Valuing Organics: Labels, People, and the Materiality of Information Infrastructure in China

ABSTRACT

Certification schemes and alternative food networks have often been studied separately or using different analytical concepts. Ethnographic research in Sichuan Province, China shows how a cooperative used both labels and people to inform consumers about the ecological quality and trustworthiness of its food, thereby obtaining price premiums. Using information infrastructure as the common analytical vocabulary for exploring both certification and performances of face-to-face relations unearths unexpected commonalities amidst presumed otherness. The infrastructure lens draws attention to the importance of scientific planning and materiality: the force of specific features of different crops and modes of production. Information infrastructures – whether labels or people – depend on, and affect, the materiality of whatever they are supposed to be informing about. They do not simply convey information, but shape value as they interact with and intervene in what is produced.

KEYWORDS:

• Alternative food networks
• certification
• infrastructure
• standards
• value

In 2010, a peasant cooperative was established in Daxi, a self-styled ‘ecological village’ of about 1,500 registered inhabitants in the hills near Chengdu, the capital of Sichuan Province in China.¹ When I conducted ethnographic fieldwork there between 2013 and 2015, the cooperative was still struggling to create a market for its agri-food products. Following the cooperative’s rice and vegetables from fields to shelf, I explored the different means the cooperative used to inform consumers about products’ features and production.

In the winter of 2014 to 2015, I accompanied Wang Zhaochen (the village party secretary and director of the cooperative), Kang Sunbin (the village leader and its vice-director), and Wen Erqiang (a former school mate of Wang and an entrepreneur who had invested in a rice mill which the cooperative could not afford on its own) to three trade fairs in the metropolis of Chengdu (eight million inhabitants) and the smaller cities Suining and Ziyang (about 500,000 and 300,000 inhabitants, respectively). There, the cooperative displayed its rice packaged in plastic bags with the ‘green food’ (lüse shipin) label. While we placed these newer bags in the most prominent positions on the provided tables and showcases, we also used older bags without the label to avoid wastage and reduce costs. The sales and networking opportunities at the fairs failed to meet expectations. Very few commercial buyers stopped at the booth and most consumers either thought the rice was too expensive or doubted whether it was really ecological.

The cooperative also sold rice to middle-class consumers in Chengdu through associations they had organised. In May 2015, I helped Wang Zhaochen collect products and haul them to a consumer association’s shop. Kang Sunbin and two members of the cooperative’s board had collected two hundred kilograms of paddy rice from villagers and husked it at the rice mill. Deliveries to consumer associations were less labour intensive to prepare than displays for the trade fairs. There was no need to fill small bags with an exact amount of rice (which required repeated weighing and adjustments) and vacuum-seal them. Instead, the cooperative used woven polypropylene sacks that held twenty to thirty kilograms. These plain white bags bore neither the logo of the cooperative nor the eco-label: Kang only wrote the variety and weight of the rice on them in large red characters. As the rice was displayed in bulk, neither the members nor any walk-in customers at the consumer association’s shop could see the label that testified to organic certification.²

Dong Jie, the township agronomist who had initiated the establishment of Daxi’s peasant cooperative, considered certification a waste of money and even feared that environmentalist consumers might interpret it as putting profit before ecological concerns. When he talked with organised urban middle-class consumers, he tried to justify the label by claiming that state subsidies had induced the cooperative to acquire certification. And indeed, the consumer association did not display the existing label, even though its customers still paid a price several times higher than that charged for non-certified rice in supermarkets.

What the association’s shop did display were posters that briefly introduced their food producers with photographs and descriptive text. The Daxi cooperative’s photo showed several people standing on a path between paddy fields with Wang Zhaochen at front and with their arms and hands posed gracefully as if welcoming the viewer to join them. The producers were thus given faces, implying that they – and by extension their products and the production process – were personally known at the consumer association. I interpret this poster as a visual representation of the idea that people – and direct links between them – offer an ‘alternative’ to certification schemes.

Food safety risks and consumer anxieties in China have received a fair share of academic attention (Yan 2012). Despite (or even because of) this known distrust, markets for food that consumers come to trust as ‘ecological’ have been growing. Some authors have differentiated between a ‘“formal” organic sector, created by “top-down” state-developed standards and regulations, and an “informal” organic sector, created by “bottom-up” grassroots struggles’ (Scott et al. 2018: i; see also Merrifield 2020). In contrast, this article strives for more analytical symmetry when examining what is often portrayed in studies of food networks as a binary of formal standards versus personal relations. Drawing on theoretical vocabulary developed in infrastructure studies, it ethnographically analyses how an agricultural cooperative in Sichuan Province informed consumers about its ‘ecological’ products. It thereby not only intervenes in the debate about conventional and alternative food networks in China and elsewhere but also contributes to understanding how the materiality of information shapes value. Designed and performed as information infrastructure, neither labels nor people simply convey information: they intervene in what is produced and valued.

Theoretical Puppetry and Analytical Symmetry in Agri-Food Studies

Critics of standardisation have sometimes indicated possible alternatives. In the conclusion to her ground-breaking monograph on certified organic agriculture in California, Julie Guthman (2004: 181–185) warns against the fetishisation of alternatives while simultaneously presenting community-supported agriculture (CSA), direct marketing through subscription farms. and farmers’ markets as ‘possible starting point[s] to a fairer and more equitable food order’. Ethnographic studies of such ‘local food systems’, ‘short supply chains’, and ‘alternative provisioning networks’ have often invoked concepts such as personal relationships, embeddedness, informal social networks, solidarity, social capital, or community (Hinrichs 2000; Sonnino & Marsden 2006; Goodman et al. 2012; Grasseni 2013; Aistara 2018; Scott et al. 2018). While often including careful reflections on differences between alternatives, tensions between decommodification and marketisation, and the dangers of conventionalisation, studies of alternative food networks thus rely on some of the very same concepts that also inform activists who initiate and run these food networks. This prevents some surprising insights.

Stefan Hirschauer (2008: 177) has discussed ‘theoretical ventriloquism’ (theoretische Bauchrednerei) as a serious methodological limitation on interview-based research. If experts are well-read in social science literature, he argued, they provide a domesticated form of empirical data that can all too easily be assimilated to social scientists’ theoretical outlook without creating any potential for theoretical innovation. But a similar issue can also emerge in ethnographic research based on participant observation, particularly when social science concepts have previously been translated into activist models and organisational practice, which thus confronts researchers with performances of what I call ‘theoretical puppetry’.

Analysing ostensibly objective, standards-based food networks through the lens of personal relations has been a first step towards a more productive analytical symmetry. In her comparative study of postcolonial transnational commodity networks, for example, Susanne Freidberg (2004: 208), found that supermarket supply chain management practices, ‘while technocratic and standardized on paper, were in fact highly subjective’.³

This article makes another step towards symmetry, starting from the other side and applying analytics developed in studies of certification schemes to their supposed alternatives. This move has been prefigured but not fully executed by economic sociologist Lucien Karpik (2010: 181). While he has categorised both labels and people as ‘judgment devices’ for ‘singularities markets’, he has also reproduced the opposition by systematically contrasting ‘network-based personal devices’ (highlighting flexibility, for example) to ‘impersonal devices’ such as labels.

Labels – such as the green food label for the cooperative’s rice – have been fruitfully studied as ‘information infrastructure’ (Frohlich 2017). This concept refers to techno-organisational networks that facilitate the production, distribution, and consumption of knowledge (Bowker et al. 2010). Both the physical object applied to the packaged food and the formalities and technicalities of certifying compliance with specified organic standards – through processes such as documentation, auditing, and testing – constitute the label as an information infrastructure. Aspects such as formality, technicality and standardisation make it counterintuitive – yet all the more productive – to apply the analytical notion of infrastructure to initiatives and movements that describe and perform themselves as networks of personal relations.

In infrastructure studies, some authors have made the provocative suggestion that people be considered as infrastructure (Simone 2004; Elyachar 2010; see also Sippel, this issue) and have been accused of thereby ignoring ‘classical’ features of infrastructures (Calkins & Rottenburg 2017). Inspired by these interventions, this article ethnographically explores people as value-adding information infrastructure, particularly in terms of scientific planning and materiality. This contributes to debates in infrastructure studies, agri-food studies and economic anthropology, demonstrating how information infrastructures shape the value of agri-food not simply by conveying information, but also by interacting with various materialities.

The first section discusses the cooperative’s green food label for rice through social-scientific literatures on standards and infrastructures, highlighting interrelations between information and materiality. The second section moves from the cooperative’s rice to its vegetables and from labels to people as information infrastructure. Activist models, informed by social science concepts, designed people as a substitute for quality-guaranteeing and value-adding labels. Diverse materialities of infrastructures, food crops and modes of production interacted to enable, shape and limit how these models were translated during consumers’ field visits. Therefore, the article concludes that neither labels nor people are neutral tools for conveying knowledge. Instead, both information infrastructures are entangled with background work and various materialities. As they intervene in what is produced – here, loss-making vegetables along with profitable rice – information infrastructures generate and destroy value.

Labels as Information Infrastructure

In 2013, Wang Zhaochen (the director of the cooperative) and Wen Erqiang (the rice mill entrepreneur) had successfully applied for the ‘green food’ label and by the time of the trade fairs were already in the process of acquiring ‘organic’ certification for the cooperative’s ‘ecological’ (shengtai) rice. To friends and acquaintances (other entrepreneurs and officials), Wen Erqiang often proudly emphasised that this ‘state certification’ (guojia renzheng) – along with all the other necessary documents and permits – would enable them to sell their ecological rice to customers throughout China. Then, he often also highlighted the high price of their ecological rice as an achievement in a way similar to the lectures he liked to give me about the high prices of certain brands of cigarettes and alcohol that he, his acquaintances, or famous people consumed or gave and received as gifts. In such conversations, he followed ideas from management literature on supply chain management that envisioned eco-labelling as an informational, and thus value-adding, tool.

Social-scientific studies have demonstrated that food labels do not simply function as neutral ‘knowledge fix’ and compensate for information asymmetries between knowledgeable producers and ignorant consumers. Touching upon the more general question of how scientific knowledge is popularised, science and technology studies (STS) have examined how food label information is interpreted by different groups and translated into knowledge about how to use these labels (Eden 2011; Frohlich 2017: 147–148). Beyond this focus on the interpretation of information, some anthropological, sociological and STS research on food labels has turned attention to the materialities of information, particularly through the productive analytical lens of infrastructure. Labels, associated standards, and processes of certification have been explored both in terms of their relation to infrastructures (in historical materialist as well as new materialist terms) and as infrastructure themselves.

Designing and Adjusting Standards to Infrastructural Materialities

Anthropological studies have shown that safety, hygiene and food standards in the EU have not been designed de novo but were based on existing infrastructures. This ‘politics of materiality’ (Gille 2016) reproduced inequality by favouring those producers whose material infrastructures already complied with the standard, mainly those in the old and more powerful Western member states (Dunn 2003; Aistara 2018; see also Ana, this issue). Similarly, the translation of models of eco-labelling from the West to China reveals that related standards and certification procedures have been designed based on, and need to be adapted to, existing political economies and infrastructures.

The first organic-certified product in China was green tea in 1990, but this was intended for export and certified by a European agency. Since then, the Chinese state itself has set up three standards of ecological certification for food: ‘hazard-free’ (wugonghai, literally ‘no public harm’, sometimes also translated as ‘pollution-free’), ‘green’ (lüse), and ‘organic’ (youji). ‘Hazard-free’ agri-food requires the environment of the production area to be tested for heavy metal residues and the products to be tested for pesticide residues but allowed a wide range of agrochemicals in the production process. ‘Green’ food allows only certain agrochemicals and regulates the amounts used and level of residues in the products. ‘Organic’ food allows neither synthetic fertilisers and pesticides nor genetically modified organisms. These eco-labelling schemes were developed after a global market for certified organic products had already been established (Thiers 2002) and inherited its dominant emphasis on technical definitions that prescribe certain inputs and prohibit others, a point to which I will return below.

Remarkably, there has been no linear move towards increasingly stringent technical standards within China. The ‘green’ food standard was introduced in 1990 and aimed mainly at the domestic market; the (higher) ‘organic’ standard was modelled on international standards in 1994; and the (lower) ‘hazard-free’ standard was introduced only in 2001 to address food safety concerns in China because the ‘green’ food standard turned out to be unachievable for most producers (Scott et al. 2014). In formulating three different standards for eco-food, the Chinese government made a plausible attempt to enable some producers to tap into profitable export markets while also protecting others from foreign competition by creating domestic alternatives to international standards that take specific materialities of agricultural production into account.

In post-Mao China, unlike North America and Europe, there was no eco-movement aiming at protection through eco-labelling. Instead, what Paul Thiers (2002: 368) calls the ‘fragmented entrepreneurial state’ has translated this market-based approach to promoting sustainable agriculture from the West for a very different political economy. Since post-Mao agricultural reforms, collective ownership rights to land have been combined with use rights to land for households. State entities at different levels and in different departments have acted both as regulators and as competitors pursuing profit. At the central level, the Ministry of Agriculture and the State Environmental Protection Agency established competing affiliated certification agencies for, respectively, ‘green’ and ‘organic’ food. At the local level, county, township and village governments instituted agricultural development plans. They used their political authority to make households comply with regional specialisation in certain value-added agri-foods to achieve economies of scale based on specialised processing, storage, transport, and marketing infrastructure (Thiers 2002; on the capitalist afterlife of socialist infrastructures of wine and raspberry production, see Ana and Thiemann, this issue).

Individual farms were not certified but rather so-called ‘production bases’ covering the land farmed by several rural households (Thiers 2002): small plots and shared rice-paddy irrigation infrastructure made eco-labelling for individual peasant households impossible.⁴ According to the township agronomist Dong Jie, households from three out of Daxi’s nine villagers’ groups participated in the first year of his training sessions and experimented with cultivating paddy rice without synthetic fertiliser or agrochemicals. In the following years, more groups joined the effort until ‘the whole village’ had transitioned to cultivating ‘ecological rice’ and joined the cooperative’s certified production base (on agronomists’ pivotal role in ‘infrastructuring’ value, see Thiemann, this issue).

Based on the Western experience with organic certification and the related economics of premiums, it may seem counterintuitive for (very) small-scale producers to transition to organic agriculture through the cultivation of grains like rice rather than producing crops from which more value per acre can be extracted (Guthman 2004: 175). In Daxi, Dong Jie argued that focusing the ecological transition on a staple like rice would get the villagers themselves to eat healthy food. But rice also made sense for a new cooperative in China that was struggling to create a market for its agri-food through eco-labelling.

The very small amount of agricultural land available per household in the hilly village meant that most of the rice produced was reserved for household consumption over the course of the year.⁵ After each harvest, any rice left over from the previous one was usually fed to chickens and pigs. The cooperative only bought villagers’ unhusked rice when there was consumer demand and villagers only sold it when they did not need it for themselves. While peasant members of the cooperative paid no membership fees, held no shares and received no dividends, the cooperative offered them a premium over the local market price for unhusked rice. Thus, neither the villagers nor the cooperative ran much risk of investing either too much labour in producing or too much money in buying an agricultural product that they would not be able to sell at an ecological premium. In Daxi Village, eco-labelling has thus been made compatible with existing materialities of specific infrastructures and modes of production by organising a ‘specialised peasant cooperative’ (in line with the relevant 2006 state law) and by selecting a specific crop.

Beyond these material links between standards (such as eco-labels) and infrastructures (such as irrigation), some scholars have suggested studying labels themselves as ‘information infrastructure’ (see also Lammer and Thiemann, this issue).

Infrastructuring Information and Producing Food with Information

Susan L. Star (1999), Geoffrey C. Bowker and their colleagues (2010) have extended conventional understandings of infrastructures to ‘the technologies and organizations which enable knowledge work’ (98), including ‘more abstract entities, such as protocols (human and computer), standards, and memory’ (97). While these scholars of science and technology have employed the term ‘information infrastructure’ mainly to analyse ‘digital facilities and services usually associated with the internet’ (98), Xaq Frohlich (2017) applied the analytics of infrastructure to other kinds of information technology.

Studying labels as infrastructure – instead of merely studying interpretations of labels – has five analytical advantages, Frohlich states: Infrastructure shifts attention to (1) ‘the problem of scale’ and ‘action at a distance’, (2) ‘realignments of the public and private spheres’, (3) ‘the role of material assemblages in binding social relationships’, (4) ‘the effects or problems raised by building on top of previous infrastructures or environments’, and (5) ‘taken-for-grantedness’ as the ‘key feature of infrastructure’s power’ (Frohlich 2017: 163–164). Beyond these insights that the analytics of infrastructure generally affords, he highlights three more specific features of information infrastructure. It draws attention to (1) ‘the material work that goes into immaterial information systems’, (2) in particular, ‘the cost of data entry work’ as a ‘common constraint on any information system’, as well as (3) ‘a tendency towards convergence between the medium and the message’ (Frohlich 2017: 164). It is this attention to the links between information and materiality, as well as to the – often invisible – work of creating, maintaining, and changing these links, that proves so powerful for studying eco-labelling (and below: people) as information infrastructure.

Besides his focus on how materialities of technology and organisation shape labels as information infrastructure, Frohlich (2017) also spotlights the material consequences of labelling. He argues that labels have the ‘tendency to be performative rather than merely descriptive’, given their ‘power to frame markets by what is measured and counted, discounting externalities that are not [measured]’ (147–149). Karpik (2010) has already argued that ‘judgment devices’ such as labels construct consumer choice and interactions; to this focus on consumption, Frohlich (2017: 156) adds that nutrition labelling also drives the production of ‘foods with desirable information’ rather than desirable food (149).

The power of information infrastructure shaping what is produced has also been observed in studies of eco-labelling. Labels are not only interpreted by different communities but as mediums also interact with the materiality of certain crops and specific agricultural inputs. In her California case, Guthman (2004: 173) observes that the technical definition of organic standards (allowing some inputs and disallowing others) meant that ‘the ability to produce organically has become less dependent on how a grower manages production than on what crops he or she grows.’ For some crops, such as stone fruits, walnuts, or grapes, certain inputs for pest and weed control were permitted as ‘organic’ and became available early on. It took more time (or has not yet been possible) to find such easily usable ‘tools’ for other crops, such as artichokes, strawberries, or avocados (150–151). Organic standards ‘designed [by the organic movement] as barriers to entry, have had the paradoxical effect of attracting entry into the sector by playing into a logic of valorization but, in turn, undermining the self-protection many growers originally sought’ (173).

In Daxi, the peasant cooperative was able to produce ‘green’ and ‘organic’ certified produce because alternatives to measures like chemical pest control were readily available. Certain kinds of alternative pest control were classified as ‘ecological’ by some actors but not by others, and were more-or-less labour intensive. In other villages I have visited in China since 2011, activists and organisers of cooperatives have introduced me to selected peasants who enthusiastically claimed that they went to their paddy fields daily to catch pests such as rice skipper butterflies by hand. This was not the case in Daxi, where the cooperative had installed electric insect traps throughout the village, a technology that was both labour-saving for peasants and in line with the ‘green’ and ‘organic’ food standards.

However, Dong Jie, the township agronomist, had been inspired by Masanobu Fukuoka’s ‘do nothing’ method of ‘natural farming’ and did not regard this technological intervention that attempted to ‘control nature’ as ecological. And here is where a critical story about food labelling – well-rehearsed both in academic and activist circles – might be cut short. By reducing complex practical knowledge (mētis) to information, standards and labels allow corporate producers interested in price premiums to gain access to niche markets, thereby marginalising those who pursue more holistic approaches.

Spotlighting existing infrastructure in the self-styled ecological village offers an opportunity to give the story a twist. Different actors involved with Daxi’s peasant cooperative strove for divergent versions of ‘ecological’ farming and preferred differing means to make consumers aware of them. If any intervention was required, Dong Jie would have preferred to substitute biogas slurry for pesticides, as he repeatedly explained to organised consumers and other visitors to the cooperative. But only some peasants used this more labour-intensive method of pest control and only occasionally; many others found it too cumbersome. More importantly, the application of biogas slurry built on an infrastructure that had been installed over the past decades as Daxi became a model village for biogas digesters. These had been constructed in every household, with state support (on the use of old infrastructural modules for new agri-food products, see Thiemann, this issue). Even if the peasants in Daxi had whole-heartedly embraced the biogas-slurry method and followed the township agronomist’s notion of ecology, production according to this unwritten standard would not have been easily replicable in other villages without the necessary material base. Like other standards (Dunn 2003; Gille 2016), it was designed based on, and suitable for, a specific set of existing infrastructures.

Moreover, as I will argue in the remainder of this article, Dong Jie’s preferred way of communicating his agricultural approach to consumers – through personal experiences and relatedness – also still involved scientific and political planning, was built on an installed base, and affected what was produced. This is surprisingly similar to the role labels play as information infrastructure.

People as Information Infrastructure

While one did see some products with eco-labels on the shelves, the consumer association in Chengdu mainly presented itself as knowledgeable about the agri-food products it offered through personal relations with trusted producers. In an interview, Wu Yingying, the woman who founded the consumer association, explained that distrust of the state’s certification schemes was one reason for setting up the organisation and establishing direct relations with peasants and other producers of ecological products. According to her, this would also avoid rigid standards that discouraged producers from transitioning to ecological agriculture. People should be substituted for labels.

Africanist anthropologists have argued that people may replace failing or lacking infrastructure. AbdouMaliq Simone (2004) suggested looking at ‘people as infrastructure’ who substitute for the lack of urban infrastructure in Johannesburg, South Africa. Similarly, Julia Elyachar (2010) also framed marginal people – women in Cairo and their practices of sociality – as ‘social infrastructure’. Against persistent discourses that continue to portray Africa as dysfunctional and Africans as passive victims, this highlights creative and effective practices of infrastructuring. Yet Sandra Calkins (2021: 708) warns that this ‘romance of improvisation … backgrounds the darker side, the humiliation and the dangers that can also be lurking when people step into infrastructural gaps’. In a postcolonial Ugandan molecular biology lab embedded in global science, ‘scientists insert themselves [and their bodies] into infrastructural gaps and forge molecular ties with toxic chemicals.’ (723)

People as infrastructure have thus become a new ‘located topic’ (Lederman 2008: 311), one located in Africa and fulfilling a critical function against both othering and postcolonial violence. Recently, this located topic has travelled to other places to serve a similar function. Studies of street food in Indonesia and Mexico have used it against negative public images in order to acknowledge the agency of street vendors and recognise the productivity and necessity of their reproductive labour (Malasan 2019; Hayden 2021). Introduced to the anthropology of China, this concept encounters quite different public discourses about infrastructure, especially since the Chinese state has launched its immense Belt and Road Initiative in 2013 (Joniak-Lüthi 2020). When applied to alternative food networks, its positive emphasis on flexibility and improvisation also risks reproducing actors’ self-representations. Only a transformed notion of people as infrastructure can inform a productive displacement.

Indeed, some anthropologists have rejected the notion of people as infrastructure altogether, warning that to call almost anything infrastructure means ‘ignoring some of the term’s classical defining features, such as its stable materiality and the techno-scientific and political dimensions of infrastructural planning, construction and maintenance’ (Calkins & Rottenburg 2017: 254). Looking backward, studies of infrastructure in STS started as an intervention against a one-sided focus on the materiality of technologies and the science and engineering that undergird it. The relational concept proposed by Susan L. Star and Karen Ruhleder (2001: 308) framed infrastructure as ‘something that emerges for people in practice’. Later studies that claimed that people are infrastructure have pushed the insight that there is no infrastructure without people to the other extreme by disregarding key features conventionally attributed to infrastructure.

Rather than debating categorically whether people should be viewed as infrastructure or not, it is therefore more productive to consider classical features of infrastructures and ask a different question: How do people become information infrastructure akin to eco-labels that enable a price premium? What role do science and politics play in the planning, construction and maintenance of people as information infrastructure? What is the material base of people as information infrastructure and what are its consequences? After discussing these questions in the following three subsections, I will turn to the insights this transformed analytics of people as infrastructure facilitates.

Information Flows between People: Valuing Trust Relations

Besides rice, Daxi’s peasant cooperative also produced vegetables, but these lacked eco-certification and sold for highly divergent prices at the consumer association in Chengdu and at the Old Market in nearby Yinhe, a city with more than 100,000 inhabitants. To understand how a price premium was possible in one place but not the other, I examine how information about the produce was communicated to consumers in both sales locations, focusing in particular on the enabling material base.

As well as the plain white sacks of rice, Wang Zhaochen also delivered vegetables to the consumer association weekly. The day before a delivery in May 2015, the eight villagers who worked at the two vegetable production bases in Daxi had harvested one bamboo basket of green chilli peppers, another bigger basket of aubergines, a plastic crate of green beans, and several bundles of long beans. This small shipment easily fit in the cooperative’s microvan. While I helped the shop assistant at the consumer association arrange the vegetables in basketwork trays on the salesfloor, she complained that they had delivered too many green chilli peppers again: those from the last delivery several days before had not yet sold out.

By six o’clock in the morning that same day, Wang’s son had already delivered a bigger shipment of a few hundred kilograms of vegetables in the cooperative’s pickup truck. He had driven this cargo, along with two villagers, to the Old Market at the East Gate in the city. Zhou Yueying, a member of the cooperative’s board and the village committee, and another woman arranged the vegetables and stayed at the market until they had sold everything.

The cooperative could only afford a cheap location on the outskirts of the Old Market and the required waste disposal fees. To avoid having to pay for a stand they displayed their vegetables on plastic bags and tarps spread on the ground. The prices for different varieties of vegetables ranged from 2 to 4 CNY per jin (0.57 to 1.14 EUR/kg) according to what everyone called the ‘market price’. The cooperative could not collect an organic premium above that. Customers at the association’s shop paid a high fixed price of 10 CNY per jin (2.86 EUR/kg) for most varieties of vegetables (such as aubergines) delivered by Daxi’s cooperative. When I told the foreman of the cooperative’s vegetable unit about this, he expressed surprise at the price difference. He said he would have expected the price to be no more than twice what they could get at the Old Market, where the cooperative sold in much greater quantity.⁶ How are we to understand this enormous price difference for vegetables grown through the same labour, from the same seeds, in the same soil, and harvested side by side in the cooperative’s plots?

The consumer association’s shop did not just display posters about selected farmers and cooperatives. Members also shared product information with each other. While some only came to pick up their purchases, others stayed longer, and some assisted the paid shop assistant as volunteers and took on a lot of the communication work. Chairs, a large table, and some very basic kitchen facilities even made it possible to eat together occasionally. Moreover, a shelf of books offered insights into healthy food and ecological agriculture, as well as peasant and consumer movements. All in all, this offered a stage to perform personal relatedness and enact shared concern about such matters. In particular, Wu Yingying and other long-standing members passed on news from producers they had recently talked with in person, on the phone, via social media, or during a visit to their farms (I will return to such visits below). Besides stories and anecdotes, the news exchanged among members included descriptions of farming processes including those that did not yet meet organic standards. For example, customers were told when certain fruits have been sprayed with pesticides in order to allow consumers to judge by themselves whether they regarded them as safe to consume. On the producers’ side, Dong Jie was the key figure in communicating the ecological quality of the vegetable production of Daxi’s cooperative to Wu Yingying and others. When Wang Zhaochen or his son, sometimes accompanied by Dong Jie or Kang Sunbin, delivered rice and vegetables, Wu Yingying welcomed them effusively. Instead of their official titles, she addressed the men as ‘elder brother’, thereby performing personal relatedness and distancing ‘the state’ and its supposedly rigid certification schemes (Lammer 2018).

At the Old Market the cooperative did not put up signs that explained its ‘ecological agriculture’ and enabled consumers to notice a difference. Consumers used their usual techniques of acquiring information about food at such markets. Besides considerations about visual characteristics and price, this included expectations regarding food safety. Zhou Yueying, the woman responsible for selling the cooperative’s vegetables, emphasised the importance of skilfully preparing and arranging the produce by, for example, removing insect-damaged outer leaves, but as late as possible in order to prevent the inner leaves from wilting. However, I overheard some customers wonder how the cooperative’s cabbages could possibly be grown ecologically since the leaves had no insect damage at all. For health reasons, some people recommended buying cheaper vegetables that had a less perfect appearance, taking signs of pest damage as an indication that no chemicals had been used.⁷ Moreover, vegetables from ‘local peasants’ were often considered safer than those sold by traders. As the cooperative usually produced a wide variety of vegetables, consumers at the Old Market often mistook it for a conventional trader selling potentially ‘unsafe’ food from a far-away and unknown elsewhere.

When time allowed, Zhou Yueying did personally inform some consumers about the cooperative and its ecological cultivation of vegetables. At first glance, this seemed not unlike what happened at the consumer association’s shop. After having sold the cooperative’s vegetables for three years, she told me in 2015 that she had made many acquaintances who trusted her at the market. Several people regularly bought vegetables at the cooperative’s stall and some of these even visited the vegetable plots at Daxi Village. These regular customers would sometimes be given vegetables for free. But normally, they paid as much as those who did not get into conversations with Zhou Yueying, did not trust her, or did not visit the village.

In both sales locations, consumers valued personal experience as reliable and personal relations as trustworthy (as in many other places around the world, such as the postcolonial Francophone food network described by Freidberg 2004: 127–166). In both, information about the ecological quality of the cooperative’s vegetables flowed from people to people. And in both, some consumers came to trust their interlocutors and appreciated the vegetables they bought as healthy and environmentally friendly. In only one did this translate into higher prices, however.

In both locations, only certain people communicated directly with producers or visited them, not all customers. While such interactions turned specific vegetables into ecological ones for individual customers at the Old Market, the consumers’ association in Chengdu successfully performed this transformation for its products and shoppers in general. As Frohlich (2017: 163) points out, infrastructure is about ‘scale’ and ‘action at a distance’. This scaling-up from individual interactions was achieved by representing and performing specific personal relations as part of a broader network of information flows. While there was occasionally time for in-depth conversations with customers when the Old Market was not bustling with life, the consumer association provided both organisational and material background work for performances of personal relations and common matters of concern. This background work relied on scientific and political planning. Widely shared discourses about trust relations have been translated with the help of social-scientific concepts into activist models of organisation. These models retained some of the functions attributed to labels in the management literature: not only reliable information for consumers, but also price premiums for producers.

Designing People as Alternatives to Labels: Social Science and Activist Models

The consumer association was established based on experiences with activist models of social organisation, such as ‘food cooperatives’ and ‘community-supported agriculture’ (CSA). The founder of the consumer association, Wu Yingying, was familiar with alternative food initiatives from living and studying abroad in the USA and in Europe. She and other members were also in close contact with activists and scientists from the ‘New Rural Reconstruction’ (NRR) movement – a rural cooperative movement that emerged in China in the 2000s (Day 2008). These links were made visible: for instance, a food blogger directed the attention of her readers not only to Wu Yingying’s rigorous ‘personal testing’ of products but also to books about NRR that she discovered in shelves at the consumer association.

Among the most prominent protagonists in NRR was Professor Wen Tiejun, an agricultural economist who had served as dean of the School of Agricultural Economics and Rural Development at Renmin University of China. There was also a lively exchange with Western scholars like Jan Douwe van der Ploeg, a retired Dutch academic from Wageningen University, who also teaches rural sociology at the Chinese Agricultural University in Beijing and included a chapter on China in the second edition of his now classic New Peasantries (2008/2018). Activist scholars such as He Huili, now a professor at the Chinese Agricultural University, organised experiments with direct marketing and subscription farming under the slogan ‘mutual aid between city and countryside’. Shi Yan wrote her dissertation about ‘Beijing’s first CSA’, which she initiated based on her experience as an intern at a CSA farm in the US.

There were several different streams of thought in NRR; cooperation was discussed in different theoretical terms such as ‘embeddedness’, ‘community’, ‘networks’, ‘social capital’, or ‘transaction costs’; and experiments have been made with different versions of direct marketing. Nevertheless, one core idea appears to be widely shared within networks of ‘alternative’ food initiatives in China (and elsewhere): ‘trust’ emerges between producers and consumers who ‘link up directly’. Some activists and academics have also contrasted state standards with alternative food movements, drawing on the work of the preeminent Chinese anthropologist Fei Xiaotong and his concept of the ‘relational morality’ of ‘the differential mode of association’ (chaxu geju) in ‘earthbound China’ (xiangtu zhongguo), along with other anthropologists of peasant China such as G. William Skinner (Merrifield 2020). In CSAs (and similar forms of direct marketing), ‘information about the farm, about organic food and about sustainable living’ is said to be ‘borne along relational ties, flowing outward to expand the network of people that know and trust the farm.’ (Shi et al. 2011: 557) This model is often explicitly presented as an alternative to certification. Shi and others (2011: 553) even describe ‘a good reputation gradually established through word-of-mouth’ as ‘informal, participatory “certification”’.

While obviously not all members of consumer associations like the one in Chengdu were so well-versed, or even interested, in social science literature, many of the organisers were and several had studied with Wen Tiejun and other NRR intellectuals. Both organisers and other members of the association communicated, alongside information about producers and their products, the basic idea that personal ties built trust and enabled a flow of information that was as reliable as – or even more reliable than – certification and therefore justified higher prices for the producers. In this sense, people as information infrastructure were planned through scientific and political background work, not unlike other types of infrastructure. In this case, social scientists and activists do for food networks what engineers do for networks of roads, wires and pipes; or, more directly related to information infrastructure, what marketing experts and lawyers do for labels.

But these design efforts alone do not yet explain why the ‘ecological’ vegetables were produced in the first place when most of them ended up being sold for the regular ‘market price’ at the Old Market. Therefore, I now turn to the materiality of people as information infrastructure and spotlight consumer visits to the ecological village.

The Materiality of People as Information Infrastructure and Its Consequences

As we argued in the introduction to this special issue, as various infrastructures of value become interlinked compatibility issues between them arise: for example, between information and transportation infrastructure. When I first visited Daxi Village in 2013, the county road was in terrible condition. Numerous potholes challenged drivers and vehicles and made delivery time – and cost-intensive. The broken concrete not only slowed transport considerably but the vibrations it caused also harmed the appearance and quality of vegetables by causing bruises and cuts. In spring 2015, maintenance work on the county road was completed, including a new asphalt surface. This not only allowed transporting fresh vegetables to Yinhe more smoothly and faster but – in conjunction with a highway opened in the mid 1990s – to Chengdu. It also enabled middle-class consumers to make a comfortable day trip from the provincial capital to Daxi Village (less than two hours each way), without worrying about damaging their cars. The improved materiality of transportation infrastructure was thus a prerequisite for people becoming information infrastructure.

Visits to the ‘ecological village’ also required material facilities suited to the consumers’ class and lifestyle. While travel by private cars depended on infrastructural investment by higher levels of the state, the cooperative built paths suitable for the clothing and shoes of the wealthy urbanites they imagined strolling through the vegetable plots. By 2013, they had already laid pre-cast concrete paths between the cooperative’s vegetable fields and at the edge of nearby rice paddies. These paths were to enable field visits by higher-level officials, journalists and urban middle-class consumers, who were assumed to want to keep their shoes clean as they informed themselves about the ecological origins of their food. Dong Jie frequently guided consumers through the village’s fields, both the rice paddies and the cooperative’s vegetable plots. Except for special occasions like transplanting rice seedlings or harvest time, visitors spent most of their time in the vegetable unit, which was located centrally at the recently built office of the villagers’ committee.

Giving consumers the possibility of visits to the village to ‘link up’ with the peasants and ‘inform themselves’ about the cooperative’s ecological food also depended on the specific materiality of different crops, related farming practices, and modes of production. In August 2015, the board of the cooperative (the village party secretary, the village leader, several leaders of villagers’ groups, and the villager who served as the technical expert for organic rice and cultivated the largest area of rice for the cooperative) met to discuss the future of the vegetable unit. A board member summarised the situation. From an ‘economic point of view’, the vegetable unit had been losing money since its establishment a few years ago. For two years, the cooperative had even used part of a state grant officially earmarked for the ‘construction of basic infrastructure’ to pay the wages of the workers of the loss-making vegetable unit. This board member, however, also highlighted its ‘representational and political value’. Nevertheless, others on the board proposed giving up on it and concentrating on the more viable rice business.

As mentioned above, almost every household in the village cultivated rice to meet its basic subsistence needs and the cooperative could eliminate any financial risk by only buying unhusked rice from villagers when there was customer demand. As that was increasing, the future challenge for the cooperative appeared to the villagers’ capacity to supply enough paddy rice. In 2015, the cooperative’s leaders were very optimistic and even tried to persuade one villagers’ group to sell its entire future harvest to the cooperative. In exchange, the cooperative offered them ‘ecological’ rice from a different cooperative in another county that had just started to transition to organic production under the guidance of the Daxi cooperative’s technical expert but not yet been certified. However, the villagers preferred to eat rice they had grown themselves and rejected the proposal.

In contrast to the rice unit, running the vegetable unit had fixed costs since the cooperative leased villagers’ use rights to land and paid villagers to cultivate vegetables there.⁸ The majority of its board felt that the economic losses outweighed the representational and political value and voted to shut down the vegetable unit. Only an intervention of the township agronomist after the meeting changed the plan. The vegetable unit was too important in his strategy of marketing rice to organised urban consumers.

This had to do with the materiality of the cultivated crops. Grains such as rice grow more slowly and are less labour-intensive than vegetables such as celtuce (stem lettuce), green chilli peppers, aubergines, tomatoes, and cauliflowers. Although it took more labour to cultivate organic paddy rice than to use agrochemicals, each season still only required a few days of intensive work in the paddy fields when preparing the soil with manure and rapeseed straw, transplanting the seedlings, weeding, and for the harvest.⁹ While some outside visitors came to Daxi specifically for those events, the vegetable unit offered more opportunities – and, importantly, more regular ones – to demonstrate ‘ecological’ agriculture to visitors. Moreover, paid workers could also be assigned to labour-intensive tasks such as applying biogas slurry that were key to demonstrating certain standards of ecological farming that Dong Jie favoured and promoted.

Because the cooperative relied on people as information infrastructure to sell valuable ‘organic’ rice in the consumer association, it thus also produced ‘ecological’ vegetables, even though this meant selling most of the surplus produce at ‘market price’ at the Old Market and foregoing the ecological premium. Making information flow through people ­ – rather than through labels – thus not only involved significant background work; it also depended on, and affected, various materialities.¹⁰ As with labels, this shaped value also in unexpected ways unforeseen by the designers of this information infrastructure.

Conclusion: Value and the Productivity of Information Infrastructures

In this article, I tried to understand the highly divergent prices for virtually identical food by following the approach pursued in this special issue: connecting infrastructure and value. Methodologically, my ethnography followed a cooperative’s products from fields to consumers. To avoid the effects of ‘theoretical ventriloquism’ (Hirschauer 2008) and ‘theoretical puppetry’ in analysing supposed alternatives to eco-labelling, I strove for analytical symmetry by looking for two specific features conventionally ascribed to infrastructures: planning and materiality.

By extending Frohlich’s (2017) argument about labels as information infrastructure to people, the opposition between ‘reductive information’ through labelling and supposedly more holistic ‘cultural knowledge’ before or without standardisation started to dissolve. Writing on people as infrastructure in Africa has thus far emphasised (tactical) improvisation, but largely overlooked the scientific and political (strategic) background work involved. In the case of people as information infrastructure in a consumer association in China, I highlighted the productive role of social-scientific concepts in undergirding activist models of alternative food initiatives. While ideals of trust-based relations between specific people circulate in many places, drawing up and promoting personal relations as a generic design for an alternative to labels produces different effects by generally guaranteeing reliable information about certain product qualities and justifying price premiums. Enacting such designs of people as value-adding information infrastructure requires specific material infrastructures. David Graeber (2001: xii) has theorised value as ‘the way in which actions become meaningful to the actor by being incorporated in some larger, social totality’ (see also Lammer and Thiemann, this issue). The focus on the infrastructural features of alternative food networks helps understand how the relevant whole is assembled (here a collective of consumers for whom a product is enacted as ecological), in relation to which certain actions (such as specific farming practices) become meaningful and valuable.

As we stressed in the introduction to this special issue, adjustment between different infrastructures of value is necessary to achieve compatibility. This holds true for both labels and people as information infrastructures. Agricultural production needs to be adjusted to the means used to communicate product information to consumers. Eco-labelling, modelled on the scale of Western agriculture, requires adjustments of the legal-organisational infrastructure (particularly the establishment of large-scale production bases such as peasant cooperatives, ‘dragon head enterprises’,¹¹ or European-size household farms) if transferred to irrigation-based farming and the extremely small scale of post-Mao agriculture in China. In the alternative food network in Sichuan, further rearrangement was necessary. People as information infrastructure, although seemingly more flexible and provisional, also required specific infrastructures, crops and farming methods. In order to perform ‘personal experience’ and ‘direct links’ with ‘the peasants’, a stable stage for regular possibilities for interaction through farm visits had to be created. This was not afforded by the profitable, but labour-extensive production of organic rice, but only by the expensive, labour-intensive production of ecological vegetables that created credibility for the former.

Rather than providing consumers with a fuller picture and freeing producers from negative effects of standardisation, people as information infrastructure are thus as productive of unexpected (or even unwanted) effects as labels. Information infrastructures – both seemingly generic labels and seemingly unique people – do not only convey information about materiality; instead, they create, shape, or destroy value as their entanglement with materiality intervenes in what and how it is produced.

The ‘politics of materiality’ and their consequences for inequality do not disappear if labels and standards are substituted by networks of people. While specificities may change, producers of agri-food are still faced with the challenge of having access to certain material infrastructures that people as information infrastructure require. The construction and maintenance of such infrastructure, for example for road transport, are outside of the power of individual producers, cooperatives, or even a whole village. Moreover, even if unwritten, specific dominant ideas about proper ‘ecological’ agriculture are conveyed through people as information infrastructure; and producers need to adopt production to crops, farming methods and modes of production with a suitable materiality if they want to participate and compete successfully in the alternative food network. While power may not be shifted to techno-scientific experts and bureaucrats as in the case of labels and related standards, it moves to other experts, including those trained as, or informed by, social scientists. For this reason, the techno-material preconditions and consequences of our own concepts being translated into organisational practice deserve more attention.

Acknowledgements

A first version of this article was presented at the workshop “Consuming the Unique: Food, Art and the Globalizing Infrastructures of Value” organised by Daniel Monterescu and André Thiemann (9-10 May 2019, Central European University, Budapest). Critical questions and helpful suggestions by André Thiemann and the two anonymous reviewers helped me clarify my argument. Last not least, I thank Daniel Flaumenhaft for language editing.

Disclosure Statement

No potential conflict of interest was reported by the author.

Additional information

Funding

Research in China (2014–2015) was supported by a Marietta Blau-Grant of the Austrian Agency for International Mobility and Cooperation in Education, Science and Research (OeAD-GmbH), funded by the Austrian Federal Ministry of Science, Research and Economy. Language editing and open-access publishing was supported by the University of Klagenfurt.

Notes

1 Names of persons and of places below the prefecture level have been anonymised.

2 On the link between containers and standards, see Cronon (1991: 104–119), also Krüger (this issue).

3 Even this acknowledgment of personal relations in standardisation falls short of true symmetry. Certain analytical terms (such as the notion of experiential knowledge, or mētis, borrowed from James Scott) as well as potential positive effects (such as reinforcing commercial obligations and tempering competition) are not applied to certification processes (Freidberg 2004: 127–210). In such an asymmetrical analysis, standards and audits necessarily end in tyranny (Freidberg 2004: 220–222; Strathern 2000: 313). More recent studies have instead pointed out that the effects of standardisation are not standardised but context-dependent and time-specific: they rely, specifically, on the actors’ numerical competence and reflexivity (Mugler 2019) – which is arguably just another kind of mētis that emerges when practicing standardisation.

4 In Daxi, an average household of three to four people had use rights to about four to five mu (0.27 to 0.33 ha) of arable land, of which only a bit more than one mu (0.066 ha) was irrigated land suited for cultivating paddy rice. Some households also planted rice on land assigned to people who had left the village, but none cultivated more than eight mu (0.53 ha).

5 Although very few households earned most of their income through agriculture, almost all households planted rice for their own consumption, even those who relied mainly on wage labour or non-agricultural small businesses.

6 At the consumer association, prices for vegetables were 2.5 to 5 times higher than at the Old Market. This price was high even compared to those at to other consumer associations in Chengdu. Another association arranged with a farmer to hire peasants in a suburban county to produce vegetables throughout the year for a fixed delivery price of 7 CNY per jin (2.00 EUR/kg). 1 jin is equal to 1 pound or 0.5 kg. 1 CNY equalled about 0.143 EUR at that time.

7 Stories circulated about hawkers poking holes in vegetables with needles if they looked too perfect.

8 About five to ten villagers were regularly employed by the vegetable unit. In 2014 and 2015, the daily wage of 50 CNY (7.15 EUR) was low, not only compared with the 100 CNY (14.30 EUR) an unskilled worker could earn in the city but also to the 60 CNY (8.58 EUR) usually paid in the countryside.

9 Rice cultivation can be de-intensified or include (at certain peak periods) very labour-intensive strategies. As Lena Kaufmann (2021: 187–227) highlights, decisions between cultivation strategies are based not only on available technologies but also integrated into larger ‘projects’ that citizens with rural household registrations pursue.

10 See also Sippel, this issue, on how land valuation practices not only measure but affect the materiality of what is supposed to be measured.

11 ‘Dragon head enterprises’ is a title the government confers on large-scale agribusiness firms that it expects to lead, and supports in leading, the way to ‘efficient’ and ‘modern’ agricultural development by replacing supposedly ‘underdeveloped’ and ‘unproductive’ small-scale farming (Schneider 2017).