Playing development roles: the political ecology of performance in agricultural development

Performance is a useful lens through which to analyze agrarian life, as performance illuminates the ways that farmers manage the complex socioecological demands of farm work while participating in social life and in the larger political economy. The dialectic of planning and improvisation in the farm field has produced scholarship at multiple scales of political ecology, including the global ramifications of new technologies or policies, as well as the hyper-local engagements between farmers and fields in the context of modernity and development. Political ecologists are also beginning to understand how affects, such as aspirations and frustrations, influence agriculture by structuring how farmers and other stakeholders make decisions about farms, households, capital, and environments. To understand farm work as a performance is to situate it within particular stages, roles, scripts, and audiences at different scales. The articles in this Special Section ask how farmers have improvised, planned, and performed in response to agroecological challenges, bridging scholarship in political ecology, development studies, and the study of agrarian landscapes through new empirical case studies and theoretical contributions. Agriculture both signals social values and fosters improvisations within farming communities' collective vulnerability to weather and the political economy. We argue that the lens of performance situates the political ecology of agriculture within the constraints of the political economy, the aspirations and frustrations of daily life, and the dialectic between improvised responses to change and planning in the field.


Performance and agriculture
Performance is a powerful and versatile analytic lens in social theory, describing ritual roles and dramas that reflect power in society (Turner 1970(Turner , 1980, the actions that signify and enact social identities that are always and already there (Butler 1990), the strategic presentations of self that help people navigate complicated social spaces (Goffman 1956(Goffman , 1959, the iterative adaptations and course corrections of skilled practitioners responding to variability in daily life (Richards 1993;Scott 1998), the interpretation of practices or events by participants and various audiences (Schieffelin 1998), or the use of scripts to justify decision-making and roleplaying in a development context (Vanclay and Enticott 2011). Scholarship in agrarian studies and development studies has used the language of performance to understand how farmers manage the complex socioecological demands of farm work while navigating postcolonial identity (Gupta 1998;Pandian 2009), state, colonial, corporate, or non-governmental interventions in indigenous technical knowledge (Richards 1985;Stone 2016), and global environmental change that threatens to impact marginal agricultural communities hardest (Crane et al. 2011). This Special Section asks how farmers have improvised, planned, and performed in response to these challenges, bridging scholarship in political ecology, development studies, and the study of agrarian landscapes through new empirical case studies and theoretical contributions.
In two influential works, Paul Richards (1989Richards ( , 1993 describes small-scale agriculture as a type of performance that requires improvisation, repertoire, experiential knowledge, stage presence, and an active audience. Just as musicians learn to 'play through' mistakes or improvise to match hall acoustics and audience moods when they play, farmers develop adaptive strategies to confront agronomic issues framed within a larger political economy. Richards argues that these completed performances, the farm fields, reflect not a preconceived design logic so much as an emergent improvisation to factors including weather, soil conditions, input costs, labor, and available seeds. In the daily performances of agriculture, farmers call upon an experiential repertoire to manage these environmental, social, and economic farm variables. To understand agriculture in practice, social scientists must understand how this improvisatory skillset is put to use in the hyper-local context of individual farm fields (Richards 1993: 67).
Reading Richards' metaphor at face value, performance is not a heuristic to describe differing presentations of self in the sense developed by sociologist Erving Goffman (1959), but a literal farm level response to droughts, pests, shortages, or other opportunities and constraints on household resources. Agricultural performances emerge spontaneously as improvisations when farmers struggle to keep their land, plants, households, projects, and livelihoods viable. Richards extends the language of performance and improvisation to describe agricultural decision-making, spurring research in indigenous knowledge (Stone 2007), sustainable development (Murdoch and Clark 1994), and adaptation in dynamic environments (Orlove et al. 2010).
The question of improvisation in situ challenges the analytic value of agricultural field trials used in many studies of agricultural development economics, and provides a data-driven critique of development programs that cast small farmers as incapable of managing farm resources or innovating in markets or households ( Brookfield 2001;Netting 1993;Richards 1985). Rather, these and other scholars described smallscale agriculture ethnographically, as an iterative practice based in a series of diverse, adaptive strategies to manage labor and resources over the long term within a limited space (Alcorn and Toledo 1998;Leslie and McCabe 2013;Netting 1993;Sumberg and Okali 1997). While providing a useful critique of agricultural development, the focus on improvisation does little to illuminate the stage. Scholarship informed by political ecology must also consider the wider context in which agriculture occurs (Batterbury 2017) -land tenure, capital, national policies, climate change, gender roles, and the other innumerable constraints of agricultural work. Political ecology and performance must be considered, then, in tandem.

The political ecology of agricultural performances
While performance has been a persistent theme in the scholarship of agrarian development over the last 25 years, performance has not always been explicitly connected to the larger political economy or theories of postcolonialism, which ask how small-scale farmers engage with development programs at local, state, and international scales (Agrawal 2005;Escobar 1995;Gupta 1998;Vasavi 1999). Empirically-grounded work has shown that farmers are not passive recipients of aid, but actively shape that development to their own needs (Vanclay 2004;Vanclay and Lawrence 1994). This shaping involves improvisations as well as planned labor and resource arrangements (Batterbury 1996) that must work within the disputes, risks, opportunities, and needs of the local community (Agrawal and Sivaramakrishnan 2000;Flachs 2016). This relationship is dynamic, as development intermediaries from states, NGOs, or the private sector intervene in farmer performances by providing new incentives, sources of expertise, and audience reactions (Kumar 2015). Small scale farming is often a moral as well as an agricultural process (Pandian 2009), and so the performance of development and the performance of agriculture are inextricably linked to quotidian farm life.
Political ecology has become a diverse term, applied with equal fervor in the analysis of capital and the socioecological consequences of US American lawn care (Robbins 2007), Costa Rican pesticide regulations (Galt 2009), obesity (Guthman 2011), famine (Ross 1998;Watts 1983), or the transnational coffee trade (West 2012) -to name a few. In the context of agriculture and performance, a political ecology perspective recognizes that farmer improvisations and repertoires are also constrained within social expectations and the political economy. As a discipline, political ecology is well suited to destabilizing the seeming inevitability of environmental decision-making and illuminating, for example, the political economy that incentivizes the consumption of sugar (Mintz 1986) or the racial disparities of migrant labor on American fruit farms (Holmes 2013). While we argue that improvisation and innovation are crucial elements in the farming process, and that this perspective corrects for some of the implicit ethnocentrism in agricultural development (Escobar 1995), they do not occur in isolation. Extending the performance metaphor, this wider context is the stage on which farmers perform. Completed farm fields tell a story enmeshed in both ecological, impromptu decisions and in a wider context of local expectations, development incentives, agribusiness quotas, and commodity pricing.
Where the notion of performance and improvisation suggest impromptu decision-making, agricultural performances also include elements of planning and staging (Batterbury 1996). In this collection, we use performance to refer to both the improvisations of farmers in the field and to the carefully constructed plans of farmers navigating a political economy.
This dialectic lies at the core of a political ecology of agricultural performance, which recognizes that improvisations occur within the constraints and opportunities of local and global socioecological life. As in Richards' description, musical metaphors are useful here. A free improvisation concert performed by a group of musicians illustrates evidence of both. The players construct their performance as a dynamic response to the sounds they heard earlier in the day, the acoustics of the room, their moods, or the inspirations they receive from each other. The process is entirely improvised from a repertoire of musical training if viewed within the context of the performance. However, other elements of the event require careful planning -set breaks are timed to coincide with drink orders, without which the venue would not be able to support experimental music; the sound equipment must be carefully arranged to ensure that musicians and audience members can hear clearly; musicians and hall managers lean on informal and formal networks to recruit friends and fans to the performance; the concert may be funded at the behest of an arts patron; and the event itself is likely to end in time for everyone involved to catch the last train or bus home.
The dialectic between planning and iterative impromptu decision-making argues against the inevitability of development programs or agricultural technologies by situating them within a wider social context. Exploring this push and pull also provides ethnographic nuance better suited to describing agricultural decisions and performances as they unfold. The improvisations of groups and individuals reveal how some farmers are learning to turn the opportunities of new markets to their own interests (Flachs 2016), as well as the agroecological consequences when local knowledge is eroded by a new private market (Stone 2007) or a new concept of environmentalist self-policing (Agrawal 2005). Similarly, attention to context can help explain why new technological innovations, ranging from hybrid seeds to keyboards to climate-smart policies, ultimately diffuse into widespread use (Rogers 2003). By analyzing agriculture as performance, political ecologists gain a new vocabulary to describe the contextual and improvisational elements of agricultural life. The constituent elements of performance, including audiences, stages, scripts, front-and back-stage personae, and repertoires, allow us to attend to two key perspectives in political ecology. First, the Marxist perspective that ecological and social relationships are based in historical and material conditions and relations, wherein capitalism extracts surplus from labor and resources (Robbins 2004;West 2012). Second, the ethnographic perspective that creativity, opportunism, frustration, lived experience, competing knowledges, and thus the variation in what people know and do shapes social, economic, political, and environmental relationships (Benson 2012;Tsing 2015).

Roles, stages, and knowledge in the field
Understanding these roles and the socioecological constraints on performances within them is a growing concern in studies of agricultural development. These performances include the "superlative sufferer" of medical development literature (Heller 2015), the "show farmer" (Flachs 2017;Stone 2014) in alternative agriculture, or the script of "insecurity" in infrastructure development projects (James 2010). For playing such roles, farmers and other development subjects receive economic rewards including access to loans, urban consumers, farm equipment, or part-time work; social rewards including a network outside of ethnic or kin relationships that provides support and resources to participating farmers; and the reward of prestige through certificates conferred for 'good' production, recognition in the media, or by gaining respect in the community because of this attention. None of these various roles can exist outside the political economy, and yet these roles, often carefully planned and cultivated, then create new opportunities and constraints for improvised performances on the farm. This dialectic between planning and improvisation presents an opportunity to consider performativity within an analysis of praxis. Theoretical critiques from gender theory (Butler 1990), practice theory (Ortner 2005), science and technology studies (STS) (Haraway 2003;Latour 2010), and postcolonial studies (Pandian 2009;Subramanian 2009) argue that performers and performances are not totally separate. That is, through quotidian practice, it may be inappropriate or misleading to understand farmer performances and farmer performers as distinct. Rather, they simultaneously create each other.
Theories of quotidian performance suggest that in agriculture, as in any daily activity, the separation between performer and performance becomes blurred with repeated practice and the building of skill. Judith Butler's (1990) work on performativity as well as the scholarship of subjectivity and practice theory (Bourdieu 1977;Ortner 1984Ortner , 2005 ask how performer and performance simultaneously create one another within the power dynamics that govern who performs on which stages. As Gregson and Rose (2000) argue, understanding performances as products of knowing, intentional agents totally distinct from a true self, other performers, audiences, and power is fundamentally inaccurate because all such performances are enmeshed in already-established norms and claims to authority. Rather, the presentations of self-emerge from iterative social and ecological feedback that determines what is learned and how that knowledge is used in the future.
With its focus on the idiosyncrasies of knowledge, context, and discovery, research in STS has linked the performance of knowledge to the performance of self. Viewing subjects as in a constant state of mutual composition (Latour 2010), in which one 'becomes with' the multitude of human and nonhuman actors (Haraway 2008) in the farm landscape is a means to understand performances of self as preconditions of material agricultural decisions. Each performance must adjust for new slight or major variations in the audience and stage, because the process of creating, adapting, and translating knowledge within this multispecies and transcultural milieu, does not allow knowledge to be separated from the actors who engage with it (Stengers 2005(Stengers , 2010. Although Goffman lacked a vocabulary to situate performance of self within power and expertise (Ong 2010) or the iterative subversions and incomplete, frictive development benchmarks (Pandian 2011;Subramanian 2009;Tsing 2005), these performances occur at the nexus of subjectivity and daily agricultural practice.
Dramaturgical metaphors help to link these quotidian performances of knowledge and self to the larger expectations of development or modernity -essentially guiding a discussion of how one ought to live (Pandian 2009). These subjective performances are inseparable from a specific postcolonial history that demands certain kinds of progress, strategic presentations of self and knowledge, and awkward reorganizations of knowledge and identity. When new technology arrives, increasing capital accumulation and commodifying labor and knowledge (Kloppenburg 2004), it also creates new authorities, experts, and subjects based on changes in and claims to agricultural knowledge, a phenomenon Akhil Gupta (1998) and A.R. Vasavi (1999) call hybridity. Yet these new elements are always reimagined by new audiences, belying a monolithic or inevitable global capitalism (Appadurai 1996;Tsing 2005) in favor of a modernity that has been "resisted, reinvented, and reconfigured in different social and historical locations" (Gupta 1998: 9).
Thus, to farm well is not merely to produce for one's household and to safeguard the future of the land (Wilken 1987) but to prove one's modernity, to show off for the community, to use new technologies, and to embrace the goals of states, agribusiness, NGOs, or patriarchs (Agrawal 2005;Pandian 2009Pandian , 2011. These new obligations and responsibilities engender new performances just as much as new technologies or policies. Indeed, striving to perform modernity or cosmopolitanism was seen to be a crucial factor in the adoption of hybrid maize (Ryan and Gross 1943), Green Revolution varieties of wheat and rice (Gupta 1998;Vasavi 1999), or tractors (Fitzgerald 2003), a move based as much in performative identity as in technological efficiency or active instruction (Stone 2016). States and agribusiness help to encourage such roles, as these generate capital and legible citizens or farms (Scott 1998). In some ways, this process can be subtle: export quotas, subsidies, sales representatives, or advertising pushes can encourage farmers or other petty commodity producers to adopt the language of agricultural capital. In other circumstances, institutions reshape landscapes in the service of a definitive political statement: defensive terraces in Peru (Langlie and Arkush 2016) and extensive territorial farming in the US Southwest secured military positions (Stone and Downum 1999) while plantations of sugar, tea, rubber, or cotton seek to civilize the land by bringing legible order and control to otherwise dangerous landscapes (Besky 2014;Dove 2011;Valenčius 2002).
Farmers must perform well in the field to fulfill duties expected of kin or neighboring households, but agricultural development expands this audience. In a development context driven by the moral question of being a good, responsible farmer, the audience extends to those gatekeepers of agrarian success: neighbors,