Anthropological considerations about cattle-related beliefs and behaviors along the lines of the bovine mystique problematic were, at bottom, irrelevant. In other words, it did not matter why West African peasants wished to keep their livestock and resisted to interventions aimed at changing their cattle-related practices. The fact was that they did, and it was up to the project to adapt to these circumstances rather than the other way round. This working notion of the milieu paysan in the C-4 countries was, in its broad outlines, the one miniaturized in quantifiable form as the control situation in the adaptive experiments set up in the various project plots. It did not correspond to how peasants actually did things, but to how they were supposed to be doing them if they followed the technical recommendations crafted by the research institutes and delivered by local extension agents. The African partners and their Brazilian counterparts did not however close their eyes to the potential – or better said, inevitable – “noise” between research stations and the peasant environment. Rather, as will be discussed shortly, they recognized that peasants were likely to both follow and not follow the recommendations. This ambiguity reflects the broader “uncaptured peasantry” problematic discussed, among others, by Mamdani . As this author remarked in his critique of two polar approaches in academia and policy-making circles,African peasants are neither capitalists in the waiting, eager to be freed from the shackles of state patronage, nor labor of a pre- or antimarket kind that have failed to be successfully captured by the state or global capital. It is neither one nor the other; the challenge is to trace empirically how processes of capturing peasant labor for global markets have succeed or not,stackable planters but most likely have been caught somewhere in-between. This “in-betweenness” resonates with some of our previous discussions: the cotton filière’s relations with global processes, the technology trap remarked in Chapter 3, or the lag between soft and hard domains discussed in Chapter 2.

In fact, in its general outlines, it could even be extended to describe the postcolonial condition at large. And as many postcolonial scholars have done , rather than striving to overcome these ambivalences and contradictions, the project tried to work through some of them, with the tools it had at hand. The next section will describe a third spatial-temporal scale assembled by the project, in fact its most visible face: the so-called parcelle.The picture above, taken in October 2011 at the Sotuba research station, shows four Malian peasants who had been randomly approached in a cotton farm not far from Bamako, and invited to come see the project grounds. They were shown how cotton was supposedly grown in the milieu paysan, alongside how cotton would be grown using the new seeds and crop management techniques brought from Brazil. Little by little, the project technician explained to the farmers what was new about the plot with the Brazilian technologies, while pushing off the branches to call attention to the residues from the previous year’s crops in-between the lines; inviting them to feel the temperature difference across the height of the cotton plant, significant enough to be noticed without the need of a thermometer ; and taking them to notill’s chief visual enabler, the pit, where they had access to what was going on with the plants’ roots under the ground . This kind of comparative device is basic to adaptive research and transfer anywhere; new technologies are not just introduced into a random sample of the local environment. It is always against a certain rendition of “business-as-usual” that they will be experimentally calibrated, and comparatively evaluated. This evaluation is done not just by researchers and technicians, but by those who are to be the ultimate recipients of the technologies: in this case, peasant farmers. This rendition of the local context is therefore as constructed as the one to which it will be compared, that includes the travelling technologies.

But the Malian peasants in the picture were impressed by both sides in the experiment: not just the crop lines managed using the Brazilian technologies, but the control situation representing the milieu paysan, yielded beautiful and loaded plants. At a first glance and without expert guidance, the differences between them would not be easily perceived, but both were clearly more productive, healthy and homogeneous than the plants found in their own fields . This is not surprising, given the discussion already made here – even if the seed sowed in the experimental fields and in the farms were the same, the assemblages of which they were part were very different. Research institutes had not only more resources at their disposal, but much more control over their experimental fields; they were in this sense closer to the original context where technologies were developed in Embrapa’s own research stations. This was especially true of the project plot in Sotuba, which enjoyed a budget of its own, full time technicians, and the dedicated presence of the Brazilian project coordinator, himself an agronomist. The only significant environmental variable that was not controlled – in other words, that was fully shared between experimental and farmer fields – was rainfall. What these peasant farmers were seeing however were not finished technologies, ready to be brought to their fields, but adaptive experiments. These fields had a hybrid character. On the one hand, they were experiments, generally structured along the same lines of the ones routinely set up in research stations both in Embrapa centers in Brazil and elsewhere in the African institutes. In this sense, they brought non-humans into the project assemblage in a controlled and directed manner. On the other, they were demonstration sites, assembling various kinds of humans as a supporting public for the project. The project’s main experimental field was located in the Malian institute’s research station in Sotuba, in the outskirts of the capital city of Bamako. Brazilians referred to it as parcela, and the Africans as parcelle – the English equivalent land parcel is close enough to be maintained here. While technically speaking, parcel refers to the area where the experiments were being performed, the word was used more broadly to identify the piece of land that the Malian institute had assigned to the project. Different from its counterparts in the other three project countries, this parcel was supervised closely by the Brazilian project coordinator along with the Malian experts in each of the three components .

The local institute provided the project with two local technicians to be full-time at the researchers’ disposal in order to execute the experimental protocols and other project tasks . The project grounds in Sotuba were made up of two parcels, side by side. The smaller included a varietal test where the ten new cotton varieties brought from Brazil were being adapted and displayed,stacking pots alongside their five regional counterparts. It also contained an in situ seed bank for cover crops brought from Brazil , and part of it grew local cereals as a preparation for future no-till essays. The same written word, vitrine, was used to designate this half of the parcel in both French and Portuguese, so I will keep it here.The remainder of the parcel, twice as large as the vitrine, contained the no-till test fields, the largest and most visible of all plots . At the left corner as one entered the main gate, one could spot the only building within the parcelle, a small shack built by the project to store equipment and seeds. It was located right by one of the few trees that were left to stand within the fenced area, under which visitors and project workers would happily gather for a break from the scorching Sahelian sun. A bit farther behind the shack, in front of the first field stood a couple of panels, identifying the project and displaying a schematic picture of the no-till production system. As one walked further through the aisle that separated this cotton field from the maize field to its right, another couple of panels stood on each side: one showed the symptoms of nutritional deficiencies in the cotton leaves, and the other did the same for maize . Like the pit , these were key visual enablers in the parcelle. Other panels, set up in the vitrine, displayed a schematic description of pest control modalities.When I got there in September 2011, the other half of the maize field grew niébé, a leguminous plant that has abundant biomass and fixates some nitrogen – two common criteria in Embrapa’s no-till research in Brazil. But different from crotalaria, niébé was used locally for food – something which could potentialize adoption of the system by peasants. Towards the end of that season, local women hired by the project harvested the niébé beans and the corn ears, while the stalks and straw – the biomass – were kept in place as a variable to be measured in that experiment. After a sample of the production harvested was carefully weighed and recorded, these and whatever other edible grain was produced were informally distributed to the local community, or donated with a bit more formality at charity events by the Brazilian ambassador and/or his wife. Between the two parcels, I would occasionally see a boy playing a monotonous beat in a drum; it was to scare away the birds, eager to get at the loaded heads of sorghum in the vitrine. On the extreme opposite side of the no-till parcel, behind the crops, a guardian – a middle-aged man from Côte d’Ivoire – would sit down everyday during the evenings and nights. Just a few meters from him stood the wall that separated the Sotuba station from the surrounding community; without surveillance, maize ears would not wait in the stalks to be harvested by the project.

These, as well as the women who harvested the crops and performed other operations such as weeding were recruited from the local community through the channels already available at the institutes, and were paid by the project. Its demonstration aspects were a bricolage of elements from different technology transfer methods commonly deployed by Embrapa in Brazil . Embrapa research units normally include unidades de demonstraçãothat display to farmers the institute’s technologies, so they can be evaluated and perhaps adopted by them. It may include a direct comparison with alternative methods deployed by local farmers, as was the case with the C-4 Project plots. Vitrine, which means “display window”, is also a demonstration device. In Brazil, they are set up in events such as agricultural fairs, where new technologies are displayed and explained through short lectures delivered to farmers and the specialized media. Treino & visita, which involves continuous training of technicians and extension agents in charge of disseminating the technologies among farmers through “visits”, resonates with the project’s capacity-building cycles and more informal training of local technicians. Finally, there were elements of dias de campo , when multiple farmer groups and individuals were invited to come see the technologies at the institute’s experimental stations, or on pilot tests set up in farmers’ land.Who were the addressees of demonstration in this case? The parcel’s first public was the researchers and technicians from the C-4 institutes who worked for the project. They weren’t however mere spectators: researchers participated in the design and evaluation of experimental protocols, while technicians worked in implementing them. Besides this core group, demonstration targeted other researchers and technicians working in the local research institutes and adjacent agencies , but not formally within the project. They were recruited as trainees for the capacity-building cycles, and were expected to become multipliers of the project’s technologies. Much of their training took place outdoors on the parcel, and the chief aim of demonstration in this case was to produce a “paradigm shift” in their minds – in particular, about not tilling the soil, which seemed counter-intuitive to most of them. For the technology to disseminate, this ultimately would have to happen along the entire transfer chain: as a Burkinabe agronomist put it, “in order to succeed in convincing others, we have to be ourselves convinced”.