Issue Number Three: Sentinel Devices

The Birds of Poyang Lake: Sentinels at the interface of wild and domestic

Lyle Fearnley looks at what happens when farmers draw a line between wild and domestic that scientists miss.

The birds of Poyang Lake, do you know their number?
Flying they obscure the sun and moon
Alighting the lake-grasses are concealed.[1]

The Poyang Lake in southern China is renowned for its flourishing birds. Each winter, over a million migratory waterfowl arrive from the north seeking open water and fresh grasses. One of China’s first wildlife protection areas was founded on one section of the lake. Over the last three decades, however, small and medium-scale poultry farming rapidly expanded in the lake region. When fears of a global influenza pandemic grew, scientists identified Poyang Lake as a source of possible influenza emergence, fearing that transmission of viruses between wild and domestic birds could produce the next pandemic strain. Although to an anthropologist the contrast between wild and domestic suggests the classical oppositions of the raw and the cooked, of nature and culture, influenza researchers had far more urgent concerns: to make this contagious relationship into a sentinel for the pandemic.

Back in the 1960s and 1970s, Robert Webster and Kennedy Shortridge proposed that birds (and especially domestic waterfowl) could serve as sentinels for human flu pandemics. Their research transposed the temporal progression of pandemic emergence onto the categorical distinctions between species of living beings, the “frontiers of the living” (Keck 2010). An emergent influenza virus begins in wild bird reservoirs, they theorized, mutating and reassorting through domestic poultry and pigs before appearing in human populations. This suggested one could anticipate the next human pandemic by monitoring virus and disease in poultry. The emergence of a novel highly pathogenic avian influenza strain (HPAI H5N1), and the isolation of the novel virus from domestic poultry, largely confirmed this hypothesis.[2]

Shortridge had also argued that southern China was the “epicenter” of influenza viruses. The southern ChiDEVICESnese ecology and agriculture system entailed close contact between ducks, chickens, pigs and humans, which he believed promoted the emergence of new viruses (Shortridge and Harris 1982). When in 1997 the HPAI H5N1 strain first infected humans, global concern grew and increased funding for research followed. Scientists began to disaggregate southern China into zones of greater and lesser influenza risk, refining the location of possible sources of viral emergence. During my fieldwork in China, many researchers pointed to Poyang Lake as one such epicenter, citing scientific findings that traced the origin of internal genes from the HPAI H5N1 strain to influenza viruses isolated around Poyang and the discovery of the virus in healthy wild birds at the lake. Beginning in 2008, a group of researchers, including both international and Chinese scientists, set out to develop sentinels for influenza emergence at Poyang Lake.

Unlike Webster and Shortridge, however, they did not prioritize the detection of novel viruses. Instead, they monitored the structural conditions that produce highly pathogenic forms of the virus, a turn of attention from viral phylogeny to disease ecology. This monitoring of the ecological conditions of diseases focused on one relation in the system above all others: what they called the “wild waterfowl-domestic poultry interface” (Xiao et al. 2010). In a 2010 review article, the Poyang Lake researchers argued that a “key factor integral to the evolution of low pathogenic avian influenza into highly pathogenic avian influenza is the interaction between wild birds and poultry” (Takekawa et al. 2010). The transmission of virus across the wild-domestic interface provided opportunities for mutation and reassortment, and researchers hoped to construct a model of these interactions that would enable the structural conditions and contact dynamics leading to viral evolution to be isolated and abstracted. These models, they claimed, could provide sentinel indications of future pandemic dangers.

Building this approach required unusual collaboration among livestock and wildlife veterinarians, ornithologists, bird migration ecologists, geographers, and economists. A set of field research projects were set up around the lake: satellite tracking of bird migrations patterns and habitat use patterns; geo-spatial mapping of rice agriculture; surveys of poultry farmers; sampling of viruses from birds; and assessments of contact rates between wild birds, poultry and humans, among other endeavors. Each research project was considered one part of an “integrated pilot study” designed to answer common questions: “where, when, and how do wild birds interface with poultry and humans?” (Xiao et al. 2010).

But what if their distinction of wild and domestic did not hold?

Scott Newman, an American wildlife veterinarian and migration expert, stared in wonder as swan geese (Anser cygnoides) one by one lurched into the sky, then circled in larger and larger rings around the farmer’s house. It was a damp day in the winter of 2009 and a stiff wind blew across the lake. Newman, working for the Food and Agriculture Organization on avian influenza and other zoonotic diseases, was capturing migratory birds at Poyang Lake. The birds would be surgically outfitted with transponders, then tracked by satellite when they returned north to Siberia in the spring.

But the geese he saw above him were not wild; or at least, they were bred, raised, housed, fed and sold commercially as meat by a farmer, a man surnamed Wang. Newman had found Wang’s farm only that morning through what he declared to be a series of chance discoveries. First, he had been impressed by the vast number of poultry farms around Poyang Lake: Driving “from any point A to any point B,” he told me in a conversation three years after his field visit, he always saw grey poultry sheds, netted ponds, and flocks of ducks or geese in the canals alongside the road. During their research, Newman and his colleagues stayed at the hotel administered by the Poyang Lake Migratory Bird Preserve in the island town of Wucheng. Over dinner, Newman began to ask the hotel staff about the poultry farms: What breeds and species of bird are being raised? Their answer surprised him: They reported species Newman did not typically see raised as domestic poultry, including swan geese.

A domestic duck farm.

A domestic duck farm.

web-Beijing 2011 536-wild ducks

When he visited Wang’s farm, the Wang family graciously invited him for lunch, refusing to be dissuaded from their misrecognition of Newman as an American investor. Showing him the flock of swan geese hundreds strong, as well as mallard ducks, Wang proudly told Newman that bird production could easily be increased, and birds could be exported overseas. Wang also emphasized that the wildness (yexing) of his geese made them particularly valuable.

Newman found himself in a deeply ironic position: Wang’s boasts were an influenza expert’s fears. When he saw Wang’s geese lift off into the sky, he thought to himself that he was looking at “what could be the link between wild and domestic birds.”

“They are the perfect intermediary. Because they look identical to their conspecifics, when they are foraging, a wild bird would come right up to them, because phenotypically they are the same. But then, they go home at night, and there are other poultry around at the farm. So there’s your transmission!”

A poster from the Wang family farm, which reads: “Provide human society with green, ecological, organic, healthful, delicious food … Promote the harmonious development of humanity and nature."

A poster from the Wang family farm, which reads: “Provide human society with green, ecological, organic, healthful, delicious food … Promote the harmonious development of humanity and nature.”

Newman came to believe that these wild bird farms—which he soon discovered throughout the Poyang Lake region—were the key “link” in the wild bird-poultry interface. Along with colleagues, he developed a number of research projects focused on farmed wild birds: farm surveys to count the number of birds, investigations of how birds were marketed, and maps of the foraging range of the birds (Newman et al. 2012).

The integrated study at Poyang Lake began to focus on farmed wild birds as the central site of “disease implication” within the ecosystem. Although Newman and his colleagues moved to confirm their insights in new research projects, the anthropologist must examine more closely how Newman discovered the unexpected. Placing significance onto the farmed wild bird as a reified “link”—to be counted, mapped, and described—displaces attention from the practices that went into farming the wild birds in the first place. To detect the unexpected, the sentinel must cultivate an ability to question scientific objects by examining how these objects are modified by practices.

From this perspective, the distinction of wild and domestic life itself—the conceptual core of the wild bird-poultry interface—can be understood as a product of strategic practices oriented towards the future (Bourdieu 1990). It is through the labor of domestication—practices of capturing, breeding, feeding, and so on—that a qualitative difference is cultivated in the continuum of living beings, a difference of quality which becomes the precondition for categorical symbolic distinction. Moreover, the wild bird farmer approaches this distinction of wild and domestic from a reflexive position (Rabinow 1996; Descola 1996). The symbolic and material dimensions of the distinction become the object of practice. The goal is not domestication, per se, but rather the manipulation of the distinction of wild and domestic to produce new matters, new meanings, and new values.

Farmer Wang’s emphasis on the value of wildness (yexing) in his farmed swan geese makes this symbolic strategy clear. In marketing pitches at their farm, or to visitors at their stall at the Forestry Products Expo, Wang and his son constantly promoted the wildness of their geese. In contemporary China, the consumption of wild or other unusual foods is an important strategy of status differentiation, particularly when banqueting important guests (Yang 1994; Zhan 2005). For Wang, this wildness was not defined ontologically as that which was outside of human touch. Neither was it a stable characteristic of certain individual birds or species of birds. Wildness was a collection of qualities which could be cultivated or lost. The qualities he identified included the taste of the meat, certain secondary sex characteristics, and above all, the ability to fly. After four or five generations in captivity, Wang explained to me at his farm, the birds begin to “regress” and lose their wild character.

Therefore, techniques for cultivating wildness as a collection of qualities constitute a central part of Wang’s farming practices. These include periodic capture of birds from the wild, “exogamous” breeding practices, and maintaining a reserve of birds far from the village farm, where they inhabit a more “wild” environment. On the other hand, a parallel set of techniques aim to domesticate the birds, including imprinting the birds when they are born to consider humans as “parents”, and habituating the geese to regular meals at the farm.

In Wang’s farming, the distinction between wild and domestic is the object of a set of strategies that reflexively manipulate the material of life in order to gain symbolic distinction. When his farmed wild geese are treated as the object of new scientific studies, the strategic art that brought the geese into being may be erased. The danger of this erasure, to put it plainly, is that there is no reason to believe that Wang (or others) will continue to farm the same things; as time moves forward, strategies will shift to maintain symbolic distinction, and change along with it the links in the wild bird-poultry interface. Moreover, scientific alerts of danger, by bringing about new monitoring regimes or biosecurity standards, may well encourage unexpected shifts in farming practice.

If raising an alarm about risk or danger must awaken actors from the routines of scientific expertise and bureaucratic inertia (Chateauraynaud and Torny 1999), the construction of scientific sentinels provides a different configuration. Neither whistleblower nor expert, a sentinel must alternate between scientific conceptualization and attending to how others modify the objects of these concepts through practice. It was only because of the concept of the wild bird-poultry interface that Scott Newman paid attention to farmed wild birds. But it was only because he attended to the practices of others—farmers in the Poyang Lake area—that he discovered the limits of his own concepts.

[1] A poem used as the leader in many Chinese newspaper articles about Poyang Lake and its wild bird preserve.
[2] See Frédéric Keck’s contribution to this issue of LIMN.


Bourdieu, Pierre. 1990. The Logic of Practice. Tr. Richard Nice. Stanford: Standford University Press.

Chateauraynaud, Francis and Didier Torny. 1999. Les sombres précurseurs: une sociologie pragmatique de l’alerte et du risque. Paris: Editions de l’EHESS.

Descola, Phillipe. 1996. “Constructing natures: Symbolic ecology and social practice.” In: Phillipe Descola and Gísli Pálsson. Nature and Society: Anthropological Perspectives. New York: Routledge.

Keck, Frédéric. 2010. “Une sentinelle sanitaire aux frontières du vivant,” Terrain (2010/1)54: 27-41.

Newman, Scott, Boripat Siriaroonrat, and Xiangming Xiao. 2012. “A One-Health approach to understanding dynamics of avian influenza in Poyang Lake, China.” Presentation at EcoHealth 2012, Kunming, China.

Rabinow, Paul. 1996. “Artificiality and enlightenment: From sociobiology to biosociality” In: Essays on the Anthropology of Reason. Princeton: Princeton University Press.

Shortridge, K.F. and C.H. Stuart-Harris, 1982. “An influenza epicentre” The Lancet 320(8302): 812-3.

Takekawa, John, Diane Prosser, Scott Newman, et al. 2010. “Victims and Vectors: Highly Pathogenic Avian Influenza H5N1 and the ecology of wild birds” Avian Biology Research 3(2): 1-23.

Xiao, Xiangming, Scott Newman, Tracey McCracken, and Ding Changqing, 2010. “Wild waterfowl-domestic poultry-human Interfaces: An integrated pilot study in Poyang Lake, Jiangxi, China”, Presentation at The 2nd International Workshop on Community-based Data Synthesis, Analysis and Modeling of Highly Pathogenic Avian Influenza H5N1 in Asia.

Yang, Mayfair. 1994. Gifts, favors and banquets: The Chinese art of social relationships. Cornell: Cornell University Press.

Zhan, Mei. 2005. “Civet cats, fried grasshoppers, and David Beckham’s pajamas: Unruly bodies after SARS.” American Anthropologist, 107(1): 31-42.