The ox fall down: path-breaking and technology treadmills in Indian cotton agriculture

ABSTRACT

Although India’s cotton sector has been penetrated by various input- and capital-intensive methods, penetration by herbicide has been largely stymied. In Telangana State, the main obstacle has been the practice of ‘double-lining’, in which cotton plants are spaced widely to allow weeding by ox-plow. Path dependency theory primarily explains the persistence of sub-optimal practices, but double-lining is an example of an advantageous path for cash-poor farmers. However, it is being actively undermined by parties intent on expanding herbicide markets and opening a niche for next-generation genetically modified cotton. We use the case to explicate the role of treadmills in technology ‘lock-in’. We also examine how an adaptive locked-in path may be broken by external interests, drawing on recent analyses of ‘didactic’ learning by farmers.

KEYWORDS:

• Agricultural industrialization
• path dependency
• knowledge production
• biotechnology

Acknowledgements

For comments and discussion we are grateful to Dominic Glover and to some particularly thoughtful anonymous peer reviewers. For assistance during fieldwork we are grateful to the Rural Development Foundation and to N. Ranjith Kumar. For the paper's title we thank Robert Hunter.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 This includes 95.9 million hectares of HT-only crops and 58.5 million ha of crops in which the HT trait is stacked with a Bt trait (James 2015). Bt traits, which provide some protection against insects, are discussed below.

2 Fieldwork was conducted between 2000 and 2015 in the Telangana region of Andhra Pradesh State, which became Telangana State in June 2014. During this time ethnography was conducted in over a dozen villages in Warangal District, selected to represent variations in soil, caste and prosperity. Statistics presented here are drawn from five of these villages, as shown in Figure 1, for which repeat agricultural household surveys were administered to random samples stratified on land holdings. Sample sizes ranged from 144 to 254 households, except for 2009–2010 when sample sizes were only 43–44, and 2015 when sample size was 311. Because our sample includes farmers from diverse castes, irrigated and rainfed farms, and from villages both proximate to cities and in isolated rural areas, we argue that our claims are widely applicable to cotton agriculture in semi-arid India.

3 For example, ‘cotton follows the classic “pesticide treadmill” … namely, once the pests become resistant, even more pesticides will need to be used’ (Jin, Bluemling, and Mol 2015, 24).

4 In this environment, regular pesticide sprays also take on a social value associated with responsible farming – in a 2013 interview, one farmer explained, ‘you should always seek to produce more than your neighbors. If they spray four times, you have to spray five. That way, you’ll always have the best yield’.

5 A pesticide treadmill intervention in one village in the early 2000s temporarily weaned farmers away from excessive sprays and encouraged the use of neem powders, pheromone traps and other IPM methods. But after a few years most participants abandoned the IPM methods, as the IPM team never checked back with the villagers and then newly introduced Bt cotton seeds controlled pests just as effectively (at first) and with less trouble. However, during that same time period the same farmers began working with university extension services. This time the intervention was mediated through a cooperative agriculture supply store rather than an outside group, meaning that the program would feature a more permanent investment in the community. With this additional social commitment, the cooperative store has had a long-term impact on farmer management decisions that short-term interventions fail to replicate.

6 Recent studies show cotton leafworms have developed resistance to four insecticides (Nandakumar 2016).

7 To reduce the selective evolutionary pressure for Bt resistance in target pests, farmers are required to plant a pest refuge of approximately 20 percent of the planted area to non-Bt cotton. Some entomologists advocate mandating larger refuges (Carrière, Fabrick, and Tabashnik 2016), but such mandates are universally ignored in India. Monsanto has proposed mixing non-Bt refuge seeds in with the Bt seeds, which would effectively reduce the refuge area to 5 percent of the field, but has been prevented by approval process mandated by the recently reconstituted Genetic Engineering Appraisal Committee.

8 In August 2016, Monsanto withdrew its application for approval for RRF seed (Bhardwaj 2016), although since India is their second largest market this was understood to be a bargaining position.

9 The Mahatma Gandhi National Rural Employment Guarantee Act.

10 Many analyses of partial budgets are based on eliciting farmer recall of specific expenses that are not recorded and often remembered neither accurately nor precisely, and farmer responses are rarely checked by any form of triangulation. Most ethnographers know the veracity of such studies to be questionable.

11 This is a distillation of Henrich’s argument. He is particularly interested in a form of social learning called biased cultural transmission (BCT). He shows that S-curves of adoption are produced by BCT, or by mixed types of learning where BCT is predominant.

12 Acharya N.G. Ranga Agricultural University.

13 Agricultural extension reports also strike a defensive, nationalistic tone when stressing India’s need to increase yield, pointing out relatively low yields per acre (tied in these reports to inefficient non-chemical, low-tech agriculture), especially compared to China (Bagla and Stone 2012).

Additional information

Funding

Primary funding for this research was from the John Templeton Foundation initiative “Can GM crops help to feed the world?” The opinions expressed in this publication are those of the authors and do not necessarily reflect the views of the John Templeton Foundation. This paper also draws on fieldwork supported by the Cultural Anthropology program of the National Science Foundation, the Wenner-Gren Foundation, and the Jacob K. Javits Fellowship Program.

Notes on contributors

Glenn Davis Stone

Glenn Davis Stone is anthropologist whose research focuses mainly on ecological, political and cultural aspects of agriculture. In recent years he has been particularly interested in issues relating to indigenous knowledge, technology change and genetically modified crops in developing countries. His recent fieldwork has been in the Philippines, India and rural Appalachia. He is currently a professor of anthropology and environmental studies at Washington University in St. Louis, on leave as a Guggenheim fellow.

Andrew Flachs

Andrew Flachs recently completed his PhD in anthropology at Washington University, and currently is a Volkswagen Exchange Postdoctoral Fellow at the Karl Jaspers Centre for Advanced Transcultural Studies at Heidelberg University. His research interests include indigenous knowledge and agricultural change in both North America and South Asia.