict with older attitudes of Han Chinese chauvinism, often expressed in racialist terms.
The “imagined geography” of my book title is intended to distinguish between the geography that exists on the ground and geography as a cultural construct. Thus, although Taiwan never moved from its position at 23.5 degrees north/120 degrees east, 96 miles from the Chinese coast, in terms of the Chinese geographic imagination, between the seventeenth and nineteenth centuries the island shifted from “far, far beyond the seas” to a location firmly situated within the Chinese empire. Then, between 1895 and 1945, the island that had only recently “become Chinese” “became Japanese,” as Leo Ching demonstrates.  In focusing on the role of discourse in these processes, I concur with Edward Said that the “struggle over geography . . . is complex and interesting because it is not only about soldiers and cannons but also about ideas, about forms, about images and imaginings.”  My use of the term “imagined geography” draws on Said’s notion of “imaginative geography,” as well as the concept of “imagined communities” introduced by Benedict Anderson.
In Orientalism, Said proposed the notion of “imaginative geography” to denote the complex set of ideas and images by which geographic entities such as the “Orient” and the “Occident” were historically produced. Said argued that it is through “imaginative geography” that meaning is assigned to the space “out there,” beyond one’s own territory. As Said wrote: “All kinds of suppositions, associations, and fictions appear to crowd the unfamiliar space outside one’s own.”  Said thus highlighted what he saw as the arbitrary and imaginative dimensions of geographic knowledge. Following Said, the term “imaginative geography” has been used to refer to the culturally constructed nature of geography and to the role of discourse in producing geographic knowledge. Whereas Said’s work in Orientalism primarily focused on European texts about the Orient, recent scholars such as Joan M. Schwartz, Anne Godlewska, and Derek Gregory have considered imaginative geography as the product of a variety of representations—literary, cartographic, pictorial, photographic, and so forth—working in concert. 
The concept of “imagined communities” derives from Anderson’s pioneering work on nationalism, in which he described the nation as an imagined political community—”imagined” because “the members of even the smallest nation will never know most of their fellow-members, meet them, or even hear of them, yet in the minds of each lives the image of their communion.”  Borrowing from Anderson, I suggest the Qing empire not as a community but as an “imagined geography,” a defined and limited spatial image that existed in the minds of Qing elites despite the fact that most would never travel to the distant reaches of the empire. This imagined geography delineated the territory that belonged to the “our land” of the Qing empire, in distinction to the “barbarian lands” that lay beyond its boundaries. As such, imagined geography describes the process by which the “geo-body” of an empire is produced.
The term “geo-body” was first used by Thongchai Winichakul in 1994 to describe the territoriality of the Thai nation:
Geographically speaking, the geo-body of the nation occupies a certain portion of the earth’s surface which is objectively identifiable. It appears to be concrete to the eyes as if its existence does not depend on any act of imagining. That, of course, is not the case. The geo-body of the nation is merely an effect of modern geographical discourse whose prime technology is a map. 
Although there are essential differences between empires and nation-states, since Thongchai concedes that the definition of the term is neither strict nor conclusive, I will also use the term “geo-body” in reference to the Qing empire, which I regard as an effect of multiple geographic discourses—textual, pictorial, and cartographic. 
My use of the term “imagined geography” thus distinguishes this particular form from other forms of imaginative geography more generally. “Imaginative geography,” as Said described it, tends to dramatize the distance and difference of what is “out there.” Following Orientalism, studies of imaginative geography have generally focused on the “barbarian land” side of the “our land–barbarian land” distinction that Said posited as fundamental to imaginative geography.  In my formulation, “imagined geography” is concerned primarily with defining “our land” and thus focuses on the other side of the equation. More specifically, the imagined geography of empire describes the process by which “their land” is converted into “our land.” Rather than simply dramatizing distance and difference, imagined geography at once exoticizes the other and attempts to convert otherness into familiarity and we-ness. I use the variant form “imagined geography” to connote this slight shift of focus.
Yet because attempts to construct identity out of difference can never be wholly successful, Qing imagined geography was characterized by an inner dissonance. The tension between dramatizing difference and domesticating it marked Qing representations of Taiwan and other frontier regions of the empire. It is this tension between difference and sameness, distance and union, the exotic and the familiar, that I explore in tracing Taiwan’s transformation from “savage island” into “Chinese province.”
Frontier Travel Writing and Pictures
Frontier travel writing and pictures had an intimate connection with the Qing imperialist project. These works served as vital sources of information about the new regions of the empire, especially during the early years of expansionism, when other sources of empirical geographic information were not readily available. Frontier officials relied on travel accounts, maps, and other topographical pictures to familiarize themselves with local conditions—the terrain, natural resources, local customs, relations between various local tribes, and so forth. Compilers of local gazetteers likewise employed such texts as sources for the production of these compendia of geographic and historical information and crucial administrative aids. Travel accounts also served as important source materials for general geographic and historical works, encyclopedias, guidebooks, and even the zhiguai (records of anomalies) collections that Chinese literati read for entertainment.
Frontier officials themselves seriously engaged in the collection of geographic information about the frontiers. An excerpt from an eighteenth-century travel account gives us a picture of how officials went about this activity:
In the course of my duties, I toured around and inquired after customs and strange products. I saw all kinds of unusual and weird things that have never been seen in China proper. . . . In my spare time from official duty, I ordered a painter to make illustrations of those concrete things that I had seen and heard. . . . I will keep [this pictorial album] in my travel trunk, so that when I return I may present it to the learned and accomplished gentlemen at the capital and thereby expand their knowledge. 
Officials produced such textual and pictorial accounts both as personal mementos and as aids for colonial administration. In addition, they sometimes submitted travel accounts, maps, or illustrations to the emperor as records of conditions on the frontiers or of their own achievements in frontier service. The court itself occasionally ordered frontier officials to submit geographic information and illustrations for use in the compilation of grand, empirewide projects such as the Comprehensive Gazetteer of the Great Qing Realm or The Qing Imperial Tribute Illustrations. 
The subjects taken up by travel writers were shaped to a large degree by the conventional categories of geographical recording—topography, climate, buildings and institutions, transportation routes, local ethnic groups and their customs and languages, flora and fauna, local products, geographic marvels, and so forth. Although the subjects of travel writing were fairly uniform, travelers chose to compose their accounts in a variety of formats—the travel diary, the essay, the notation book, or the geographical record (a genre consisting of short entries under various categorical headings) —each with its own generic conventions. 
Topographical illustrations, which were produced either by the travel writers themselves or by professional painters in their employ, generally focused on similar subjects (see Figs. 6–11). For the most part, frontier pictures can be divided into three main categories—maps or landscape pictures (ditu or shanchuantu), ethnographic illustrations (fengsutu), and illustrations of flora and fauna (fengwutu)—but a wide range of tu exist outside these categories. There are architectural drawings of civil, military, and religious structures, pictures of local industries, and illustrations commemorating events such as the annual review of troops or an official’s tour of inspection. As with travel writing, a variety of generic conventions shaped these pictorial representations. 
Figure 6. “Drinking Party,” from the Zhuluo County
Figure 7. “Pounding Rice,” from the Zhuluo County
Figure 8. “Hunting Deer,” from the Zhuluo County
Figure 9. “Fishing,” from the Zhuluo County Gazetteer
Figure 10. “Picking Betel Nuts,” from the Zhuluo County
Figure 11. Figure from the “Country of Wild People” from
Wang Qi’s Pictorial Compendium of the Three Powers (1609)
The development of frontier travel literature in the Qing was encouraged not only by the demands of expansionism but also by the new status accorded geography as a discipline. According to Benjamin Elman, the kaozheng (evidential scholarship) movement—one of the major intellectual trends of the late imperial era—elevated geography to a key discipline through important methodological innovations in the seventeenth century.  The new geography emphasized empirical observation, the systematic gathering of data, and philological research. It often focused on topics related to issues of frontier or maritime defense. Evidential scholarship also generated interest in historical geography, including the historical geography of China’s frontiers and borders. Thus, in frontier geography colonial imperatives coincided with the scholarly interests of Qing literati, and these interests reinforced one another.
Other cultural trends of the late imperial era also stimulated interest in the frontiers. This era has been described by many literary scholars as an age of surfeit, an age when everything seems to have been done, everything seems to have been said.  This was particularly true in the case of travel literature, for all the famous mountains, all the scenic spots in China, had been overinscribed with the writings of earlier travelers.  In conventional travel writing, the traveler was expected not so much to describe the scene before him but to meditate on his relationship to those who had come before him. Paintings of the famous scenic spots similarly were bound by convention and the precedents of famous masterworks. Frontier travel gave the literatus the opportunity to do something new and fresh, to cover new ground, as it were. Liberated from the need to dwell on historical models, the traveler as explorer and eyewitness observer took center stage. Travelers to Taiwan had perhaps the greatest leeway for originality, for unlike travelers to the Central Asian frontiers, for example, there were virtually no literary precedents for their journeys.
Qing travel literature evinced a new attitude toward the frontiers. Canonical Tang dynasty (618–907) literary treatments of the frontiers were filled with images of hardship and suffering—whether the bitter winds of the Central Asian frontier or the miasmas of southern border lands. The theme of exile permeates this literature, with unfamiliar terrain causing tears of homesickness and alienation to well up in the eyes of the poet. As Han Yu wrote from exile in Chaozhou in the south:
Typhoons for winds, crocodiles for fish—
Afflictions and misfortunes not to be plumbed!
South of the county, as you approach its boundary.
There are swollen seas linked to the sky;
Poison fogs and malarial miasmas
Day and evening flare and form! 
In contrast, we find Qing travelers, like Yu Yonghe, proclaiming a passion for “distant travels” (yuanyou) and even relishing the danger and strangeness of the frontiers. As Yu Yonghe declared: “In searching for the exotic and visiting scenic spots, one must not fear terrible inclinations: if the voyage is not dangerous, it will not be exotic; if the inclination is not terrible, it will not be exhilarating.”  In similar terms, another traveler vowed: “If the journey is not distant, then it will not be lusty; if the journey is not dangerous, it will not be exotic.”  These sentiments were echoed by numerous other frontier travelers who insisted on the unconventional and adventurous journey as the only authentic form of travel.
In the classic Chinese World Order (1968), John King Fairbank argued that traditional Chinese relations with non-Chinese peoples were colored by the concept of sinocentrism and the assumption of Chinese superiority. One would expect Chinese accounts of frontier peoples to be marked by this attitude of Han Chinese superiority. As I shall demonstrate, in frontier travel literature this was not, however, uniformly the case. Rather, Chinese views of the other were complex and often contradictory. As in Western travel literature, encounters with the other provided Chinese travelers with an opportunity to look back at the self, and literati representations of Taiwan and its indigenous people frequently expressed, and were colored by, their author’s political, social, or philosophical concerns.
Encounters with difference on the frontiers also prompted travelers to engage in cultural reflection, leading them to new understandings of Chinese culture and often to cultural relativism. Questioning the universality of Chinese culture, travel writers suggested that it was fitting for each place to have its own customs and tastes: as many a writer proposed, “Perhaps our customs seem just as strange to them.” Traditional Western historiography has presented China as culturally static and self-satisfied until the encounter with the West in the nineteenth century. However, an examination of frontier travel writing demonstrates that long before the “response to the West,” Chinese intellectuals were interested in exploring the ways in which other cultures challenged their own societal norms.
Qing literati expressed a penchant for reading about exotic geographies and for collecting ethnographic illustrations of exotic peoples such as the Miao of southwestern China or the “savages” of Taiwan. Their enthusiasm for the subject is reflected in the number of anthologies devoted to travel writing produced during the Qing. Travel accounts and pictures originally circulated in manuscript form among the author’s friends and colleagues, as well as among frontier officials. Many such manuscripts were subsequently published, either as part of an author’s collected works or in collectanea. Wang Xiqi’s mammoth Geographic Collectanea of the Little Fanghu Studio (Xiaofanghuzhai yudi congchao, published 1877), for example, reproduced over a thousand travel accounts representing exotic locales from Taiwan to Turfan.  Travel literature and topographical pictures thus appealed to a dual audience: frontier officials and others who needed practical geographic information and armchair travelers who sought to experience the thrills of the frontier vicariously.
Unfortunately, although a significant corpus of late imperial travel writing was preserved through reprinting in collecteana, due to the difficulties of reproduction, pictures were generally not included in such reprintings.  Subsequently, many pictures have been lost or exist only in rare manuscript editions in museums or private collections. Sometimes our only clue that pictures once accompanied a particular travel account is a colophon writer’s lament that “it is a pity that the pictures have long been lost.”  Thus, although travel literature and topographical pictures originally circulated within the same milieu, the modern reader of Qing travel anthologies generally reads as though texts existed in isolation from pictures—a practice reinforced by the academic distinction between literary studies and art history. I argue, however, that travel writing is best understood within a system of geographic representation that includes visual materials.
Travel and Visuality
Indeed, visuality plays a central role in the practices of travel and travel writing. As Mary Louise Pratt has shown in Imperial Eyes: Travel Writing and Transculturation, it is primarily through the sense of sight that the traveler constructs the other.  The claim to have “seen for oneself” is a recurring motif in late imperial Chinese travel literature, with travelers insisting that “I have been there myself and seen with my own eyes,” or “I am only recording that which I have seen with my own eyes.”  Another frequent move is for the traveler to refute common beliefs about a place based on what he has seen firsthand. As Ming literatus Zhang Hong explained after a journey to eastern Zhejiang in 1639, “About half [of the things I saw there] did not agree with what I had heard. So when I returned I got out some silk and used it to depict what I had seen, because relying on your ears is not as good as relying on your eyes.”  The privileging of the eye as the most reliable sense is related to the privileging of experiential knowledge in travel writing.  The ears, in contrast, are associated with hearsay, a type of knowledge regarded as particularly suspect in travel literature.  It is the traveler’s claim to have been an eyewitness to all he records that confers authority on the travel account. 
The role of the eyewitness acquired a special importance in Chinese accounts of the frontiers, as it did in European accounts of the New World. In the absence of a canon of texts concerning these “uncharted terrains,” only firsthand experience could lend credibility to the explorer’s report.  As one Qing literatus asserted:
Of all the books written about Taiwan, works such as Ji Qiguang’s Brief Account of Taiwan [sic] and Xu Huaizu’s Random Jottings on Taiwan are based on unsubstantiated rumor. . . . Only Lan Dingyuan’s Record of the Pacification of Taiwan and Huang Shujing’s Record of a Tour of Duty in the Taiwan Strait are written by men who really went there themselves and traversed the territory. Therefore, what they have to say about the mountains and streams, the environment, customs, and products can be trusted. 
Thus, among the various forms of geographic records available to Qing readers, travel writing, as a document of personal experience, had privileged status. Travel writers themselves frequently claimed that only those with firsthand experience could produce reliable geographic knowledge—thereby bolstering their own authority. Qing literati also considered it crucial for topographical pictures, especially maps, to be based on empirical observation. This does not mean that artists necessarily drew pictures from life. Rather, pictures might be produced (either by the traveler or by a professional painter) based on the traveler’s memory of what he had seen or perhaps samples of plants and other products that he had collected. Nonetheless, the basis in firsthand experience is what theoretically separated the topographical picture from an imaginative painting, just as it separated the travel account from fiction. 
Given the importance of “seeing for oneself,” it is not surprising that travelers like Zhang Hong chose to record their experiences in visual as well as textual forms. Indeed, art historians have written much on the association between domestic Chinese tourist travel and landscape painting (shanshui hua). But comparatively little has been written on the traveler’s involvement with the class of visual materials known as tu, a broad term that includes pictures, illustrations, maps, charts, and diagrams.  This neglect is largely due to the low status of vernacular visual forms within the discipline of art history. As art historian James Cahill writes of Chinese pictorial maps: “Many such picture-maps were painted in China from early to recent times, but they have received little attention from either Chinese or foreign scholars, because they have been considered (usually with good reason) to have practical rather than aesthetic value.”  At the same time, historians and scholars of cartography have traditionally discounted pictorial maps on the grounds that such maps have more aesthetic than practical value.  Pictorial maps and other kinds of topographical pictures have therefore largely fallen between the cracks of disciplinary divisions.
Craig Clunas has recently argued for the importance of pictures (tu) and visuality in early modern China and demonstrated that pictures permeated virtually every aspect of life in this period.  Certainly, visual materials, especially maps and astronomical charts, had long been considered vital to geographic knowledge in China.  By the Ming, illustrated books of all kinds, including geographic works, were widely available. Thus, readers were accustomed to looking at pictures in conjunction with texts. Robert Hegel’s study of Ming and Qing illustrated fiction has also greatly added to our understanding of the theory and practice of reading texts with pictures. Following the approach suggested by scholars like Clunas and Hegel, I seek to reinsert the reading of pictures into the reading of travel literature. 
The idea that topographical pictures represent an important complement to written texts in the production of geographic knowledge was succinctly expressed by Xia Xianlun, the compiler of a nineteenth-century collection of Taiwan maps (see Figs. 12 and 13):
The ancients had histories on their right and maps (tu) on their left, granting equal importance to visualizing and perusing (guanlan). When [Han dynasty general] Xiao He entered the passes, the first thing he did was to collect maps and written records. Without maps, one cannot have comprehensive knowledge of all the roads and their obstacles, of the terrain and its strategic passes. 
Xia emphasized the importance of looking in addition to reading, implying that images allow for a different way of comprehending space and place than words alone. Thus text and pictures are essential to one another: visual knowledge was an important counterpart to textual knowledge.
Figure 12. Xia Xianlun’s “General Map of the Cismontane and
Transmontane Territory” from Maps of Taiwan with
Figure 13. Xia Xianlun’s “General Map of the Transmontane
Territory” from Maps of Taiwan with Explanations (1879)
Word and image enjoyed a kind of complementary division of labor in late imperial geographic representation. Cordell Yee and others have demonstrated that gazetteer maps, for example, were primarily intended as illustrative accompaniments to the gazetteer text, which contained verbal descriptions of the geography. Maps provided a general idea of the topography and aided in understanding the spatial relations between these landmarks (see Figs. 14a and b). The text supplied such detailed information as distances between locations and the names of villages, mountains, and other topographical features. Map and text thus assumed complementary functions, with maps allowing for qualitative understandings of the terrain and texts providing quantitative geo-graphical information. This complementary relation was expressed in the idea that the “narration of events without maps is not clear, and maps without explanation are not intelligible.”  A similar dynamic can be seen in a genre known as “pictures with explanations” (tushuo), which combined pictures and explanatory text. Thus, word and image worked together in the production of geographic knowledge.
Figure 14a. Map of Fengshan County from the Taiwan Prefectural
Figure 14b. Map of Zhuluo County from the Taiwan Prefectural
The Qing court clearly recognized the political importance of visual in addition to textual knowledge of the frontiers. For the Qing, the visual representation of the frontiers, especially mapping, was bound up with the assertion of imperial power on both the practical and the symbolic levels.  And so the Qing sponsored works such as the Kangxi-Jesuit atlas and the Qing Imperial Tribute Illustrations. The Qianlong emperor also commissioned a series of French copperplate engravings to commemorate Qing conquests on the frontiers. Such pictures helped not only to visualize the extent of Qing imperial possessions but also to define, order, and celebrate these possessions. Pictures therefore played an important role in the fashioning of empire.
Although the surviving visual record of the Qing frontiers is far smaller than the textual record, it is no less significant. Like travel writing, pictures are a highly mediated form of representation: thus, these images reveal a great deal about how Qing travelers “saw” the frontier. In reinserting pictures into the reading of travel literature, I demonstrate that an examination of pictures may bring to the fore issues or perspectives that do not emerge from an examination of literary texts alone. Therefore, a more complete understanding of the cultural meanings that the frontier had for the Qing can be gained by examining pictures in conjunction with texts such as travel accounts and gazetteers.
As a study of Qing colonial discourse, my work necessarily privileges both texts and the perspective of the Qing elite.  This is largely a function of my sources: travel accounts, gazetteers, maps, pictures, and other documents produced by Qing literati. Due to this limitation, neither the perspective of the Taiwan indigenes nor that of the Han Chinese settlers is represented.
The Qing Transformation of Taiwan
After the Qing conquest of the island, the court’s Taiwan policy went through a number of phases over the course of the next two centuries. John Shepherd’s Statecraft and Political Economy on the Taiwan Frontier, 1600–1800 gives a detailed and thorough account of these policy shifts during the first century of Qing rule on Taiwan. According to Shepherd, the Qing court alternated between a pro-quarantine approach and a pro-colonization approach to Taiwan policy throughout this period.  Pro-quarantine policies sought to preserve the status quo on the island by restricting Chinese immigration to Taiwan and protecting the indigenes’ land rights. Pro-colonization policies promoted Chinese immigration and the aggressive appropriation of indigenous lands for Chinese settlers. Policymakers alternated between these two orientations, as they sought to balance the interests of the indigenous people and the Chinese settlers and thereby avoid costly conflict on the frontier.
Despite the efforts of pro-quarantine officials, the Qing could not stem the tide of Han Chinese immigration to this frontier, and by the nineteenth century, the court had decided to proceed with the final colonization of the island as a whole. In 1875, the Qing adopted the “Open the Mountains and Pacify the Savages” (kaishan fufan) policy.
This policy legalized the entry of Han Chinese settlers into the last of the remaining indigenous territory on the island. In order to accomplish this appropriation of lands, the Qing employed the military to “pacify the savages.” With the adoption of this policy, the tenuous balance between Han Chinese interests and indigenous interests definitively tipped in favor of the Chinese settlers. When Taiwan was promoted to provincehood in 1887, it seemed that the island was to be once and for all Chinese terrain.
When the Qing first conquered Taiwan, there were only a handful of firsthand accounts of the island. Thus, the Ming image of the island as a “ball of mud” predominated. Over the course of two centuries of Qing colonial rule, Chinese literati produced a significant corpus of travel accounts, maps, and pictures of Taiwan, providing a wealth of knowledge about the once-unknown island and concomitantly transforming its image. The pioneering Qing writers strove to make the island known and struggled with the question whether the island was worth colonizing. In the eighteenth century, when the issue of annexation had been settled, colonial officials recognized the need for accurate geographic information about the island. Eighteenth-century authors rejected the works of the earlier period as unreliable and attempted to replace these writings with their own empirical observations. It was during this second phase that the dominant tropes of Qing colonial discourse about Taiwan emerged. By the nineteenth century, Chinese attitudes toward Taiwan and the material conditions of the colony had changed so dramatically that “the ball of mud” was now considered a “land of Green Gold.” 
In 1895, only a short time after Taiwan had become an official province of China, the Qing were forced by their defeat in the Sino-Japanese war to cede the island to Japan. The reaction of Chinese elites to the signing of the Treaty of Shimonoseki demonstrates how far Chinese ideas about Taiwan had come since annexation. Officials and students in China vigorously protested the Treaty, signing declarations condemning what they called the “selling of national territory,” and the “severing of the nation.” Whereas Chinese officials two centuries earlier had protested the annexation of Taiwan as a waste of money, these protesters now declared that Taiwan should not be sold for any price. Pessimists predicted that once this piece of China was lost, the rest would soon fall like dominoes to imperial aggressors.
Writers also mourned the loss of Taiwan in their private writings. As one nostalgic traveler wrote: “Does this not hurt? Is this not cause for regret? How I blame the responsible officials who severed our national territory to end the [Sino-Japanese] war!.... I cannot stop the flood of my old tears. Alas!”  That the loss of Taiwan could evoke such emotions in writers like this reveals the profound change in the idea of Taiwan’s place in China’s imagined geography. The image of Taiwan severed, or cut off, from China implies that Taiwan had come to be conceptualized as an integral part of the Chinese geo-body. Taiwan was no longer a lone ball of mud beyond the seas, but a full part of China’s terrain, which would have to be “reunified” in order to make the national body whole again. The act of severing leaves a scar, a constant reminder of the pain of the knife: hence the “old tears” that our author sheds. The “loss” of the island to the Japanese only served to convert Chinese nationalists even more ardently to the idea that Taiwan rightfully belonged to China. The history of frontier travel literature thus traces the emergence of Chinese nationalist sentiment toward Taiwan.
The scar that was left when Japan annexed Taiwan was temporarily healed when China— this time the Republic of China—once again took possession of the island following World War II. But the wound would be opened four years later by the Chinese Civil War. With the victory of the Chinese Communists (CCP) imminent, the Chinese Nationalists (KMT) packed up their government, army, and many of the treasures of the Qing Imperial collections, and fled to Taiwan, establishing a provisional government of the Republic of China on the island.
It was out of this history of the Chinese Civil War that the current “Taiwan issue” emerged, with the KMT declaring the provisional government on Taiwan the legitimate government of “Free China,” and the PRC staking its claim over the “renegade province” of Taiwan. In order to bolster the legitimacy of these claims, both sides promoted the historically inaccurate contention that Taiwan has been “a part of China” since antiquity and effectively erased the rich history of the Qing colonization of the island. This Chinese nationalist discourse (both the KMT and the CCP varieties) has naturalized the idea that Taiwan is an inalienable part of China’s sovereign territory, a sacred part of the Chinese geo-body. The success of this discourse can be measured by the disappearance from the Chinese collective memory of the pre-Qing conviction that Taiwan was “beyond the pale.”
In Taiwan’s gradual transformation from a “savage island” into a “Chinese province” we see the profound changes in the imagined geography of the Chinese domain wrought by Qing expansionism. In the contemporary construction of Taiwan as a “renegade province” that must be “reunified” in order to restore China’s territorial integrity we see the lasting impact of Qing expansionism on the imagined geography of the modern Chinese nation-state.
Emma J. Teng is an Associate Professor of Chinese Studies at the Massachusetts Institute of Technology. She is the author of Taiwan’s Imagined Geography: Chinese Colonial Travel Writing, 1683–1895, from which the present text is drawn. She is currently working on Chinese and Chinese American representations of Eurasian interracialism at the turn of the 20th century. Her publications on this subject include “Eurasian Hybridity in Chinese Utopian Visions: From ‘One World’ to a ‘Society Based on Beauty’ and Beyond,” positions: east asia cultures critique (Volume 14.1, Spring 2006). Posted at Japan Focus on June 14, 2007.
1. There is some ambiguity in the English word “Chinese,” since it refers to two different Chinese concepts, Hua and Han. The latter is a more restrictive term that refers to the majority Han Chinese ethnic group.
2. Yu Yonghe, Pihai jiyou, p. 1.
3. “Zheng Chenggong” in Mandarin Chinese.
4. See Yu Yonghe, Pihai jiyou, p. 31.
5. Shi Lang, Jinghai jishi, p. 60.
6. Millward, Beyond the Pass, p. 37; Waldron, The Great Wall of China, pp. 42–43.
7. In this work I use the word “comparable” in the sense of “suitable for comparison.” See Perdue, “Comparing Empires,” pp. 255–61; Adas, “Imperialism and Colonialism in Comparative Perspective”; Harrell, “Introduction”; and Millward, “New Perspectives on the Qing Frontier.”
8. See Perdue, “Boundaries, Maps, and Movement.”
9. The term “renegade province” is used in U.S. media to represent the PRC’s position on Taiwan, but it is not a Chinese phrase. My thanks to Tom Christensen for pointing this out.
10. Foreign Desk, “China’s Statement,” p. 10.
11. See Christensen, “Posing Problems Without Catching Up.” See also Cohen and Teng, Let Taiwan Be Taiwan.
12. This is not to deny that ethnic Chinese families on Taiwan, including my own, have ancestral ties to China, just as many Americans have ancestral ties to England.
13. For a critique of this “China as victim” approach, see Millward, Beyond the Pass, p. 16. On the problematic of applying “postcolonial discourse” to China, see Lee, Shanghai Modern, pp. 308–9.
14. For example, a recent publication of the National Museum of Chinese History includes this statement under the heading of “National Unification”: “During the Qing Dynasty, the political, economic and cultural relations among various nationalities became increasingly harmonious, reinforcing the foundation of national unity” (Yu Weichao, A Journey into China’s Antiquity, p. 164).
15. Doyle, Empires, p. 45.
16. Parry, “Problems in Current Theories of Colonial Discourse,” p. 34; Thomas, Colonialism’s Culture, p. 9.
17. See the Epilogue.
18. Said, Culture and Imperialism, p. 9.
19. Spurr, The Rhetoric of Empire, p. 5.
20. Williams, Keywords, p. 160.
21. Said, Culture and Imperialism, p. 9.
22. Thomas, Colonialism’s Culture, p. 2.
23. As Peter Perdue (“Comparing Empires,” p. 255) has provocatively argued: “The Qing Empire of China was a colonial empire that ruled over a diverse collection of peoples with separate identities and deserves comparison with other empires.”
24. See ibid.; Perdue, “Boundaries, Maps, and Movement”; Shepherd, Statecraft and Political Economy on the Taiwan Frontier; and Hostetler, Qing Colonial Enterprise.
25. Ding, Dongying zhilüe, p. 79.
26. Hostetler, Qing Colonial Enterprise, pp. 37–41.
27. For example, the Qing required Han Chinese men to adopt the Manchu hairstyle of the queue.
28. On the multiethnic composition of the Qing military, see Crossley, “The Qianlong Retrospect of the Chinese-Martial (hanjun) Banners.”
29. In Taiwan, the Qing recruited indigenous troops for deployment against rebellious Han settlers. In response, Han Chinese settlers sometimes forged alliances with rival indigenous groups against these troops; see Shepherd, Statecraft and Political Economy on the Taiwan Frontier. For more recent work on “ethnic politics” on the Taiwan frontiers, see Ka, Fan toujia.
30. Thomas, Colonialism’s Culture; Stoler, “Rethinking Colonial Categories”; JanMohammed, “The Economy of Manichean Allegory”; Bhabha, The Location of Culture.
31. See, e.g., Millward, Beyond the Pass; Perdue, China Marches West; Gaubatz, Beyond the Great Wall; Shepherd, Statecraft and Political Economy on the Taiwan Frontier; Herman, “Empire in the Southwest”; Giersch, “‘A Motley Throng.’”
32. Said, Orientalism; Said, Culture and Imperialism; Lowe, Critical Terrains; Greenblatt, New World Encounters; Spurr, The Rhetoric of Empire.
33. All this is not to argue, of course, that Qing colonial discourse is identical to European Orientalism (which itself had many variations, British, French, etc.). It is simply to say that Orientalism was not a uniquely European phenomenon and that the Chinese had a roughly equivalent discourse by which they produced their others (a category that included Europeans). Indeed, the very idea of comparison undermines the Orientalist notion that China is “utterly different” from the West and therefore beyond comparison. At the same time, this is not to say that Western colonial discourse theory is universally applicable without regard to cultural specificity. This study attempts to walk the fine line between these two extreme positions.
34. Gladney, “Representing Nationality in China,” p. 94; Schein, “Gender and Internal Orientalism in China,” p. 70.
35. See Crossley, “Thinking About Ethnicity in Early Modern China.”
36. Despite the fact that the recent mapping of the human genome has demonstrated that there is no genetic basis for race, the current obsession with Tiger Woods shows that the idea of race is alive and well in twenty-first-century America.
37. For the development of nineteenth-century racial theory, especially the debate over monogenesis and polygenesis, see Young, Colonial Desire. See also Stocking, Race, Culture, and Evolution.
38. My research into issues of race and ethnicity builds on the work of scholars such as Pamela Kyle Crossley, Frank Dikötter, Stevan Harrell, Dru Gladney, and others. See Crossley, “Thinking About Ethnicity in Early Modern China”; Dikötter, The Discourse of Race in Modern China; Harrell, “Introduction”; and Gladney, Muslim Chinese.
39. Although I use the terms “race” and “ethnicity” in this work to describe Qing ideas about human difference, in referring to historical aspects of Qing frontier management I follow James Millward in using the term “ethnic.” As Millward (Beyond the Pass, p. 14n) notes: “In describing these categories in general terms, it is extremely convenient to have a single word for this sort of distinction.” Thus I speak of “ethnic policy,” “ethnic groups,” and “interethnic conflict.” Whenever possible, I have avoided the awkward construction “race/ethnicity.”
40. Crossley, “Thinking About Ethnicity in Early Modern China,” p. 8.
41. Ching, Becoming “Japanese.”
42. Said, Culture and Imperialism, p. 7.
43. Said, Orientalism, p. 54.
44. Schwartz, “The Geography Lesson,” p. 36; Godlewska, “Map, Text, and Image”;
Gregory, Geographical Imaginations.
45. Anderson, Imagined Communities, p. 15.
46. Thongchai, Siam Mapped, p. 17.
47. I propose that one key difference between the territoriality of nations and empires is that the boundaries of the nation are generally imagined as fixed, whereas the boundaries of the empire are often imagined as expandable.
48. Said, Orientalism, p. 54.
49. Liu-shi-qi, Fanshe caifeng tukao, p. 99.
50. Such information was either culled from the local gazetteers or based on officials’ own observations during tours of inspection.
51. Travelers also composed poetry, which they often included in their narrative accounts.
52. The notion of the “travelogue” (youji) as a distinct Chinese genre is essentially a modern invention.
53. These include the genres of shanshui hua (landscape painting), renwu hua (figure painting), huaniao hua (bird-and-flower painting), and jiehua (ruled-line or architectural drawing). The format of such works included the painted handscroll, the hanging scroll, the painted album, the painted fan, the woodblock print, and the line drawing. The tu produced by such professionals are finely rendered gongbi works, painted in color on paper or silk. Less privileged travelers, and gazetteer compilers who did not enjoy the services of a professional painter, had to settle for producing their own drawings. These drawings served as the basis for the woodblock prints reproduced in the gazetteers. The quality of tu thus ranges from the painterly to the sketchy.
54. Elman, “Geographical Research in the Ming-Ch’ing Period.”
55. See, e.g., Zeitlin, Historian of the Strange, p. 11.
56. Strassberg, Inscribed Landscapes.
57. Schafer, The Vermilion Bird, p. 128.
58. Yu Yonghe, Pihai jiyou, p. 27.
59. Dong, Taihai jianwen lu, p. 1.
60. Other similar, but less ambitious, anthologies include the Geographic Collectanea of Imperial Dynastic Barbarian Colonies (Huangchao fanshu yudi congshu) and the Five Collected Works on the Frontiers (Bian¬jiang wuzhong).
61. Wang Xiqi’s mammoth travel anthology, for example, contains no pictures.
62. Liu-shi-qi, Fanshe caifeng tukao, p. 20.
63. Pratt, Imperial Eyes, p. 5.
64. It was the desire to see for himself that motivated Yu Yonghe, for example.
65. Quoted in Cahill, Chinese Painting of the Late Ming Dynasty, p. 39. Zhang Hong’s comment further suggests that he viewed mimetic representation as an important function of painting. Zhang implied that what is pictured on silk directly correlates to, or imitates, that which is seen with the eyes. In representing “what he has seen” on silk, Zhang attempted to translate visual experience directly into visual artifact. In contrast, travel writing translates visual experience into words. Through the production of such an image, Zhang hoped to enable the viewer to “rely on the eyes,” thereby deriving more authoritative knowledge, perhaps, than could be found through reading a travelogue.
66. The importance of visuality in late imperial Chinese travel writing is underscored by the fact that vivid description is commonly praised as ruhua, “like a painting,” or rutu, “like a picture.”
67. Travelers, including the famous Xu Xiake, frequently mention being inspired to travel by a desire to ascertain whether what they have heard about a particular place is either right or wrong. Moreover, travel writers often emphasize that they record only what they personally witnessed and exclude any material that could be considered hearsay. Thus, despite the fact that one subgenre of travel writing is called wenjian lu (records of things heard and seen), I would argue that seeing is still the privileged term: that is, wen is no good without jian.
68. Similarly, Anthony Pagden (“Ius et Factum”) and Clifford Geertz (Works and Lives) have argued that it is through the sense of sight that the travel writer constructs the authority of his account.
69. See Pagden, “Ius et Factum.”
70. Wu Xiqi, “Preface,” in Zhai, Taiyang biji, p. 1.
71. The notion that only firsthand experience could guarantee the reliability of geographic information became a particular problem on the frontier. In Taiwan, for example, the dense jungle of the mountainous areas was often impenetrable to Chinese travelers. Such terrain was, therefore, impossible to map and difficult to document. When firsthand information was unattainable, hearsay had to substitute for empirical observations. For the frontiers, then, the expansion of geographic knowledge often went hand in hand with the extension of Chinese control.
72. Unlike landscape paintings, or other kinds of hua, which could stand on their own as objects of appreciation, these tu were generally produced and consumed in conjunction with written texts.
73. Cahill, Chinese Painting of the Late Ming Dynasty, pp. 206–7.
74. Cordell Yee is one notable exception.
75. Clunas, Pictures and Visuality in Early Modern China.
76. Maps and other tu were a central component of local gazetteers, as well as the more comprehensive Yitong zhi. Cosmological diagrams and charts were produced with the hopes of making visible the “patterns of Heaven” (tianwen). Emperors as far back as the Liang dynasty (502–77) commissioned the painting of tribute illustrations (zhigong tu) to record the appearance of foreign peoples. Illustrated versions of the Shanhaijing (Classic of mountains and seas) were in circulation at least by Tao Qian’s (365–427) time. Many of these tu have been analyzed by scholars of geography, cartography, and the history of science in China.
77. The role of the visual in anthropology has also become a subject of renewed interest in recent years; see Banks and Morphy, Rethinking Visual Anthropology.
78. Xia Xianlun, Taiwan yutu, p. 1.
79. Yee, “Chinese Maps in Political Culture,” p. 91.
80. Strategic maps, of course, greatly aided the Qing in the conquest of frontier lands.
81. I include both Han Chinese and Manchus who wrote in classical Chinese. Because of my own linguistic limitations, I do not use Manchu sources.
82. These terms are used by John Shepherd in his analysis of Qing Taiwan policy in Statecraft and Political Economy on the Taiwan Frontier.
83. “Green Gold” is a Chinese term for tea; see Etherington and Forster, Green Gold.
84. Chi Zhizheng, Quan Tai youji (Travelogue of all Taiwan), preface.
Adas, Michazel. “Imperialism and Colonialism in Comparative Perspective.” International History Review 20, no. 2 (1998): 371–88.
Anderson, Benedict. Imagined Communities: Reflections on the Origin and Spread of Nationalism. Rev. ed. London: Verso, 1991.
Banks, Marcus, and Howard Morphy, eds. Rethinking Visual Anthropology. New Haven: Yale University Press, 1997.
Bhabha, Homi K. The Location of Culture. London: Routledge, 1994.
Cahill, James. Chinese Painting of the Late Ming Dynasty, 1570–1644. New York: Weatherhill, 1982.
Chi Zhizheng. Quan Tai youji (Travelogue of all Taiwan). In Taiwan youji (Taiwan travelogues), ed. TYJYS, 1–17. TWWX, vol. 89. Taipei: Taiwan yinhang, 1960.
Ching, Leo T. S. Becoming “Japanese”: Colonial Taiwan and the Politics of Identity Formation. Berkeley: University of California Press, 2001.
Christensen, Thomas J. “Posing Problems Without Catching Up: China’s Rise and Challenge for U.S. Security Policy.” International Security 25, no. 4 (2001): 5–41.
Clunas, Craig. Pictures and Visuality in Early Modern China. Princeton: Princeton University Press, 1997.
Crossley, Pamela Kyle. “The Qianlong Retrospect of the Chinese-Martial (hanjun) Banners.” Late Imperial China 10, no. 1 (June 1989): 63–107.
———. “Thinking About Ethnicity in Early Modern China.” Late Imperial China 11, no. 1 (June 1990): 1–34.
Dikötter, Frank. The Discourse of Race in Modern China. Stanford: Stanford University Press, 1992.
Ding Shaoyi. Dongying zhilüe (Brief record of the Eastern Ocean). TWWX, vol. 2. Taipei: Taiwan yinhang, 1957.
Dong Tiangong. Taihai jianwen lu (Record of things seen and heard in the Taiwan Strait). TWWX, vol. 129. Taipei: Taiwan yinhang, 1961.
Doyle, Michael. Empires. Ithaca: Cornell University Press, 1986.
Elman, Benjamin A. “Geographical Research in the Ming-Ch’ing Period.” Monumenta Serica 35 (1981–83): 1–18.
Etherington, Dan M., and Keith Forster. Green Gold: The Political Economy of China’s Post-1949 Tea Industry. Hong Kong: Oxford University Press, 1993.
Fabian, Johannes. Time and the Other: How Anthropology Makes Its Object. New York: Columbia University Press, 1983.
Foreign Desk. “China’s Statement: ‘The Right to Resort to Any Necessary Means.’” New York Times, Feb. 22, 2000, Sect. A, p. 10.
Gaubatz, Piper Rae. Beyond the Great Wall: Urban Form and Transformation on the Chinese Frontiers. Stanford: Stanford University Press, 1996.
Giersch, C. Pat. “‘A Motley Throng’: Social Change on Southwest China’s Early Modern Frontier, 1700–1880.” Journal of Asian Studies 60, no. 1 (Feb. 2001): 67–94.
Gladney, Dru. Muslim Chinese: Ethnic Nationalism in the People’s Republic. Cambridge, Mass.: Council on East Asian Studies, Harvard University, 1991.
———. “Representing Nationality in China: Refiguring Majority/Minority Identities.” Journal of Asian Studies 53, no. 1 (1994): 92–123.
Godlewska, Anne. “Map, Text, and Image: The Mentality of Enlightened Conquerors. A New Look at the Description de l’Egypt.” Transactions of the Institute of British Geographers, n.s. 20 (1995): 5–28.
Greenblatt, Stephen, ed. New World Encounters. Berkeley: University of California Press, 1993.
Gregory, Derek. Geographical Imaginations. Cambridge, Mass.: Blackwell, 1994.
Harrell, Stevan. “Introduction: Civilizing Projects and the Reaction to Them.” In Cultural Encounters on China’s Ethnic Frontiers, ed. idem, 3–36. Seattle: University of Washington Press, 1995.
Herman, John E. “Empire in the Southwest: Early Qing Reforms to the Native Chieftain System.” Journal of Asian Studies 56, no. 1 (Feb. 1997): 47–74.
Hostetler, Laura. Qing Colonial Enterprise: Ethnography and Cartography in Early Modern China. Chicago: University of Chicago Press, 2001.
JanMohammed, Abdul R. “The Economy of Manichean Allegory: The Function of Racial Difference in Colonialist Literature.” In Critical Inquiry 12, no. 1 (1985): 59–87.
Ka Chih-ming. Fan toujia (The aborigine landlord: ethnic politics and aborigine land rights in Qing Taiwan). Taipei: Institute of Sociology, Academia Sinica, 2001.
Lee, Leo Ou-fan. Shanghai Modern: The Flowering of a New Urban Culture in China, 1930–1945. Cambridge, Mass.: Harvard University Press, 1999.
Liu-shi-qi. Fanshe caifeng tukao (Illustrations of the savage villages). TWWX, vol. 90. Taipei: Taiwan yinhang, 1961.
Lowe, Lisa. Critical Terrains: French and British Orientalism. Ithaca, N.Y.: Cornell University Press, 1991.
Millward, James A. Beyond the Pass: Economy, Ethnicity, and Empire in Qing Central Asia, 1759–1864. Stanford: Stanford University Press, 1998.
———. “New Perspectives on the Qing Frontier.” In Remapping China: Fissures in Historical Terrain, ed. Gail Hershatter, Emily Honig, Jonathan N. Lipman, and Randall Stross, 113–29. Stanford: Stanford University Press, 1996.
Pagden, Anthony. “Ius et Factum: Text and Experience in the Writings of Bartolome de Las Casas.” In New World Encounters, ed. Stephen Greenblatt, 85–100. Berkeley: University of California Press, 1993.
Parry, Benita. “Problems in Current Theories of Colonial Discourse.” Oxford Literary Review 9, no. 1–2 (1987): 27–58.
Perdue, Peter C. “Boundaries, Maps, and Movement: Chinese, Russian, and Mongolian Empires in Early Modern Central Eurasia.” International History Review 20, no. 2 (1998): 263–86.
———. “Comparing Empires: Manchu Colonialism.” International History Review 20, no. 2 (June 1998): 255–62.
Pratt, Mary Louise. Imperial Eyes: Travel Writing and Transculturation. New York: Routledge, 1992.
Said, Edward. Culture and Imperialism. New York: Vintage Books, 1993.
———. Orientalism. 1st ed. New York: Pantheon Books, 1978.
Schafer, Edward H. The Vermilion Bird: Tang Images of the South. Berkeley: University of California Press, 1967.
Schein, Louisa. “Gender and Internal Orientalism in China.” Modern China 23, no. 1 (1997): 69–98.
Schwartz, Joan M. “The Geography Lesson: Photographs and the Construction of Imaginative Geographies.” Journal of Historical Geography 22 (1996): 16–45.
Shepherd, John. Statecraft and Political Economy on the Taiwan Frontier, 1600–1800. Stanford: Stanford University Press, 1993.
Shi Lang. Jinghai jishi (Record of pacifying the seas). TWWX, vol. 13. Taipei: Taiwan yinhang, 1958.
Spurr, David. The Rhetoric of Empire: Colonial Discourse in Journalism, Travel Writing, and Imperial Administration. Durham, N.C.: Duke University Press, 1993.
Stocking, George W., Jr. Race, Culture, and Evolution: Essays in the History of Anthropology. Rev. ed. Chicago: University of Chicago Press, 1982.
Stoler, Ann Laura. “Rethinking Colonial Categories: European Communities and the Boundaries of Rule.” Comparative Studies in Society and History 31 (1989): 134–61.
Strassberg, Richard E. Inscribed Landscapes: Travel Writing from Imperial China. Trans. idem. Berkeley: University of California, 1994.
Thomas, Nicholas. Colonialism’s Culture: Anthropology, Travel, and Government. Princeton: Princeton University Press, 1994.
Thongchai Winichakul. Siam Mapped: A History of the Geo-Body of a Nation. Honolulu: University of Hawaii Press, 1994.
Williams, Raymond. Keywords: A Vocabulary of Culture and Society. Rev. ed. New York: Oxford University Press, 1983.
Xia Xianlun. Taiwan yutu (Geographic maps of Taiwan). TWWX, vol. 45. Taipei: Taiwan yinhang, 1959.
Yee, Cordell D. K. “Chinese Maps in Political Culture.” In The History of Cartography, vol. 2, book 2, Cartography in the Traditional East and Southeast Asian Societies, ed. J. B. Hartley and David Woodward, 71–95. Chicago: University of Chicago Press, 1994.
Young, Robert. Colonial Desire: Hybridity in Theory, Culture and Race. New York: Routledge, 1995.
Yu Weichao, ed. A Journey into China’s Antiquity, vol. 4. National Museum of Chinese History. Beijing: Morning Glory Publishers, 1997.
Yu Yonghe. Pihai jiyou (Small sea travelogue). TWWX, vol. 44. Taipei: Taiwan yinhang, 1959.
Zeitlin, Judith. Historian of the Strange: Pu Songling and the Chinese Classical Tale. Stanford: Stanford University Press, 1993.
Zhai Hao. Taiyang biji (Notation book on Taiwan). TWWX, vol. 20. Taipei: Taiwan yinhang, 1958.