When the H1N1 influenza pandemic began spreading around the world in April 2009, I was nearing the end of a year of ethnographic fieldwork at several local, government-affiliated public health institutions in and around Tianmai, China.Footnote 1 , Footnote 2 Tianmai is a large urban center located near the border between Mainland China and Hong Kong in the Pearl River Delta (PRD), the largely Cantonese-speaking area of southeastern China that was at the epicenter of the 2003 epidemic of Severe Acute Respiratory Syndrome (SARS). SARS, a novel and highly fatal influenza-like virus, killed about 800 people worldwide, including over 500 in China, over a period of 8 months from November 2002 through July 2003. Ever since the disappearance of SARS, global health authorities had been waiting for another deadly virus to emerge from the PRD, which had long been known as an incubator of novel flu strains (WHO 2004, 2005). And so, when reports emerged of a brand new influenza virus that looked poised to become a pandemic, some flu specialists were surprised to identify the first cases of H1N1 in California and to trace the virus’ probable origins to an industrial pig farm in Mexico (Mason 2010).

The epidemiologists, microbiologists, geneticists, and other professionals who worked in public health institutions in Tianmai were not so much surprised as relieved. Not only would they not be blamed for this pandemic, they thought, the virus’ origins in North America meant that it was unlikely to spread widely in China and thus to spur the kind of economic and political consequences that they had suffered during SARS—when Chinese bodies were stigmatized as infectious, trade and tourism ground to a halt, and Chinese businesses around the world experienced losses. H1N1 was “not a Chinese virus,” my informants repeatedly told me. Though they expected that the virus would eventually make its way into China, Tianmai’s public health professionals were equally confident that its geographical origins would accompany it. In an inverse of biomedical and epidemiological understandings of immunity, they told me that because it arose in North America, the virus’ “Euro-American genes” (oumeiren de jiyin) uniquely suited it to infect Euro-American (White or Mexican) bodies, shielding those without these genes from the worst of its wrath.Footnote 3 Unlike SARS, which became “Chinese” due to its geographical origins in China, Tianmai’s public health professionals told me that a virus that arose in North America would reflect that continent’s dominant racial group. “It seems that mostly white people [bairen] get it,” Dr. Yang, a microbiologist explained, “the susceptibility seems to have something to do with race.”

In this article, I trace how the race-making of people, viruses, and the places they share became a powerful means by which Chinese public health professionals made sense of infectious disease spread after SARS.Footnote 4 In the wake of the 2003 epidemic, in which international blame for a novel infection was placed on the backward predilections of a racialized Chinese population, Chinese public health professionals in Tianmai engaged in a racialization project of their own. They cast H1N1 as a EuroAmerican disease and diverted any blame for the spread of foreign viruses inside China to two racialized, mobile Chinese subgroups already associated with infectious disease: the Cantonese-speaking population largely blamed for SARS, and the rural-to-urban migrants who crowded into Tianmai and other PRD cities looking for work.

Public health workers both inside and outside of China have long employed the racialization of foreign, immigrant, or ethnic minority populations as a means to separate out the infectious from the non-infectious (Briggs and Mantini-Briggs 2003; Hyde 2007; Farmer 1999; Shah 2001). Mobile people in particular have often been the target of racialization discourses aimed at keeping them from crossing boundaries guarded by those in power (Massey 1994; Markel 1997; Briggs and Mantini-Briggs 2003). In the contemporary Chinese case, however, Tianmai public health professionals’ racialization of certain groups thought to be infectious joined with the racialization of the infections themselves. H1N1 could not easily infect most Chinese because both the virus and its hosts were racially alien.

This dual racialization process functioned as a tool with which Chinese public health professionals attempted to stabilize and contain viral threats. By inscribing geographical stability onto infected bodies in motion through the languages of race and genetics, my informants sought to constrain the mobility of infection and, in doing so, to contain the symbolic and material threats to China’s modernity and development that flu-like infections, and the people who carried and spread them, had come to represent.

A White Virus, a Mexican Virus—but Not a Chinese Virus

During the early days of the H1N1 pandemic, those working on disease control in Tianmai, along with the local media, took pains to emphasize the virus’ non-Chinese genetic genealogy (Palsson 2007). Health education videos and posters featured cartoon pigs and birds combining viral material in North America that then skipped across the ocean to China. Several informants told me that the apparent increased virulence of the virus among those infected in Mexico, and the failure of the virus to spread quickly to China, could be directly attributed to racial-genetic differences between Chinese and Mexicans (renzhong bu yiyang), as well as to the genetic origins and characteristics of the virus itself. A microbiologist, for example, explained to me,

Look at how many cases there are in the US. I think it’s because the virus’ genes have been found to be of Euro-American origins—so it is more virulent in people with those genes, in the US and Mexico and Europe—it passes more easily between people because they are better adapted to these genes. But in China it doesn’t pass very easily—it’s because our genes are different. So although it keeps spreading in the US, after a few generations here it gets weak—that’s why all our cases are imported.

Said a leader in Tianmai’s influenza control department:

The virus genes are Euro-American genes, so it’s easier for Americans to catch—it needs to mutate to suit our genes more closely before it can infect Chinese people on large scale. So after it gradually circulates here and develops Chinese genes maybe it will adapt to infect us. But right now it is not going to spread quickly. You can see that in Asian countries it is not spreading, I don’t think it is adapted to infect Asians.

The genetic immunity to H1N1 that Chinese in general supposedly enjoyed was contrasted with genetic compatibility between Chinese bodies and the SARS coronavirus. Explained one middle-aged infectious disease specialist, “[H1N1] hasn’t really spread to Asia—it’s been almost all foreigners. SARS was the opposite, it didn’t really get to Europe or North America, right?” When I reminded her that there was a major outbreak of SARS in Toronto, she continued, “Right, but there it was mostly huaren [people of Chinese descent], right? I think it has something to do with race [renzhong]. Because in SARS it was all Asians and few Europeans, and now it’s the opposite.

During the SARS outbreak, according to most of my interlocutors, Chinese people who immigrated to Toronto 30 years prior to the appearance of SARS were still genetically “Chinese,” and thus were vulnerable to a virus incubated in China in a way that White Canadians were not (if not quite as vulnerable as those who had never left China). Conversely, when H1N1 appeared in Mexico and then spread to the USA, it might travel to Tianmai and infect Mexican and American expatriates, but because of its origins in North America, Chinese people would, at least initially, be mostly immune.

“Mexicans”—members of a “race” that most Chinese had previously thought little about—were during the H1N1 outbreak both grouped with other North Americans as “Euro-American” and at the same time singled out as unique. One lab biologist, upon assigning a Euro-American and then a Mexican race to the new virus, paused to reflect on this classification decision before asking: “What race are Mexicans, anyway?” (Moxigeren daodi shi shenme renzhong?). A young epidemiologist from Shanghai who was training in Tianmai told me that as soon as H1N1 had spread to nearby Hong Kong and threatened to cross the border into Tianmai, the mayor of Tianmai called her infectious disease team and insisted that they track down all the Mexicans living in Tianmai and test them for H1N1, regardless of whether they had recently been to Mexico. Their Mexicanness automatically made them suspect, victims of their inherent compatibility with a Mexican-born virus.

Thus for all that global pandemic preparedness rhetoric emphasized the universal vulnerability of humanity to emerging infectious diseases (Chan 2007; King 2002; Lakoff 2008), H1N1 and SARS were, for Chinese public health professionals, remarkably discriminating. This is one reason why early in the initial H1N1 outbreak my informants displayed great confidence that H1N1 in fact could be both kept out of China and quickly brought under control. For them, the preferential infection of bodies associated with a virus’ home region assured that though a virus may leave its originating space, the harm it could inflict on foreign spaces would be limited. SARS spread around the world, and it went to Toronto, but it remained Chinese through its genetic makeup and ability to target Chinese genes. And so when H1N1 came, it was this tendency to travel with a racial group that my informants imagined would protect them.

Epidemiological Disjunctures

While discursive connections between infection, geography, and race have broad historical and global echoes, the particular racialization process that I describe here contrasts sharply with present-day epidemiological and immunological understandings of infectious disease. These biomedical understandings generally hold that those native to a place where a virus emerges and circulates are less, rather than more, likely to be infected because long-term exposure allows the body to adapt and learn how to defend itself. By repeatedly being exposed to local pathogens, natives of a particular place build up partial immunity to those pathogens. Insofar as genetics may be involved, the same principle applies: sharing a common geographic ancestry with a pathogen allows for genetic adaptation over time.

The gene for sickle cell anemia, for example, which also provides some protection against malaria, is thought to be more common in populations that have historically been exposed to malarial parasites. Exposure to this pathogen over time, the story goes, selected for a genetic adaptation that protected against malaria infection but had the unfortunate side effect of causing another debilitating disease. Thus, populations labeled as “black” in the USA are categorized as more at risk for sickle cell anemia than Caucasian populations, due to an assumed link between blackness and genetic African ancestry (see Fullwiley 2007, 2011; Wailoo 2001). As with H1N1 and SARS, in the case of genetic malaria protection, shared geographical origins link an infection to a genetic population through the language of race. The difference in the case of the former is that shared geographic ancestry with a pathogen implied adaptive susceptibility to that pathogen, rather than protection.

From a biomedical standpoint, then, Tianmai’s epidemiologists might seem very confused indeed. According to immunological principles, H1N1 and SARS hadn’t circulated nearly long enough to effect a genetic adaptation in any particular human population. Insofar as general immunological adaptation might occur, the longer H1N1 circulated in North America and not in Asia, the less susceptible those living in North America should become relative to Asians—and this should apply to Asian Americans and White Americans alike. So how was it that public health professionals with degrees in epidemiology, microbiology, and genetics all espoused views of infectious disease spread that seemed so contrary to epidemiological teachings?

One way of understanding this disconnect is to look at it as a sort of epidemiological bilingualism akin to the medical bilingualism common among Chinese biomedical practitioners and laypeople (Scheid 2002). That is, my informants subscribed to standard epidemiological explanations of infectious disease spread at the same time that they subscribed to the views described in the previous section. As Volker Scheid (2002) has shown, most Chinese patients and doctors move fluidly between Western and Traditional Chinese Medicine (TCM) explanations for disease, and modern Chinese medicine has not only mapped the Chinese body onto the Western anatomical body but has also mapped Western diseases onto the Chinese body (Kuriyama 1999). A similar type of fluid movement between languages and interpretations was evident in my conversations with Tianmai’s public health professionals. In the same conversations in which they warned me that I had to exercise more caution than they did when interacting with H1N1 contact cases (because as a White American, I was more susceptible than they were to catching the virus), they would also trot out the latest epidemiological models from WHO and CDC to show me how the disease was moving from continent to continent.

Developmental Threats and Anxieties

The racialization of viruses and people in China fed at least two deep anxieties on the part of many of Tianmai’s public health professionals: anxiety about the instability of China’s positioning as a modern, developed nation, and anxiety about their own unstable place in that nascent Chinese modernity. The 2003 SARS epidemic made clear that an infectious epidemic could threaten China’s goals of becoming “modern” (xiandai) and “developed” (fazhan)—two tropes that the central Chinese government had been heavily promoting since the Communist revolution and had explicitly made the cornerstone and raison d’etre of nearly all of its economic and social reforms since the death of Chairman Mao in 1976 (cf. Greenhalgh and Winckler 2005; Solinger 1999).

It is difficult to overstate the extent to which these twin goals of development and modernity—supported by a third goal of building a stable (wending) and harmonious (hexie) society—were central to Chinese politics in the 2000s when I was doing my fieldwork. This was not just empty rhetoric espoused by the central government. Spurred by memories of what they referred to as the “chaotic” (luan) and “backward” (luohou) Mao years, the professionals who worked in local public health institutions, like other members of China’s newly emergent middle class, both publicly and privately articulated fierce commitments to creating a prosperous nation that would adhere to accepted norms of modernity and would not be plagued by the kinds of instability that characterized its past.

SARS threatened all of this. On one level, it endangered China’s burgeoning modernity on a symbolic level by blaming Chinese consumers’ insufficient grasp of modern norms for the launch and spread of the SARS virus. Scientists both inside and outside of China blamed the zoonotic emergence of the SARS virus—that is, the virus’ leap from animals to humans—on Chinese consumption of civets, a raccoon-like mammal sold as a delicacy in some local “wet markets” in the Pearl River Delta region (Kan et al. 2005; Lau et al. 2005; Brookes 2005).Footnote 5 Although scientists later cleared civets of culpability—finding that humans infected civets, rather than the other way around (Janies et al. 2008)—a Western presumption of Chinese culpability in starting the SARS epidemic through the production, sale, and consumption of strange animals that do not belong on modern dinner tables never quite disappeared. As Mei Zhan points out in her 2005 analysis of SARS, “the story of ‘zoonotic [animal] origin’ did not blame nature itself for the SARS outbreak; what went wrong was the Chinese people’s uncanny affinity with the nonhuman and the wild” (Zhan 2005, p. 37).

As Charles Briggs points out, the American media sensationalized this so-called exotic consumption, such that “visual images powerfully link racialized bodies with descriptions of the disease … featur[ing] images of cooks cutting up animals not found on U.S. plates and customers selecting live, seemingly exotic animals for their meal (Briggs 2005, p. 276).” In a widely cited article in the New York Times published at the height of the epidemic in May 2003, for example, Keith Bradsher and Lawrence Altman wrote of “exotic animals” being sold as delicacies” in “filthy, crowded markets in Guangdong [province].” Civets, Bradsher and Altman said, were “trapped in the wild for Chinese kitchens”—setting up an implicit comparison with non-Chinese kitchens, where such abominations would never enter (2003, p. 1). The sale of civets inside crowded Chinese cities represented matter deeply out of place, as Mary Douglas would put it (2002): live animals, including strange, exotic animals not usually consumed by modern urbanites, were being sold in some of China’s most modern cities (see also Wald 2008).

And yet the relatively widespread sale and consumption of civets was a product of the same modernity that it seemed to threaten. Civets are not cheap. Despite the association of civet consumption with traditionalism, markets offering wild animals as culinary fare have proliferated only recently, and primarily only in southern China, as a result of “increasing affluence” among largely south Chinese consumers (Peiris et al. 2004, p. S89). In other words, the consumption of civets—and therefore the purported origin of SARS—was a paradoxical product of both traditional food choices and of recent modern monetary means to more frequently make those choices.

Beyond this powerful symbolism of a corrupted modernity, SARS also threatened China’s place as a political and economic powerhouse in the modern world economy. Racism against Chinese-looking people and the restaurants they owned led to losses during the epidemic of up to 60 % in New York City’s Chinatown (AsiaSource 2003). The more significant economic impact was felt in China itself. At the height of the epidemic, people panicked, trade and tourism nearly ground to a halt, and both the local and central Chinese governments suffered debilitating losses of face under a torrent of international criticism for their early mishandling of the crisis. The epidemiological mobility of a virus and its resultant racialization had shut down the productive mobility of China’s economic machine. In this way, the SARS epidemic of 2003 frightened Tianmai’s public health professionals more in its capacity for economic and political destruction than in its biological virulence.

My informants expressed these fears through the redirection of racial blame to enemies both internal and foreign. Several public health professionals theorized that foreign forces had exploited the predilection of SARS for Chinese bodies for evil purposes. Following popular conspiracy theories that circulated widely during the SARS outbreak, others told me that SARS was actually a biological weapon constructed for the purposes of harming the Chinese nation-state, via a targeted biological assault on Chinese bodies. “Many people don’t like China and want to attack us,” one such informant explained. “They don’t want us to be strong.”

Wen-Ching Sung (2010) notes that a best-selling book published in 2003, entitled The Last Line of Defense, suggested that SARS was an American invention that was designed to recognize “Chinese genes” (Zhongguoren jiyin) in order to specifically target the “Chinese race” (Zhongguo renzhong). In addition to preferentially killing off Chinese, some saw the stigma and discrimination against Chinese abroad that became associated with SARS as further evidence of a racial-biological attack with the intended consequence of holding back the success of biologically Chinese people, no matter where they lived.Footnote 6 Racially specific biological weapons in this imaginary were impressively powerful. By casting viral DNA as racially specific, my informants gave biological weapons the crucial power that has so far (at least according to mainstream theories) prevented them from being used widely: the ability to distinguish between self and enemy within the same space, to infect the bad guy and keep the attacker safe.

A few interlocutors surmised that H1N1, like SARS, also resulted from a bioterrorist attack—but that this time the attack was directed toward the former aggressor. They pointed to the timing of the origins of the disease in Mexico with President Obama’s trip there in April 2009 as evidence that terrorists had perhaps released the virus in a failed attempt to assassinate the President. Though it was targeted at Obama’s “mixed black and white genes,” however, H1N1’s Mexican origins ended up making it more deadly to Mexican natives, they told me. H1N1 was a EuroAmerican virus but also a specifically Mexican virus, due to its origins in a Mexican environment. It could have infected Obama, but it was unsurprising that it infected local Mexicans more effectively.

SARS, too, had a dual genetic identity: it was a Chinese virus that those both inside and outside of the public health world widely believed was adapted to infect members of the Han Chinese race. But it was also a specifically Cantonese virus, due to its origins in Guangdong Province—home to tens of millions of Cantonese, a racialized subpopulation of the Han. A common thread that runs through both biomedical and TCM origin stories of the SARS outbreak provided a basis for the diversion of racial blame to the Cantonese. Microbiologists and epidemiologists associated the emergence and spread of the SARS virus not just with Chinese people as a general category, but more specifically with the cultural practices and urban environments associated with southern Chinese people, especially Cantonese-speaking people from Guangdong Province. Similarly, practitioners of TCM blamed the SARS outbreak on the peculiar geography, culture, and type of body that TCM teachings also associate with southern China. Both origin stories promoted the view that the region of south China in which SARS originated fostered the development and the spread of infectious diseases, and both implied that these diseases might preferentially attach to south Chinese bodies. The resonance between these two origin stories promoted a logic of like-infects-like, with the originating continental and regional environments of both virus and carrier being the critical risk factors for infection. Thus, the internal racialization of Chinese subpopulations became a key component of the racialization of both Chinese and foreign viruses.

Racial Classification in China

Chinese revolutionaries first began promoting the concept of a unified Chinese “Han race,” as part of their opposition to the Qing dynasty, which fell in 1911, ending thousands of years of imperial rule. The leaders of the revolution considered the Qing to be outsiders due to their identity as members of the Manchu, an ethnic group concentrated in northeastern China. Anti-Manchu revolutionaries promoted Han nationalism as a way to define China in opposition to this hated imperial power, as well as in opposition to Japanese and Western colonial powers that vied for control of a weakened Chinese state. Han identity quickly became rooted in the presumption of a shared biological ancestry. Leibold argues that anti-Manchu revolutionaries in the late Qing transformed earlier dissidents’ ideas about a non-Manchu Han ethnicity into a “new hard boundary of blood cast around the ‘unsullied descendants of the Yellow Emperor’” (qingqing baibai huangdi zhi) (2011, p. 352; see also Dikotter 1992).

The problem with a biologically rooted Han-based nationalism, as some reformers soon realized, was that the idea that China belonged only to an immutable Han race threatened to alienate minority groups. Building upon the ideal of an inclusive “Chinese nationality” (Zhonghua minzu), Republican era leaders in the 1930s thus declared China to be a “Republic of Five Peoples” (wuzu gonghe): Han, Manchu, Mongolian, Tibetan, and Hui (Chinese Muslim) (Mullaney 2011; Leibold 2011). Supporters of this approach described China as a single nation (minzu) with both racial (zhongzu) and ethnic (minzu) diversity.

The Nationalist (GMD) and Communist (CCP) parties, which were fighting for control of China in the 1930s and 1940s, promoted different concepts of Chinese racial unity and diversity—the GMD promoting strict “monogenic similarity” as Mullaney puts it (2011, p. 54) with the CCP promoting polygenic diversity. In 1943, the GMD published a political manifesto, “China’s Destiny,” in which, according to Leibold, “all non-Han minorities were recast as ‘lineage branches’ (zongzu/zhizu) of a single, consanguineous Zhonghua minzu (Chinese nationality/race)” (2011, p. 360). This unified Chinese race was in turn the most significant member of an even more inclusive “yellow race” (huangzhong) common to all of Asia. Thus, Han, Tibetans, Manchus, and others all became racially Chinese and Asian at the same time, while also maintaining some elements of intra-Chinese ancestral uniqueness. The yellow race was described in opposition to other continentally based races, including Black (hei) and White (bai).

When the CCP won control of China in 1949, however, internal diversity became the focus of Chinese ancestry studies, and the language of race blended into the language of “nationality” (minzu, sometimes translated as “ethnicity”). Seeking to bolster the concept of China as a “unified multiethnic nation-state” (tongyi de duo minzu guojia), researchers set to work classifying the nationalities, or ethnic groups, living within the new republic. As Mullaney describes, however, “the Communists unwittingly stumbled upon a remarkably complex problem: if China was home to multiple minzu, who were they?” (2011, p. 56).

Teams of social scientists spent decades trying to answer this question, eventually completing a list of 56 officially recognized minzu. Of these, 55 were shaoshu minzu (minority nationalities), in contrast to a Han majority. Notably, the Cantonese were categorized as an integral part of the Han population and were not considered for minzu status (Carrico 2012). Steve Harrell has argued that, “Hanness is like Whiteness in the United States; it is an unmarked characteristic that can be delineated only in contrast to an ethnic other.” (2001, p. 295). As Karin Brodkin describes for Jews in the USA (1998), Cantonese blended into the dominant Han population even as they recognized themselves, and were recognized by others, as both culturally and biologically distinct.

Though the minzu categories quickly took on racial connotations, with those claiming Han status frequently citing differences in physical features and bio-genetic ancestry as indicative of membership in different minzu groups (see Sung 2010), researchers at the time did not turn to biology or physiology in categorizing the 56 nationalities. Instead they drew upon Stalin’s model of the “four commonalities”: common language, territory, mode of economic production, and psychology (Mullaney 2011). The importance of common territory formed the basis for the establishment of five autonomous regions in the Western part of China where single minority groups were concentrated in high proportions (Tibet (Tibetan or Zang), Xinjiang (Uyghur), Ningxia (Hui), Guangxi (Zhuang), and Inner Mongolia (Mongol or Meng)). The establishment of the autonomous regions helped to reinforce a spatial and geographical dichotomy of center and periphery—the center being the Han, symbolically and physically located in China’s major population centers; the periphery being the other 55 nationalities, symbolically and physically located in the peripheral parts of the nation-state. At the same time, the naturalization of the 56 number became a means of permanently tying the periphery to the center: if Tibetans are one of the 56 that make up China, that made Tibet an inalienable part of a multi-ethnic whole and a greater Chinese race (Litzinger 2000; Harrell 2001; Mullaney 2011).

The minzu categorization process immobilized most racial-ethnic minority groups in China, keeping them, along with their diseases, securely in their proper places. Sandra Hyde, for example, describes how the HIV virus in China was imagined as residing specifically in certain regions of China considered the “periphery” and attaching to shaoshu minzu bodies. The unique environments of southwestern China, combined with the exoticized and sexualized behaviors and dispositions attributed to minority peoples, were jointly associated with the emergence and spread of HIV. The virus seemed to preferentially infect these not-quite-Chinese bodies living in their not-quite-Chinese home locales (Hyde 2007).

On the other hand, the shaoshu minzu rarely emerged as particularly at-risk populations in Tianmai. Most migrants to Tianmai were Han, and most minorities that did live in Tianmai had the ability to “pass,” thus rendering their minority status somewhat invisible (see Harrell 2001). Their minzu identities placed them on the frontier, their frontier status assured that they would stay put, and when they did not abide by this for the most part, shaoshu minzu migrants ceased to be identifiable as minorities at all.

Thus for public health professionals in Tianmai, the official 55 minority groups were relatively inconsequential to local disease spread or control. Far more significant were two populations that were officially members of the Han majority: southern, or more specifically, Cantonese populations, and rural-to-urban migrant workers.

Dark, Clever, and Sick

Wen-Ching Sung describes how scientists working on the Human Genome Diversity Project in China, in addition to documenting genetic differences among minzu groups, also “reported a clear genetic distinction between southern and northern Chinese populations.” (2010, p. 277; see also Reardon 2004). Leibold (2011) noted that while “a group of Chinese geneticists now argue that ‘Y chromosome and mitochondrial (mt)DNA data have demonstrated a coherent genetic structure of all Han Chinese,’” this structure still “exhibits a distinct pattern of variation among northern and southern Han people.” And Gladney (2004) points out that most people in China distinguish physiologically between Northern and Southern Chinese, and especially between Northerners and Cantonese. Public health professionals hailing from both Southern and Northern regions frequently spoke of racialized differences in facial features and stature between Northern and Southern Chinese, the latter (especially the Cantonese) being described generally as darker, shorter, and less beautiful than their northern brethren, but also more “clever” (congming) and, historically, more mobile. Gladney points out that these attributes are given racial-genetic significance, despite official assertions that the Han is racially homogenous (2004, p. 433). According to Carrico, some Cantonese have embraced this claim of genetic difference, asserting that they have a separate ancestry from ordinary Han and even declaring themselves to be “an independent race” (2012, p. 39).

These kinds of racialized distinctions between southern and northern Chinese also applied to the diseases from which they suffered. For example, Traditional Chinese Medicine has long linked so-called “warm diseases” (wenbing)—which include acute upper respiratory infections like SARS—with southern China and in particular with southern Chinese bodies (Hanson 2011). Southern bodies were thought to be constitutionally weak and soft, and southern people were described in TCM texts as tending toward indulgence and excessiveness. Foreigners and those from outside the south Chinese region did not become ill with wenbing as easily because their constitutions were not as well adapted to local disease patterns. Wenbing was a southern Chinese disease.

Historian Marta Hanson details how, given this background, TCM practitioners explained the emergence of SARS in these terms: “In southern China, a warm damp climate keeps the defensive and constructive qi [vital energy] of the local population at the surface.” (2010, p. 245). But because of a cold snap that occurred in the spring of 2003, the warm constitutions of the southern (especially Cantonese) population was out of step with the unusually cold environment, rendering them more susceptible to the entrance of “heat-wind,” (refeng) which could bring on flu-like symptoms. Hanson explains, “According to some TCM physicians, this susceptibility was increased by the allegedly sedentary habits and rich diet of Guangdong’s modern urbanites.” (2010, p. 242). The congruence of southern Chinese bodies, habits, environments, and climates together created the disease that became known as SARS.

Hanson’s account resonates with biomedical understandings of SARS’ origins; together the two accounts reinforced a racialized narrative of blame that centered around the deficient behavior and bodies of Cantonese people who ate strange animals. Racial distinctions within China were, in an international context, usually erased by the more easily observable racial distinctions between Chinese and non-Chinese. Because foreign media rarely differentiated between Chinese bodies in general and Cantonese bodies in particular, the supposed sins of the Cantonese quickly spread to anyone who looked Chinese, leading to discrimination against Chinese and Chinese-looking people in the USA and elsewhere (Kleinman and Lee 2006, p. 181). However, the Cantonese-ness of SARS was very important to those of my informants who were not themselves Cantonese.

Tianmai’s non-Cantonese public health professionals acknowledged that SARS was Chinese, but insisted that it was also Cantonese; its origins could be found not in ordinary Chinese behavior, but rather in the practices of a racialized subgroup that they both admired and reviled. Cantonese consumers, more so than most other Chinese, had enough money to enjoy a middle-class lifestyle—a valued aspect of “developed” societies. It was how they chose to pursue this lifestyle that did not sit well with normative Western concepts of what developed societies should act like—norms that many Chinese were trying to attain. At the same time, in keeping with TCM explanations for SARS, the decadent practices of Cantonese people rendered them more likely to spark, spread, and suffer from local pestilences.

Kevin Carrico notes that instances of “scapegoating Cantonese hedonism” occurred throughout China during the SARS epidemic, with one commentator referring to the SARS outbreak as “the revenge of the heavens’ for the Cantonese people’s decadent lifestyles” (2012, p. 35). Because they were weak and indulgent, Cantonese people both ate civets and fell victim to them. Unsurprisingly, the public health professionals I knew who had Cantonese origins tended to ignore or underplay this specificity and instead emphasized the unity of Chinese as common victims.

In keeping with these patterns of blame, some of Tianmai’s non-Cantonese public health professionals attributed the relative resistance of their entire city to SARS to its status as a “city of immigrants” (yimin chengshi). Though surrounded on all sides by Cantonese communities, Tianmai was not a Cantonese city. While it had its share of Cantonese residents, a large majority of the population had migrated to the region from other parts of China. According to some of my informants, its migrant status helped to keep the city resistant to SARS: non-Cantonese simply were not very susceptible to a Cantonese disease. Thus, the leaders of Tianmai’s public health institutions, most of whom were not Cantonese, boasted that Tianmai, despite being what one leader called “the filling in the cookie” between Guangzhou (Canton City) and Hong Kong, was the only one of the three cities to have had no deaths and only 56 cases during the SARS epidemic—compared with at least 1,000 cases each in Guangzhou and Hong Kong, along with many deaths.

Some of my informants applied the same logic to H1N1. As H1N1 infections began appearing inside Tianmai in June 2009, some of the epidemiologists and flu virologists I knew asserted that all of the cases in Tianmai were shuruxing bingli, or “imported cases,” which they opposed to the “community outbreaks” that nearby cities were experiencing by that time. Drawing upon the legacy of SARS, they told me that Tianmai’s immigrant identity would, at least in part, protect the city.

Though the H1N1 virus was labeled as North American, once it began circulating inside China, some still saw it as having a stronger affinity for Cantonese bodies than for other Chinese bodies—thus explaining why Guangzhou had homegrown cases while Tianmai did not. A few of the Northern public health professionals I interviewed vaguely blamed the emergence of H1N1 on the same type of dangerous Cantonese predilections that had given birth to SARS, even while insisting that it was a North American disease. For example, Dr. Gong, a public health professional from Shandong Province in northern China who headed a small department in one of Tianmai’s disease control institutes, told me soon after H1N1 arrived in China, “this new virus is a result of nature getting angry with us, and us not treating animals right, having them be too cooped up, and we eat too many of them—I mean Cantonese people eat everything—is that really necessary?” She went on to emphasize again that H1N1 was spreading within Guangzhou and not Tianmai.

Here, Gong picks up on a theme from the SARS epidemic, in which the overindulgent Cantonese consumer launches an epidemic through his unnatural appetites. In this case, however, she extends the logic of Cantonese culpability to viruses that everyone agreed Cantonese behaviors did not create. For Gong, H1N1 was a North American virus, but it still somehow adhered more readily to Cantonese people than to other Chinese, due to a common origin story in the mistreatment of animals and nature. Gong also used this story to place herself—she may be living in a Cantonese region, but she was not Cantonese. Instead, like other transplants that migrated to Tianmai from the North, she carried with her more civilized customs and more sanitary cuisine. This time, mobility protected rather than exposed. In racializing herself through her own mobility, Gong allowed place to travel through space with her, protecting her from infection through a set of Northern genes that were perhaps a bit less compatible with the diseases of animals.

Mobility, Migration, and the Racialization of the Floating Population

As the case of Tianmai shows, the hypermobility of bodies, things, and microbes that has accompanied China’s rapid development and rapid integration into global flows meant that the geographical distinctiveness of racialized groups—as implied by the racial categories of “EuroAmerican” and “Chinese,” as well as “Northern” and “Southern”—was highly tenuous. SARS may have arisen in a Cantonese region , but Cantonese people might live anywhere in China or even anywhere in the world—transforming geographical spaces that might have once seemed safe from Cantonese diseases into places at risk. In the epidemiological imaginary, mobility was by definition infectious. Where mobile people moved, and stayed, they brought the viruses that had attached to them. At first, their racial compatibility might contain the virus, but eventually, if they stayed long enough, new diseases might take hold and circulate in places they did not originally belong. Northerners might well begin to suffer from hot winds.

The Cantonese brought SARS out of Canton, their racialization both emerging from their mobility and serving as a tool with which to symbolically hold them in place. But it was another mobile subgroup of Han that seemed most capable of allowing the 2009 H1N1 virus to take hold inside Tianmai and to eventually transform into a “Chinese virus”: the city’s floating population of rural-to-urban migrant workers.

Of Tianmai’s estimated 20 million population, 60 to 80 % are considered part of the “floating population” (liudong renkou) of rural-to-urban migrant workers who began flooding China’s large cities in the 1990s and 2000s, looking for economic opportunity (Mason 2012). The hukou system of household registration, established in the 1950s as a means for the Communist government to control the mobility of Chinese citizens, impeded the floating population’s ability to migrate permanently, effectively creating a class of illegal internal immigrants within all of China’s major cities. Violating one’s hukou to move to the cities without a proper permit could, until very recently,  subject a rural migrant to deportation back to the countryside.Footnote 7

In addition to the floating population, Tianmai also had a sizable population of permanent, middle-class, hukou-holding migrants, many of whom came from the same parts of China from which the floating population came, and many of whom also had rural backgrounds. It was this latter group that included most of the public health professionals I knew in Tianmai. However, as migrants themselves who had only recently escaped poverty into the middle class, many of those who worked in public health in Tianmai felt that their own identities as “modern” and “developed” urban citizens—statuses they had obtained through a combination of education, skills, connections, and luck—were frighteningly unstable. The large numbers of rural migrants who filled their clinics and crowded into tenement housing seemed chaotic, unruly, and backwards; they represented both that which they had fled and that which they feared they still in a sense were. Tianmai’s public health professionals reactively maligned this backward threat as dirty, dangerous, and backwards (see Mason 2012). Even as floating migrants contributed much of the labor crucial to the Chinese development project and were officially included in this project’s vision (Lee 2014), they also were cast as that project’s Achilles heel, the piece capable of bringing the whole structure tumbling down. Tianmai thus became a site where boundaries between “good,” modern migrants like public health professionals and “bad,” backward migrants like the floating population were widely drawn and carefully guarded (Mason 2012; Infectious Change: Reinventing Chinese Public Health after an Epidemic (unpublished)). One of the ways that middle class Tianmai citizens like my informants guarded these boundaries was through a process of racialization.

My own introduction to the racialization of floating migrants came when I was teaching English at a university in Guangzhou in 2001. A student who resided in Tianmai invited me to tour a China Folk Culture village (Zhongguo minsu wenhua cun) in her hometown. We boarded a bus mostly full of what I only several years later was able to identify as rural migrant workers bound for the city. As we approached the city limits, two police officers boarded the bus, checked my passport, and checked everyone else’s identification cards. Two dark-skinned men with short stature, carrying battered plastic plaid tote bags, were quietly escorted off the bus. I was young, I was not yet an anthropologist, and I did not think about why these men were headed to Tianmai, what they might plan to do there, or why their ID cards were scrutinized so thoroughly.

Dong Han argues that the racial profiling of Han migrant workers that I observed on the bus that day is common to many Chinese cities, in which poor migrants have become the most visible racial Other, a stand-in for all that is threatening to the Chinese middle class. Han writes, “By relying on bodily features to identify subjects of hukou enforcement, policing sets up direct links between physical appearance and particular social, economic and political status” (2010, p. 603). As Brodkin has argued in the case of the USA, the construction of race among immigrant populations goes hand and hand with the construction of class, such that differentiating by color or appearance provides a means of justifying the oppression of a working underclass (1998; see also Omi and Winant 1994). Smaller and darker on average than urban Chinese due to a history of outdoor labor and poor nutrition, and cast as dirty and “low quality” (suzhi di), the rural Han trying to circulate into and out of Chinese cities formed a massive racial underclass that was paradoxically composed of members of the dominant Han race (see Han 2010; Solinger 1999).

At the China Folk Culture Village, however, this massive underclass was nowhere to be seen. Instead, all 55 of China’s ethnic minorities were on proud view here. As small children pointed excitedly the way they might point at elephants in a zoo, young women declaring themselves to be Miao, Yi, and Hui dressed in colorful traditional costumes, danced, and played instruments (Chio 2014). As we watched the spectacle, my student leaned over and whispered that all of the performers were likely really Han. In Tianmai, there were very few Miao, Yi, or Hui people to play the parts, she explained. There were, however, many millions of Han migrants like the ones who were taken off of the bus, looking for work. The rural-to-urban migrants dressed in minority costumes illustrated how officially recognized racial or ethnic categories often diverged from, and at the same time merged with, the categories that took on the most significance in practice. Here, two different kinds of racialized Others merged into each other, the mundane but highly mobile Other coming to stand-in for the exotic that never quite managed to leave home.

The racialized mobility that characterized the floating population seemed to my Tianmai informants to be putting their city at risk. While “good” migrants provided resistance to Cantonese infections, the floating migrants, in their failure to become permanent residents or to embody modern citizenship, carried susceptibility. Due to their huge numbers, harsh working and living conditions, and the inability of government officials to control their mobility, Cantonese and non-Cantonese public health professionals alike saw the members of the floating population as out-of-control viral vectors, capable of spreading even foreign diseases to Tianmai’s permanent, legal population of middle-class migrants.

What was most unsettling to some of my informants was that unlike the shaoshu minzu or even the Cantonese, the true “home” of the floating migrants was exceptionally difficult to identify. Because they were a mix of Han from all over China, the floating population’s racialization explicitly did not translate to a coherent genetic or ancestral difference. It also did not necessarily translate to a difference in geographical location: the floating population was officially rural, but it also made up a large proportion of the populations of most major urban centers in China. What’s more, its members often hailed from the same areas of rural China from which my now-middle-class informants—both Cantonese and non-Cantonese—once migrated. This was especially the case for older public health professionals, almost all of whom had experienced deep rural poverty during their childhood and young adult years, and thus shared more with the migrants than they cared to admit. Thus, the racialization of the floating migrants remained incomplete. Other than the very fact of their continued mobility, it was difficult to differentiate between Us and Them, when Us and Them came from the same places and were thought to share the same genes.

The “low quality,” lack of education, and poor living and working conditions of the floating population made rural-to-urban migrants seem more susceptible to H1N1 than even the Cantonese. At the same time, presumed racial-genetic and ancestral similarities between floating migrants and middle-class migrants to Tianmai allowed for the transfer of foreign germs from “bad” migrants to “good.” The floating population was like an inverted Trojan horse—Other on the outside, but on the inside dangerously the same. I was told over and over again during this time period that because rural-to-urban migrants were uniquely capable of spreading H1N1 widely inside Tianmai, containment efforts had to be focused on keeping the virus out of this population:

Said one epidemiologist:

[The migrants] are a really special problem here—everyone is terrified of them! Because there is no way to keep track of them, no one has any idea where they are, and if there is really a pandemic, then we’re in big trouble! Their wenhua [cultural level] is low, a lot are from the countryside, they don’t understand basic biological facts, and they don’t have any responsibility, no sense of that at all, and so if we come looking for them to check up if we think they have flu, they’ll just think, ‘you’re trying to do what??’ and they’ll run away to some other place and go find work there.

Another epidemiologist in Tianmai who himself had grown up in a poor rural village told me, “If the virus got into that population, we would never be able to control it. So we have to work hard to keep the disease out of the floating population. You know, we are still quite backwards” (emphasis added). The floating population represented the backwardness within on two levels: the backwardness within a modern China, and the backwardness within my informants’ own modern selves. Tianmai’s public health professionals feared that with the arrival of H1N1, both might come to subvert and destabilize China’s modernization project.


The racialization of humans and viruses in China inscribed relative geographical place onto racialized groups and in the process served to partially immobilize mobile infectious threats. Most racial Others could not infect the Han majority because their similarly racialized viruses had never quite left the place that it was imagined made them Other in the first place. The shared geographical origins of virus and carrier not only provided both with racial identities but also tethered them both to a common racial home through deep biological ties that transcended physical space. Infectiousness thus became a test of racial purity, and vice versa. Foreign viruses could cross geopolitical borders inside the bodies of mobile Others, but in the Chinese public health imaginary, they were least likely to cross into the bodies of those who truly belonged.

But this immunity of dissimilitude was partial, and it was time-limited. Chinese public health professionals, many of them having earned degrees in epidemiology from Western institutions, knew that viruses spread, and that they adapted. If they were not careful, if they did not take great care to stop carriers of the newest foreign virus from spreading the disease inside China’s borders, and from replicating inside the hybrid bodies of those Han-who-were-not-quite-Han, it would only be a matter of time before H1N1 did transform into a Chinese virus.

When H1N1 finally landed in Tianmai, it arrived in the lungs of two Chinese American tourists, who had come to China for the first time, to visit relatives in a nearby city. My informants referred to the girls as “American Chinese” (meiguo huaren), who had caught the virus by virtue of their Americanness and now stood to spread it to Tianmai by virtue of their Chineseness. As mobile bodies with a Cantonese heritage that had nevertheless adapted to a foreign virus, they were uniquely well-positioned to transform the H1N1 virus from a foreign into a Chinese pathogen, their bodies serving as figurative and literal bridges between America and China. Genetically and physiologically Han, they were geographically and culturally foreign. They were raced as Chinese, and yet their bodies had known only North American terrain, and only North American viruses—representing a radical divergence of genetics and geography. As Cantonese who were foreign to China, as mobile people twice over, they were an especially dangerous form of hybrid.

Faced with the arrival of H1N1 at their doorstep, Tianmai’s public health professionals threw themselves into the containment of H1N1 to a degree that few others in China or elsewhere were willing to do. All other disease control work ground to a halt. Young public health workers were sent to the border with Hong Kong, and to the receiving hospital, where they lived in dormitories and remained on-call to interview suspect cases all night long. Contacts of suspect cases were placed in quarantine, and workers were assigned to long shifts at these facilities as well. Even after the virus began spreading inside Tianmai, public health professionals spent long weekends calling up every person who had crossed the border from Hong Kong and listed Tianmai as their destination, just to make sure that he or she wasn’t feeling sick.

All of these measures together created the impression of an overreaction to what in the end the World Health Organization characterized as a relatively minor pandemic. International organizations and foreign governments criticized China for quarantining people unnecessarily and subjecting travelers to unnecessary tests and harassment. And yet in the end, though H1N1 did eventually spread inside China, and though it did infect and kill people from all regions of China, it killed fewer people overall, and wrought less damage, than in many other parts of the world (Wong 2009; Lurie 2009). Though China’s modernity was again briefly questioned under a barrage of criticism about the use of “nineteenth century” quarantining techniques (Mason 2010), it would seem that, perhaps thanks to the efforts of Tianmai’s public health professionals, H1N1 never did become a Chinese virus. “This virus is no SARS,” a Tianmai epidemiologist explained to me as the pandemic wound down and life began to return to normal. “It is nothing for us Chinese to be afraid of.”