Influenza’s Historical Present
I delivered the following speech, co-written with economic geographer Luke Bergmann, at a NIH-FAO-sponsored workshop held in Beijing earlier this month. The speech is based on a book chapter to be published later this year in Influenza and Public Health: Learning from Past Pandemics (EarthScan, London). The text is slightly edited.
This is the first of two talks I’ll be giving. Both I believe attempt to address one of the key concerns of our workshop: how do we work together?
And work together we must. Influenzas operate on multiple levels of biocultural organization: molecularly, pathogenically, and clinically; across multiple wildlife biologies, epizoologies, and epidemiologies; evolutionarily, geographically, agro-ecologically, culturally, and financially.
But it’s more than just a complicated story. The expanse of influenza’s causes and effects play out to the virus’s advantage. As I discussed at last year’s workshop, influenza appears to use opportunities it finds in one domain or scale to help it solve problems it faces in other domains and at other scales.
Collaboration, then, is mission-critical, even as the logistics are difficult. How do we get different research disciplines to talk to each other in such a way as to address influenza’s full dimensionality?
We should remind ourselves, however, that success isn’t merely a matter of concatenating data sets that have long been segregated by disciplinary boundaries. Despite some commonalities, each discipline thinks in its own way. Each discipline imagines problems differently.
It is my view, then, that listening to each other, however important, isn’t enough. It is when we assimilate each other’s professional imaginations–whatever the risks we personally take–that we will begin to cleave influenza’s Gordian Knot. And figuring out influenza is, after all, why we find ourselves in the same room today.
Although trained as an evolutionary biologist I will in this talk address an albeit rough attempt to assimilate another imagination, in this case, that of human geography, and more specifically economic geography.
Let’s start out with an interesting result that requires explanation. Lenny Hogerwerf and her colleagues (in press) separated out the world’s countries by the argo-ecological niches they host. Many of those countries in red and yellow here have supported persistent outbreaks of highly pathogenic H5N1.
In this version of the map, five niches are differentiated on the basis of four agro-ecological variables: agricultural population density, duck density, chicken production, and purchasing power per capita. Here some niches support more ducks than others, etc.
The whys and wherefores one niche supports H5N1 over others we will leave for another day. But one immediately observes that in spite of a few exceptions, the niches themselves are clearly structured by geography, with the most H5N1-vulnerable niches arrayed across South and East Asia, especially along the Chinese lowlands into the river basins of Indochina and, further south, to Indonesia.
Why are countries within each agro-ecological niche for the most part geographically contiguous? I’ll give you a short answer before entering some detail.
First, yes, of course, shared enviro-climatic conditions may contribute to the spatial autocorrelation in niche geography. There is, however, another possibility. Through history, agricultural innovations have emerged locally and undergone bouts of regional geographic diffusion. And prevalent modes of regional agriculture have influenced subsequent developments.
In China, rice cultivation marked the transition between Mesolithic foragers and the surplus food-producing economies of the Neolithic. Domestic ducks were deployed in rice paddies for pest control as early as 500 years ago. And finally, the last layer in our rough cultural sedimentation, Western-style poultry intensification was introduced at scale during the economic liberalization of the past thirty years.
We hypothesize that the duck-rice-intensive poultry niche in Asia resulted from a series of changes in agricultural practice—ancient (rice), late imperial (ducks) and present-day (poultry intensification)—melding in such a way as to uniquely support the evolution and persistence of multiple influenzas.
Now to the details. I will avoid giving a complete history of Chinese agriculture but will hit on some key events and circumstances that I believe can inform our thinking how to move forward our analyses of influenza in Asia.
First, domesticated animals in China have long been integrated, not merely juxtaposed, with other elements in local agro-ecological systems.
The domestication of ducks has been placed at least 3000 years ago, with funerary art dating from the Eastern Han Dynasty depicting agricultural scenes with rice fields, and ducks and fish in ponds.
Rice-duck systems, in which herds of ducks, whether backyard or more nomadic, are allowed to graze on fields after the harvest, have long been in place. By about 500 years ago, in the middle of the Ming Dynasty, ducks were very popular for pest control in the rice paddies of the Pearl River Delta. From various points in the Ming and Qing dynasties, ducks were also promoted for the control of locusts in Fujian and northern China, a mode of control still in practice today.
Fredrick Simoons reviewed a number of foreign accounts that indicate early husbandry bore markings of high-order poultry intensification. One 16th-century account, quoting Simoons,
described a sophisticated system of Chinese duck husbandry, with thousands of ducks kept in cages on boats at night. In the morning the ducks were permitted to leave, entering the water by means of bamboo bridges, feeding in paddies during the daylight hours, and returning when their owners, as evening approached, signaled them to return.
Contemporary practices bear similarities. An international team, including several attendees here today, witnessed several years ago a similar permeability, wherein domestic ducks intermixed with wild birds on their daily commute into Lake Poyang in Jiangxi Province.
Simoons also summarizes a late-19th century account of a proto-commodity chain that began with hatcheries which, again quoting,
sold the young ducklings to duck merchants who raised them in enclosures. When sufficiently grown, such ducks were sold by the merchants to itinerant duck vendors who transported them by water, as many as two thousand to a boat. While he kept the ducks, a vendor permitted them to feed twice a day along the river or in nearby fields, thereby saving the cost of feed…Though the itinerant sold ducks retail in communities along the way, most found provision dealers who specialized in salting and drying them.
By early-20th century agricultural surveys, ducks and chickens were found in much greater densities in rice-growing regions, especially areas of double-cropping. Densities found nowhere else in the world at that time.
A second observation that should inform our thinking: Integrated farming practices in China, layering and interweaving multiple types of farming, were long diverse in their character. Not just duck-rice.
For instance, the origins of rice-fish farming extend back at least 1700 years and records from over 2000 years ago describe other aquatic plant-fish systems. Whereas various livestock-crop systems are widespread and relatively well-known, livestock-carp systems are a somewhat more unique contribution less widely understood today but dating in China at least back to the Ming Dynasty.
By about 400 years ago, fruit tree-fish and mulberry-silkworm-fish dike-pond integrated systems are documented. Into these domestic ducks were integrated in the 1860s. The mulberry-silk-fish system is of interest in and of itself because it speaks to two additional observations:
Third, the dynamics of integrated farming are locale-specific. Case in point, the southern Chinese province of Guangdong, a contemporary epicenter for multiple influenzas.
We will focus on those processes that have brought humans, livestock and wild birds together in the Pearl River Delta, shown here, at the core of the province, around which contemporary centers of industrial production and population such as Guangzhou, Shenzhen, and Hong Kong were built. Dynamics elsewhere in the region are critical, of course, but we must start somewhere.
Much of the Delta itself emerged over the past two thousand years, some causes of which were, and are, anthropogenic, from conscious acts of reclamation to increased siltation from deforestation upstream. Beginning perhaps in the Song Dynasty about a thousand years ago, Delta wetlands were increasingly converted to ponds divided by soil piled onto dikes, forming the first iterations of what would be known as the dike-pond system. Fish were raised in the ponds, then fruit trees and various crops were planted on the banks, with chickens and ducks potentially integrated.
By the latter half of the 16th century, in middle of the Ming, however, instead of fruit trees, mulberries were increasingly planted on the banks in order to feed silkworms, helping close a rather efficient nutrient cycle between banks and ponds. By 1581 the mulberry dike-pond system occupied about 30 percent of certain key counties in the Delta. By the early 20th century almost all the land in a number of parts of the larger area had been converted to this silk-producing system.
The changes to the Pearl River Delta were tied not only to their locales but to their global context too. In other words—a fourth observation—changes in the kinds of integrated farming have long been related to the state of global and regional economies. Guangdong, for instance, was a key point for foreign trade and a long-distance international market for silk explicitly drove the development of local land-use and the rise of the dike-pond system.
The effects of globalization have almost always run in multiple directions. Whatever the socioecological virtues of dike-pond ecosystems, in no sense were they simply ‘sustainable’ systems of locally-closed loops. The mulberry-dike system was functionally open and at the center of many flows, sustained by products exported internationally and by substantial food and potentially other inputs imported interregionally.
By the middle of the 18th century, the Delta and a vast hinterland stretching into neighboring provinces had been functionally integrated within a single differentiated agro-ecosystem. The whole landscape of the region would then also have been more directly coupled to the dynamics of an emerging global political economy, influenced whether by the expansion of world silk markets or by the emerging crises of capitalism in distant lands.
Within this larger interconnected region, but further away from major population centers and their resources than the dike-pond regions, other agro-ecosystems were prevalent. For example, the zheng gao system is a variant of paddy-rice intercropping requiring less labor and resources to expand output in the Delta peripheral.
Extensive use was made of ducks as well, to eat crab pests in the paddies and to recycle grain waste in the fields after the harvest. The system grew alongside the dike-pond systems from the Ming to the Qing until state initiative in water-management projects to raise yields eliminated zheng gao in the early years of the socialist period in the 1950s.
The Revolution helps illustrate another observation. The state of the agro-ecology can be reset by broad shifts in the society at large.
The Maoist period brought a number of agrarian reforms, including changes in cropping systems, in land-tenure, in labor conditions, in social relations, and in extra-regional economic linkages. Development was, in theory, aimed at decreasing polarization and dependence on the export markets of the global core, and toward spatial and social parity, focusing on a national economic space. The Pearl River Delta’s previous international trade linkages and associated flows of commodities and money were thereby radically reshaped.
A doubling in population required a focus on developing grain production in order to meet the basic caloric needs of the people, a task which met with several setbacks and which was only secured in the 1980s after the maturation of the fertilizer industries in which heavy investments were made in the 1970s.
During this period integrated farming wavered in the Pearl River Delta. The silk-mulberry economy had already collapsed during the Great Depression and the zheng gao cropping system was now replaced, as we alluded to earlier. Rice-fish farming was promoted early in the socialist period, with 700,000 hectares nationally by the late 1950s, but disruptions, whether political, policy, or pesticide in origin, resulted in a sharp decline during the 1960s and 1970s. In Guangdong, rice-fish acreage declined from around 40,000 hectares to a mere 320 by the beginning of the Cultural Revolution in 1966.
By the middle and late 1970s, interest in ecologically-integrated farming arose again, with communes establishing farms and conducting research on optimizing combinations of rice, silkworms, chickens, ducks, fish, and pigs. At the same time, other communes were leading the way, in something of another direction, in researching and implementing local versions of industrial livestock intensification.
Which brings us to the recent era. Economic liberalization changed China’s agro-ecological landscape yet again.
In the late 1970s, China began to move away from a Cultural Revolution policy of self-sufficiency, in which regions were expected to produce most foods and goods for their own uses. In its place the central government began an experiment centered about a reengagement with international trade in Special Economic Zones set up in parts of Guangdong, Fujian, and later the whole of what would become Hainan Province. In 1984, fourteen coastal cities were opened up as well, though not to the extent of the economic zones.
Starting in 1979, annual foreign direct investment (FDI) increased from zero to 45 billion US dollars by the late 1990s, with China the second greatest recipient after the US. Sixty percent of the FDI was directed to manufacturing. Given the extent of China’s smallholder farming, little FDI was initially directed to agriculture.
That soon changed. Through the 1990s poultry production grew at a remarkable 7% per year. Production for domestic consumption and investments were not confined to chickens, given the longstanding interest in the consumption of duck and goose. Yet processed poultry exports grew from 6 million US dollars in 1992 to 774 million US dollars by 1996.
The changes were more than merely emergent. They were structured by new legislation and diplomatic efforts. China’s Interim Provisions on Guiding Foreign Investment Direction aim to encourage FDI across a greater expanse of the country and in specific industries, agriculture included. The government’s 2005 five-year plan set sights on modernizing agriculture nationwide. Since China joined the World Trade Organization in 2002, with greater obligations to liberalize trade and investment, agricultural FDI has doubled.
By the late 1990s Hong Kong and Taiwan’s contribution to China’s FDI had declined to 50% of the total, marking an influx of new European, Japanese, and American investment.
Economic liberalization, particularly its changes in ownership structure and its geographic integration within and beyond southern China, has had a fundamental effect on regional husbandry. By 1997, and the first H5N1 outbreak in Hong Kong, Guangdong, home then to 700 million chickens, was one of China’s top three provinces in poultry production. Some of Guangdong’s poultry operations were by this point technically modernized for breeding, raising, slaughtering, and processing birds, and vertically integrated with feed mills and processing plants.
Foreign direct investment helped import grandparent genetic stock, support domestic breeding, and update feed milling. The majority of breeds used in industrial production were now imported, bred for profit and high rates of capital turnover. At times, production has been somewhat constrained by access to interprovincial grain and the domestic market’s preference for native poultry breeds less efficient at converting feed. Of obvious relevance, production also suffered from less-than-adequate animal health practices.
Today, expansion is robust and so-called ‘high quality’ chickens, longstanding domestic breeds or hybrids, are increasingly being produced, despite their higher costs and longer turnover times. Guangdong is producing approximately one billion of these broilers a year.
The province’s economic ascension was not without its detractors, however, a dynamic with potential consequences for influenza.
Among them, landlocked provinces chafed at the liberalization the central government initially established in coastal provinces alone. With so much domestic currency on hand, the coastal provinces could outcompete inland provinces for livestock and grain produced by the inland’s own town and village enterprises.
The coastal provinces were able to cycle their competitive advantage by turning the inland’s cheap grain into more profitable poultry or flat-out re-exporting the inland’s goods, accumulating still greater financial reserves. At one point rivalries became so intense that Hunan and Guangxi, bordering more prosperous coastal provinces, imposed trade barriers upon interprovincial trade.
The central government’s efforts to negotiate interprovincial rivalries included spreading liberalization inland. Provinces other than Guangdong and Fujian were also becoming entrained into market agriculture, albeit, in something of a reprise of pre-Revolution dynamics, at a magnitude still outpaced in certain sectors by their coastal counterparts.
Industrial poultry’s expanding extent—by re-exporting and inland development—increases the geographic scope for the emergence of market-oriented influenzas and may explain in part, as shown by the phylogenetic evidence, the roles Yunnan and Hunan appear to have played in serving up H5N1 abroad.
Despite these developments, diversity remains the order of the day. Numerous forms of ownership, organization, and production co-exist. Foreign investment and intensive production have not eliminated all backyard producers. Equally, not all small-scale producers are operating independently.
Instead, thick webs of contractual obligations interweave a diverse ecology of economic actors.
For instance, Guangdong Wen Foodstuff Group, the largest chicken producer in the province as of early in this decade, drew revenues of 1.6 billion renminbi in 2000, employed about 4400 employees in the central company and 12,000 household contract farmers, and maintained a close relationship with South China Agricultural University. As of the 1990s, provincial feed mills were operated publicly, by village cooperatives, by joint-ventures, and by private capital all at the same time.
Such developments in the poultry industry, many of them in the greater region around the Pearl River Delta, are not occurring in isolation, but in the midst of a period of extremely rapid urbanization, suburbanization, inward migration, industrial expansion, interregional integration and economic differentiation, and export-led growth.
The greater delta’s agro-ecological landscape mosaic was built, and is being built, on the historical dynamical patterns we described earlier. At the same time, the development represents a historically unprecedented density and juxtaposition of activities, with potentially fundamental consequences for influenza evolution.
In other words, influenza’s regional epizoology arises out of a complex interplay of agro-ecologies past and present, in what the philosopher Louis Althusser called the ‘historical present’.
That mix of past and present is arrayed across the region’s geography. We see here the first of two maps from studies of land-use in the Pearl River Delta (Weng 2002), approximately ten and twenty years after the start of economic liberalization, respectively. In 1989 huge swaths of land are still cropland, cities relatively compact (Guangzhou, to the upper left, is spatially the most prominent), and the dike-pond land relatively isolated from concentrated populations.
By contrast, in 1997 urbanization is vastly greater, spreading out not only from a number of centers but also creating a number of suburban filaments stretching through the countryside.
Studies have indicated that much land-use change in the Pearl River Delta is fragmenting in nature and is directly related to foreign direct investment. The question remains, however, as to the exact relationships livestock industry investments may have with such patterns. Landscape ecological metrics also suggest significant increases in developed land ‘edge density’ over this time period, on the fringes of the major cities in the Pearl River Delta. Cropland in the later time period is almost absent, having been replaced by horticulture, development, and ponds.
The story of the growth of aquaculture is also apparent in the maps. As aquaculture’s economic returns were perhaps two to three times those to cropland in the time between these two maps , it is not unexpected that much land would be converted, often by village cooperatives.
Much of what is described as ‘dike-pond’ land in these maps may actually reflect a greater emphasis on ‘pond’ than previously, as the land is converted into aquaculture. However, some observers still suggest that aquacultural ponds are also likely to be used for the production of waterfowl. As ponds have spread out from the traditional core of the dike-pond region in the west section of the area, the amount of built-up land spread across the pond regions appears to have increased greatly, potentially bringing increasing human population densities and aquatic habitat into proximity.
So new layers of export-driven landscape are being superimposed upon those of previous eras. However, as the Delta’s export-driven economy develops, diversifies industrially, moves toward more costly technologies, and further urbanizes, there is pressure to move livestock industries farther away from valuable urban land. Other large producers have long been located in smaller urbanized areas in the peripheries of these larger interurban systems.
There appear to be certain similarities between the resulting landscapes and peri-urban regions of other newly-industrializing states across Southeast Asia called alternately desakotas (“city villages”) or Zwishenstadt (“in-between cities”). Many may very well share the same agro-ecological combinations described in the niche modeling of Hogerwerf and her colleagues with which we began.
The sum effect for the Pearl Delta, and further afield across southern China, may include the possibility that poultry intensification and the pressures placed on agro-ecological wetlands have squeezed a diversifying array of influenza serotypes circulating year-round through something of a virulence ratchet.
The resulting viral crop—for 1997, H5N1 by molecular happenstance—may have the opportunity to be subsequently exported out by international trade facilitated by Hong Kong and diasporic capital.
Additional research for such an assertion is required. But we may have found a peg on which to place such work. Namely, history and context matter, as much for pathogens as the humans they infect.
Indeed, at the risk of reifying arbitrary blocks of time, pathogens have their own origins, diasporic migrations, classical eras, Dark Ages, and Industrial Revolutions. As human pathogens evolve and spread in a world of our own making, these analogous eras are often coupled with our own.
What we find in southern China today for influenza is neither effortlessly remade independent of history nor enslaved to a static past. The region has neither been unconnected from the rest of the world nor had its specificities erased by a wave of recent generic globalization. The socioecological environment in which influenzas are evolving there is the complex and layered product of past and present, of global and local. The causes of emerging influenzas in southern China today are threads which may bind many places, peoples, and times together, though never evenly, and in a place-specific way.
In other words, southern China’s role as a primary influenza epicenter is far from inherent, instead arising from a contingent confluence of local and global factors in what the geographer David Harvey would call an ‘active moment’ in spatial configuration. The mechanics of that dynamic configuration as it relates to influenza remains largely a mystery. For the landscape itself:
What are the exact locations and practices of intensive livestock operations across the region? What is their proximity to smallholders practicing duck-fish aquaculture? Is the effect of present-day industrial siting of significance? Are there synergies in the landscape today that were not present in the less spatially heterogeneous rural systems still prevalent thirty years ago?
Efforts at putting numbers on the relations between the historical environments and the evolution of influenzas are correspondingly more challenging—of another order of complexity altogether:
Are these mixed landscapes what produce the aggregate averages of the most epidemiologically vulnerable agro-ecological niches Hogerwerf et al. introduced? Are some molecular phenotypes repeatedly selected for by specific micro-niches? Is influenza’s repeated parallel molecular evolution a marker of the virus’s ability to evolve in response to the landscape features defined by both past and present together?
In other words, getting back to how we might work together, when we whittle analyses of Southern China’s economic geography against its agro-ecological history we can ask new questions about influenza’s evolution. And asking questions no one has yet bothered with is half the battle.