Document(s) 4 de 38

Abstract: The erosion of South Asia's rich heritage of landraces has been paralleled by the erosion of farmer skills and community processes which created that diversity in the first place. This workshop brings together different groups with a shared, practical concern with the loss of both genetic resources and local control, with a view to exploring the common ground. Presentations cluster around three themes:

1. Diagnosis of varietal diversify. Key here are understanding, in particular situations, the forces that create and degrade diversity; differences among individuals and groups in knowledge and skills; and the efficacy of farmer-to-farmer and other seed diffusion channels. Diagnosis aims at identifying opportunities for useful interventions, primarily in the following areas.

2. Decentralized breeding and selection. Methods that involve farmers in the early decisions of what to retain or discard are being used to develop varieties of several crops that are adapted to local conditions, meet diverse preferences and that, as a set, are often more productive than what emerges from conventional processes. Farmers' selections should, in principle, enhance diversity on-farm, but evidence on this important point is still scarce.

3. Seedbanking and seed supply. In various South Asian environments, community-based initiatives are attempting to provide farmers continued and secure access to diverse seeds. On-farm testing of varieties is frequently a feature, as is small-scale seed multiplication that enlarges livelihood options, particularly for women. Initiatives also frequently aim at enhancing farmers' skills in managing genetic and other resources.

A number of opportunities are evident for increasing the scope and impact of field-based initiatives. Many of these will require collaboration between groups and institutions that, up to now, have not worked together. These include: making farmer-bred varieties from similar agroecosystems over a wider area available to farmers' selection; where feasible, genetically improving local landraces to overcome specific deficiencies; and creating functional links between local seed and format genebanks to enhance seed security and improve the banks' efficiency.

Introduction

To a large extent in the world outside, and even more sharply at this meeting, these perspectives are segregated by institution. Distinct camps have emerged around on-farm conservation and genetic improvement and enhancement, in which different issues are given priority and different vocabularies favored. In the sessions that follow over the next three days, it is striking that only one nongovernmental organization (NGO) representative will speak on breeding and selection, while no scientist from national agricultural research institutes is scheduled to present in the sessions on seedbanking and supply.

Can common ground be found? We believe it can, though how large is one of the issues this meeting may help illuminate. Common ground, we contend, lies in joining a dynamic and responsive conception of on-farm conservation, one that seeks to secure for the mass of small farmers continued access to the seed of varieties they value, to a decentralized, participatory process of genetic enhancement aimed at better meeting the diverse needs and preferences of those same farmers. Conservation and enhancement must be centered on farmers' own systems of use and production; both must be based on a deeper understanding of the uses to which people put diversity and of the forces that shape it. Among the key issues that presentations will highlight are:

What should we seek to conserve?

Several of the presentations during the next three days will describe the erosion of farming communities' control over the processes of crop evolution, which has paralleled the erosion of varietal diversity itself. A number of forces are at work, but all have the effect of limiting the choices available to farmers and skewing decisions among them. For example, P.V. Sateesh will describe the devaluation of locally-adapted varieties of small millets and sorghum in rainfed villages of Andhra Pradesh and the promotion of rice and wheat by a range of government programs, including the Public Distribution System. S. van Oosterhout recounts how, in Zimbabwe, public policy constrained farmers' options, first pushing them out of the market in the colonial period, then into it after Independence. Waves of genetic loss have been associated with the changes of regime, adding to the effects of drought, and threatening household food security. In the lowland rice-farming communities of Mindanao, Philippines, with which F. Magnifico's CONSERVE project is working, credit and extension programs have for many years promoted modern varieties at the expense of local cultivars.

Which diversity: varietal or genetic?

In principle, there is ample scope for collaboration between the two. Seedbanks require dependable back-up for security and to maintain varieties that are not immediateiy in demand but that may be in future. Genebanks would benefit from links with the growers of landraces who could identity material of potentially wide interest that easily escapes collectors' attention, and, at least as importantly, information concerning its traits. In practice, however, relationships between the two sorts of banks are at best embryonic. As brought out by several participants, among them F. Mazhar and R. Khedkar, considerable mistrust exists and issues of rights in and access to material and information must be overcome. We return to this issue later.

Diagnosing change

Farmers are likely to participate in a diagnosis if they can see a prospect of something useful emerging. Diagnostic methods must meet both researchers' need to understand and measure the dynamics of diversity and the very practical concerns of farming communities. Participants will describe their practical experiences in two principal realms: alternative breeding/selection approaches, and community-based seedbanking and multiplication initiatives.

Decentralizing breeding and selection

A few points bear emphasizing. First, decentralized breeding and selection will tend to produce varieties adapted to specific natural environments. It has long been argued that, as a set, these varieties should be more productive than the one or a few, selected for broad adaptation, that typically emerge from conventional procedures. The papers presented here lend welcome empirical support to the argument and show that by associating farmers when there is still meaningful variation to choose from, varieties can be developed or identified that satisfy a range of needs and preferences, including but going beyond grain yield. Further follow-up is needed to confirm that these selections are indeed more appreciated and retained by farmers, both those who participated in the initial selection and their neighbors. Evaluation of these programs has been limited and what there is has seldom been "at arm's length."

Second, beyond improved performance, decentralized selection and breeding should, in principle, result in greater varietal diversity on-farm than do conventional approaches. However, as J. Witcombe and A. Joshi show, there is as yet little field evidence on this score. They point to one case where participatory selection identified a variety that proved popular over a wide area and that may, in fact, have reduced diversity. Nonetheless, they remain confident that when farmers have more varieties to choose from, more will be chosen. Much will depend on the quality of what farmers have to work with and on the care with which parents are chosen in the case of breeding, or varieties assembled, in the case of selection. Over the longer term, the impact of participatory approaches will also depend on the response of downstream structures, notably seed multiplication enterprises, and their ability to assure farmers' continued access to diverse varieties. The response of policy makers will also be crucial, for example in reviewing varietal release procedures.

Finally, breeders working in some of the most difficult environments, where stresses occur in complex patterns, see a clear need for greater decentralization and participation in selection, but have yet to attempt it. These include B. Mishra, who has worked for many years selecting cereals tolerant of problem soils and J.L. Dwivedi, a deepwater rice breeder in eastern Uttar Pradesh. The latter describes scientists' meager success in developing varieties that survive the myriad combinations of flooding depth, duration and time of onset that farmers must contend with, sometimes within the same field. Instead, Dwivedi intends to give farmers a "genetic soup," i.e., variable, early generation populations from crosses of local varieties. Farmers will add their own selection criteria to those the natural environment imposes and he will monitor and further refine what results. This is essentially the "evolutionary breeding" approach that T. Berg points out was first proposed for stress breeding 40 years ago, but which Berg believes can be made more efficient in meeting farmers' needs by drawing them into the selection process. Novel and potentially fruitful approaches like this demand a new 'division of labor' between farmers and scientists.

Making "hidden" diversity visible

J. Witcombe and A. Joshi describe another form of diversity which may be hidden from farmers. In India, more than 500 rice varieties have been released in recent years, but most are not widely available, primarily because public seed enterprises concentrate on only a few, typically ones released more than 10 years ago. The situation is similar for other major crops. Slow and uneven diffusion, through either the formal or informal sectors, effectively limits farmers' choices, but also provides new opportunities for useful and effective interventions.

Seedbanking and seed supply

• R. Khedkar's Academy of Development Sciences (ADS) has collected more than 300 rices from the Konkan region of Maharashtra and is distributing 1 kg samples of some 60 of these to farmers in several districts. After testing, farmers return 2 kg, either to ADS or to their neighbors.

• In Bangladesh, F. Mazhar of UBINIG describes how women of the Nayakrishi Andolon dislike the idea of a centralized "bank," and are instead developing community "seed wealth centers" where farmers can obtain and exchange seed samples at no cost. Gram karmis orvillage workers also grow nurseries from which larger quantities of seed are sold, thus helping to sustain their own efforts and the initiative as a whole.

• The farmer volunteers of V. Jardhari's Beej Bachao Andolan (Save the Seeds Movement) are screening more than a hundred local varieties each of rice and common beans, as well as varieties of other hill crops, such as amaranth, that have generally received little attention from formal research. Those found promising are multiplied and made available to neighboring farmers.

• The sanghams (organizations of mostly lower caste women) with which the Deccan Development Society (DDS) is working in Zaheerabad, Andhra Pradesh, are rehabilitating patches of degraded land owned by these women and turning them into small but productive seed farms. As P.V. Satheesh explains, these are expected to form the basis of a distribution network for the seed of increasingly rare local varieties, at the same time expanding the livelihood options that are open to sangham women.

• F. Magnifico describes the efforts of the Community Based Native Seeds Research Center (CONSERVE) to select, bank, and disseminate seed of traditional rice and maize germplasm in S. Cotabato, Philippines, an area where farmers have long experience of Green Revolution technology. CONSERVE puts strong emphasis on safeguarding local knowledge of crop varieties as well as the germplasm itself.

These different efforts share a number of features. Most are working on varietal erosion as one facet of a wider process of agricultural change and aim to influence more than just varietal diversity. DDS, for example, also works on reclaiming "wastelands" and reestablishing a place in agriculture for coarse grains that are crucial for food security in poor households; UBINIG is concerned with a range of consequences of intensive rice cultivation, such as pesticide pollution and groundwater depletion. All groups appear to be working to empower farmers by enhancing their skills: ADS, for example, in hybridization, DDS in small enterprise management. Increased community control over agricultural production and innovation is the wider goal towards which most groups appear to be striving.

The programs are in each case quite young, some still embryonic. In all but one or two cases, it is too early to expect an evaluation of their reach, the extent to which they have touched different social and economic groups and the cost-effectiveness of their interventions. However, a number of concerns deserve further consideration. First, the links among these local initiatives and between them and institutions that might provide useful support appear to be poorly developed. The issues of seed storage and longer-term security have been alluded to above. It is unlikely to prove very efficient for communities themselves to maintain large numbers of varieties that are not in current use. Economies of scale will accrue to centralized genebanks or regional ones, such as that being developed by the M.S. Swaminathan Research Foundation in Madras. However, rights of access and control must first be worked out, which requires a foundation of mutual trust that is not yet evident. The genebanks have historically taken in "deposits" of landrace collections, but "withdrawals" by the communities that bred them have been much less frequent and indeed are difficult to accommodate within their operational designs. The emergence of community-based conservation initiatives requires new thinking in this area and a wider role for the formal sector banks in supporting local level "seed security."

A second area where broader institutional linkages may be of use is in helping these initiatives enlarge the varietal options available to farmers, making conservation more dynamic. This might take the form of access to a wider range of genetic material, which could extend to collaboration on enhancing local varieties through breeding. The need for such support is recognized by several of those directing the seed initiatives. R. Khedkar, for example, points out that the local rice varieties ADS is making available to farmers in the Konkan area are highly susceptible to the yellow stemborer, an insect pest that has apparently increased in severity in recent years.

Is erosion inevitable?

The experience in areas where Green Revolution technologies have been taken up generally supports such pessimism. The adoption of packages that include irrigation, agrochemicals and "improved" varieties has typically left in its wake a reduced number of cultivars, often closely related genetically. The outcome is unsurprising: as T. Berg points out, external inputs are intended to reduce environmental heterogeneity that people had earlier coped with by means of varietal diversity. But recent years have seen evidence of the limits of the intensive strategy. Real farm incomes have stagnated or declined in major cereal producing regions as crop prices have fallen relative to those of inputs and as environmental consequences from groundwater pumping and pesticide use have made themselves felt. In response, lowland farmers in several areas, such as those CONSERVE works with in the Philippines and UBINIG in Bangladesh, are gradually moving toward more diversified farming systems that rely less on purchased inputs, in which a larger number of more locally-adapted varieties are likely to find a place.

Whether renewed diversity in varieties and farm enterprises plays a part in increasing agricultural production or whether it serves mainly to reduce farmers' costs and secure their subsistence needs will depend on a number of factors, including the response of research and extension to these trends. But there are clearly grounds to believe that it is not only in the economically and socially marginal areas where Green Revolution approaches have made little headway that diversity-based crop development strategies have a future.

Conventional wisdom also has it that, alongside the pressures from intensive cultivation practices, market forces will lead farmers to favor a narrower range of cultivars as they gradually move away from subsistence production. But there is inconclusive evidence on this score, in particular, evidence that would help to distinguish where these forces originate. One study from Colombia (Janssen et al. 1991) suggests that consumer preferences in terms of bean types may be broader than those of merchants, and that it is the latter who are constraining choice further along the commercial chain. Merchants will, at least eventually, respond to demand for diversity when it is expressed by wealthier consumers, as is indicated by the growth of'organic'outiets in industrialized and some developing country urban centers. However, other strategies, through or around the market, may be required when, as in the Colombian case, the demand for diverse varieties originates from poorer consumers.

References

Document(s) 4 de 38