The team on the Brazilian side underwent some changes after this initial period, as the first coordinator left and new Embrapa researchers joined the project team in Brazil while others dropped it. Still in 2009, the project hired a coordinator to stay in Bamako on a more permanent basis. He was recruited from within Embrapa’s ranks, and even though not a researcher in the cotton center, he was an agronomist well experienced in the interface between research and farmers. The only Embrapa cooperante fully dedicated to the project, the coordinator was a key mediator between the Brazilian researchers and their African counterparts, and even between researchers and managers on the Brazilian side. In the opinion of some Brazilians involved in the project, much of the project’s success stemmed from his personal dedication and skills, and of those before him. Indeed, he not only executed basic managerial functions of making planned project activities happen on the ground and reporting regularly to Brasília, as did technical work at the experimental fields on a daily basis – weekends included. The annual cycle of project activities converted the project’s abstract orientations into onthe-ground practice. As remarked above,hydroponic gutter objective prescribed the construction of new research infra-structure in Sotuba: a biotechnology lab, a pest control lab, a cold chamber for storing germplasm, a meeting room, some office space, and a generator to guarantee reliable power supply.190 When I first visited Sotuba in March 2011, only the buildings’ basic structure was in place.
When I returned later on that year, the buildings were almost done, and much of the machinery to equip it had arrived; however, the actual equipping of the facilities was not yet on the near horizon. The new facilities were supposed to provide support for the project’s experimental activities and host capacity-building trainings, remaining as a legacy for the Malian institute after the end of the project. However, they could not be used at all during Phase I; this component of the project was the one to suffer most from delays in acquisition processes.Meanwhile, basic operations such as soil analysis were made through the infra-structure already in place at the institutes. The capacity-building trainings, on their turn, were carried out every year during the crop season, when the experimental parcel was in place. Typically, two or three researchers recruited from Embrapa’s cotton center would come to conduct them, one for each of the project’s technical axes: typically, the first one would be entomology, secondly agronomy , andthirdly breeding. I participated in the ones carried out in Sotuba in 2011. In that occasion, the first week of trainings targeted researchers and technicians who were participants or potential participants in the project, and the second included non-research personnel . In 2012, due to the coup d’état in Mali, trainings took place separately in each country . It was not until 2010 that the experiments were set up in the other three countries; they were always, therefore, one year behind Sotuba. Of these, I have been twice to the one in Burkina Faso, located in Finkolo, about 10km from Bobo-Dioulasso, in the country’s cotton producing region.
From the beginning, it was decided that each institute would set up its own experimental fields, following the project’s general outlines but adapted to local conditions and priorities. Taken together, the project’s three technical components cover much of what agricultural production systems are about: soil, water and nutrients ; the crop plants themselves ; and their predators and parasites . Knowledge transfer in each of these components involved key individual researchers from the four partner countries, who crossed the Southern Atlantic several times for special training at Embrapa units, presenting papers in Brazil’s annual cotton conference, and co-designing experimental protocols along with their Brazilian counterparts. They also participated in, and were expected to gradually take the lead of, the project’s annual cycle of capacity-building trainings, which was carried out separately for each project component. Different from the CECAT trainings, here capacity-building had a fundamentally technical character; but technical content was conveyed as much in the form of abstract knowledge as part of a demonstration of the situated context in which the technologies were developed and disseminated in Brazil. Neither did the trainings take the form of unilateral transfer of knowledge. They sought, rather, to elicit from the African partners both their interest in and commitment to the project on the one hand, and data on the local contexts that could be put to practical use in transfer to farmers on the other. But as in CECAT, the concrete effects of these comparative exercises did not immediately come to surface during the trainings themselves, even if there were methodologies devised to this end.
More fundamentally, it was expected that the African partners, especially the researchers who were directly involved in the experimental activities, would spontaneously take up a central role in figuring out how to put each technical component to work for the project. The three-pronged technical package assembled by the project was made up of technologies that are not necessarily found together in their original context. Each emerged from interactions between researchers and farmers, not necessarily in the same research centers and agricultural regions. In Brazil, cotton production cuts across the dichotomous configuration between large agribusiness and smallholder agriculture described in the previous chapter. It was originally a smallholder crop, grown in its perennial, creole version mostly in the Northeast, whose semi-arid portions have been posed by internal coloniality as the epitome of backward Brazil. Following the guideline of establishing research centers close to production areas, Embrapa inaugurated its cotton center in 1975 in the city of Campina Grande, Paraíba state . Less than a decade later, however, the boll weevil – a small beetle that had caused major losses to American cotton growers since the early twentieth century –arrived from the U.S. to Brazil in 1983, devastating production in these areas. Cotton would only reemerge as a significant commercial crop a decade later, and elsewhere: in the new agricultural frontier opened up in the Center-Western cerrado. In this new environment free of its nemesis pest, cotton disseminated as a large-scale, highly technified commercial crop. It was brought into a new production system that included other commodity crops, modern machinery, and conservation tillage techniques . In this process, the cotton plant itself changed significantly: from a semi-perennial plant in Northeastern agriculture, harvested by hand several years in a row by family farmers, in the cerrado it became a short-lived annual crop that is planted anew each year by cutting-edge machinery.Embrapa itself came to reflect this new assemblage. While its cotton center maintained its headquarters in the Northeast,hydroponic nft channel research nuclei and experimental stations were established closer to the newest production areas . These became responsible for developing cotton varieties and crop management systems adapted for large-scale commercial agriculture , while the center in Campina Grande became more restricted to niche domains such as breeding colored cotton varieties and agroecologic production systems, besides expanding its portfolio to other crops such as castor oil, sisal and groundnuts. One of these colored cotton varieties was included in the C-4 Project’s breeding portfolio, even if the African institutes’ chief interest was in the white ones demandedby the mainstream global textile industry. It is also in Campina Grande that entomologists did research on another project component: biological modalities of pest control, based on reduced use of pesticides and deployment of natural enemies to fight cotton pests. While Embrapa researchers participating in the project were drawn from both centers, the no-till experts have typically come from the Goiás or other off-campus nuclei . In what follows, I will describe, for each project component, how co-production between context and technology was taking place: that is, how the farming context to which the technologies were expected to travel was assembled, how the Brazilian technologies were being accordingly adapted, and how the immediate context for transfer at the research institutes was being shaped through capacity-building and other project activities.
This component was considered by many to be Project’s mainstay . From my perspective, most relevant is the fact that it was here where visible results were the most advanced at the time I did fieldwork; moreover, its systemic aspect made of no-till a good articulator of the project’s non-human assemblage at large. No-till was chosen as the main technique in the project’s soil component “due to the advanced stage of degradation of African soils, and Brazil’s vast experience in this sector” . Here, the availability of a successful experience in the original context for dealing with a problem diagnosed in the recipient context is attributed to common environmental conditions . Both within Brazil and in its South-South engagements, no-till is lauded as an effective soil and crop management technique especially suited for tropical soils, due to the latter’s higher potential for leaching and erosion relatively to their temperate counterparts. I was therefore taken by surprise by one of the Embrapa agronomists who one day explained to me the technique’s origins – a story the general lines of which he also presented to African trainees during the trainings. As it turns out, no-till is, as many others, a travelling technology: just like soybeans, it went from the United States to the colder areas in the Brazilian South, before it was “tropicalized” for the cerrado agriculture in the Center-West. Also unexpected was to learn that the introduction of no-till in Brazil did not originate primarily in government policy; it was, rather, the outcome of a joint initiative by farmers and research institutes, partly funded by resources from international development aid. One of these foreign partners was the German Cooperation Agency , same national origin of the first farmer to try no-till in Brazil in the early seventies, out of his own initiative. During the nineties, the system followed the expansion of the agricultural frontier and spread across the cerrado, coming to be deployed at increasingly larger scales for the production of export commodity crops such as soybeans, maize, and cotton. It is estimated that no-till accounts for around 70% of the total agricultural area in Brazil, today a “world leader” in this research field.No-till is not really one technique but a multifactorial soil and crop management system; as such, I will suggest, it is an especially fluid technology, in the sense put forth by De Laet and Mol . This system is always gauged in comparison to the so-called “conventional system”, and its principal aim is to enhance soil conservation – to sustain fertility and humidity over the long run –, often as a fix to a problem created by decades or even centuries of the latter. In the Portuguese/French common terminology , “direct” means that sowing happens without previously tilling the soil. In conventional agriculture, continuous tillage, especially by heavy machinery, creates over time an underground compaction layer that prevents the infiltration of water and roots deeper into the soil, leading to surface runoff and erosion. Nutrients present in the topsoil are then washed away, and the soil becomes increasingly poor and unable to hold fertilizers, water and other inputs. Part of the project work, in its capacity-building power points and demonstration fields, involved rendering visible these processes, which were normally concealed underground, out of the view of farmers or even researchers and technicians. In thus exposing them, the researchers simultaneously created them as problems for the project to tackle. According to the project, this situation of degraded soils, caused by continuous deployment of the conventional system, was the key factor behind low cotton productivity in West Africa relatively to other world producers. And this difference was not trivial: while in the C-4 countries productivity averaged between 800 and 1,200/ha, in Brazil it would reach up to, or even surpass, 5,000 kg/ha. Here, another scale came into play. These figures in themselves are not problematic, but they become so when brought together comparatively according to a common reference: on the one hand, a liberalized world market on which both Brazilian and West African cotton is sold, and on the other, a shared global technology treadmill. Productivity was indeed a key concern shown by both Embrapa front liners and their partners in the African research institutes before they came together for the C-4 Project. This lag, found in facts and figures standing for these broader scales, could also be dramatically visualized in the materiality observed in micro-practice.