150-200 km away from the capitals Rio de Janeiro and Porto Alegre, the serra fluminense (Rio de Janeiro) and the serra gaucha (Rio Grande do Sul) are lead providers of fruit and vegetables for the metropolitan areas. They are both highly specialized, taking advantage of the mountain agro-pedo-climatic conditions. In both cases, health is a key issue for agroecological transition, and has been mentioned as the gateway by producers who have colleagues or relatives seriously affected by diseases such as cancer, and health-conscious consumers.

The serra gaucha food system in Antônio Prado / Ypê is an expression of the development of Italian immigrants who arrived in late 19^th century. Those small communities of colonos in isolated regions built complex agricultural and food systems that would fit all nutritional needs. Work was a key virtue, and still is. Kinship and geographical origin relations had a leverage effect to attract qualified labour, mainly from the family, but also to build adapted machinery with a local network of SMEs, and to weave a commercial network, all Brazil through. Agroecological farms are many and organized around participative certification (Ecovida network).

In Rio de Janeiro, former capital of the Empire, slavery was central in all sectors. Petropolis, in the serra fluminense, was the imperial city providing leisure to the court. Although immigration appeared in an early stage in Nova Friburgo (from Switzerland and Germany) the work force kept mainly provided by slaves, and land tenure was dominated by noble or patrician families. Their relation to land is more functional and less emotional: their sale and move to another land is more frequent than in Rio Grande do Sul. A large number of intermediaries and middle-men collect the products in warehouses (galpões) to transport and deliver them in the capital. A network or organic fairs (circuito carioca de feiras orgânicas) is proposed by Rio’s municipality.

Both regions are highly affected by climate change, in radically different ways. On the 15^th February 2 event occurred simultaneously : in Porto Alegre there was a farmers' demonstration to demand public funds in response to the 3-year drought, while the city of Petropolis was ravaged by floods causing over 200 deaths.

Through comparing both regions, I started to ask some questions to our ATTER colleagues, to check the accuracy of my fieldwork.

In Rio Grande do Sul, a strong tradition is associations and cooperatives leads to more than 40% of family farmers participating in a cooperative (highest rate in Brazil). But the participation is mainly led by technical consideration: “I participate to improve the results of my own farm”. Considering the 3 pillars of agroecology, “set of techniques” and “scientific field” are very lively while “agroecology as a social movement” remains weaker, from the interviews I could have. In Rio de Janeiro on the contrary, I more frequently heard a mention of “participation to improve the world or save the planet”. But a bias is possible and event probable, since the numbers of interviews was low.

In the communes of serra fluminense, I did not hear people mention the whole territory as a territory for agroecology. They spoke of groups or events in their municipalities, not in the territory as a whole. On the contrary, the serra gaucha seems to make senses for the farmers. But in both cases, we have to consider the interactions between the mountain regions and the metropolitan areas to understand the territorial dynamics. I was surprised by the fact that in Rio Grande do Sul the objective of “feeding our neighbours with good stuff”, now a prominent motto in Europe, is a second rank priority. It appears more appealing in Rio, with a higher proportion of neo-rural farmers, although a research shows that building short food chains is made difficult because of the social links between farmers and middle-men.

Another question ins about the definition and limits of agroecological transition. In Rio de Janeiro, most research has been made upon organic agriculture, as an advanced post of agroecology. The information about the influence on conventional farmers is still scarce. In Rio Grande do Sul, many of my interviewees spontaneously mentioned that neighbours adopted some “ecological” (this terms is preferred to agroecology) techniques either to improve their own farm management or to comply with the neighbour's needs.

The long-term configuration of local agriculture and food trade leads to contrasted situations for the newcomers. In Rio Grande do Sul, many young farmers keep improving their parents’ land and activities (many times with processing and specific commerce chain), while the influence of neo-rural farmers, who first have to access to land tenure, is higher in Rio de Janeiro’s case.

In this fieldwork, I tried to remain inside the limits of what is usually done as a consultancy: to deliver some “strong feelings” to be discussed and amended to Brazilian colleagues in ATTER, just as I do when a municipality asks for the expertise of Terralim for the local food system. ATTER allows us to confront this approach to the scientific knowledge developed by local research institutions members of ATTER.

As a conclusion, I would like to stress the importance of the leverage effect of a research-action methodology: coming from another country and pulsed by a perspective than highly different from local research institutions, I happily participated to raise some curiosity from a range of local actors, who are used to frequent “the usual researchers”: it was my role and my pleasure to play the role of an alien that people come to see.

All pictures by Gilles Maréchal