Part I: Introduction

Part I: Introduction

1 New Road Map Paul Van Mele

"The rigour of relevance is sustained by virtuous circles of social energy. People do it, and do it well, because they enjoy it and see a point in it." Chambers, 1997: 161

DOCUMENTATION When asked to develop a strategy for the writing of this book, late October 2003, it struck me how the philosophy of organisational learning had become an integral part of PETRRA's daily life. Ahmad Salahuddin asked me at the time whether I could already write one of the case studies, although we hadn't discussed this previously. It took some convincing from his part, but eventually we both saw the benefits of putting the framework we were developing into practice: making the first case study a study case. "That's the beauty of PETRRA, everything is in the air," said Salahuddin. Funded by the UK Department for International Development (DFID), The Poverty Elimination Through Rice Research Assistance or PETRRA project approved, managed and supported 45 sub-projects between 1999 and 2004. These had a respective focus on three broad areas: pro-poor policy (6), technologies (19), and uptake and extension (20). This last group was to become the subject of this book. In November 2003, we sent a first example of the Farmseed case study, given in Chapter 18, to all the sub-projects, along with guidelines for writers. Conveying the concept of doing extension method research, rather than extension per se was PETRRA's first challenge. But partners were also inexperienced in analysing, reflecting and documenting processes that underpinned their innovations. People had a strong tendency to only think about technology, not about the broader context and forces shaping it. I returned to Bangladesh and worked intensively with all project partners from January to September 2004 to help them articulate their experiences. Above all their 3

Innovations In Rural Extension

writing needed to stimulate reflection, since PETRRA emphasised that the learning from each sub-project had to be further institutionalised. Formal reports gave us insights in mainly quantitative impacts, but to evaluate processes and uncover the human and institutional dimensions of each sub-project we used a broad range of tools, including narratives. The concept was relatively simple, let people tell a story while reflecting on key points. "Stories help explain themselves; if you know how something happened, you begin to see why it happened." Fernández-Armesto, 2003 Institutional learning was further stimulated in workshops in which we used photographs, enterprise webs, actor linkage maps and various other innovation systems analysis tools (Biggs and Matsaert, 2004; Hall et al., 2003). We also held a one-day writers' workshop for the project leaders, who later on presented their draft case studies to a broad audience of service providers at a national uptake workshop, held at the premises of the Department of Agricultural Extension (DAE) in Dhaka in April 2004. Most cases highlight people's behaviour, values and commitment and is a deliberate attempt to bring alive in a distinctive personal manner the progress achieved by PETRRA.

LEARNING WITH PETRRA Through a competitive bidding process, PETRRA encouraged organisations to develop pro-poor production and marketing systems, and demand-driven advisory and research services through collaborative learning at the field level. Achieving impact is not the responsibility of a single agency, nor can extension be seen as a separate subset of the innovation system (see Box 1.1). It is in this broad context that we present extension. Private sector, the government research and extension system, along with non-governmental and community-based organisations experimented with new partnerships, organisational models and methods, which grew organically within the context of each organisation; none were forced upon them. Many innovations emerged from building on the organisations' own strengths and enabling cross-fertilisation between sub-projects. Each case study describes the origin Box 1.1 Innovations

4

An innovation is a new way of doing things by applying technical, methodological or organisational knowledge. This knowledge might be acquired through extension, media, research, experience or any other source. The new idea may come from several actors, including farmers, NGOs, public and private sectors. An innovation is a change in behaviour, even a small one. For example, each person who plants a new rice variety, or buys seed from a coop for the first time, is innovating.

New Road Map

of ideas and their evolution. Learning by doing requires time; this meant that successes in PETRRA did not come quickly or easily. But in the end, it ensured that the majority of methods were mainstreamed into the organisations that developed them. To mainstream methods, flexibility and ownership were key. PETRRA worked like a learning organisation, stimulating new thinking among all its members, both at management and sub-project levels, through the sharing of experiences. PETRRA linked underlying values of the learning organisation - empowerment of its members, rewards and structures fostering initiatives, and experimentation (Ayas, 1999; Stroud, 2003) - with values required to address poverty in rural development (see Box 1.2). Partners interested in experimenting with technologies and extension methods were identified competitively, a first step to ensure that sub-projects reflected these values. In early 2000, PETRRA established the uptake forum to stimulate communicative learning between its seed sub-projects. Participation of the poor during needs assessment, technology validation and dissemination Poor households with 3-8 months food security from own rice production, with some flexibility depending on region, actor and technology

Box 1.2 PETRRA's Values

Partnerships for better access to the poor and synergy of skills Gender issues addressed in all project phases Demand-led research based on stakeholder analysis (PETRRA, 2004)

In 2002, they launched another initiative to stimulate sub-projects to interact more and with other rice-related projects in their region, leading to the formation of two focal area forums in the Northwest and Northeast. In these, several national research institutes, DAE, non-governmental organisations (NGOs), and representatives of the private-sector and farmer communities started making joint decisions to improve sourcing, packaging and delivering of information, seed and services to the poor. The prominent involvement of grassroots organisations ensures feedback from the poor to the researchers and the agricultural inputs industry. The forums were endorsed by the state minister for agriculture, and consensus was reached among members on cost-sharing after the life of PETRRA. In early 2004, PETRRA worked with more than 12,000 farmers, of whom 40 percent were women, in more than 500 villages across 37 districts in Bangladesh (Magor and Salahuddin, 2004). During the project, the Bangladesh Agricultural University signed a ten-year memorandum of understanding with a major NGO to conduct action research with resource-poor farmers. Learning from PETRRA will also be carried forward in a new EC-funded project from 2004 to 2008. 5


PATHWAYS OUT OF POVERTY Rural Bangladesh is differentiated and dynamic, both in terms of economy and farming, across all her ecosystems. Wealthier households are generally moving out of agriculture, creating space for poorer households to rent land (Orr and Adolph, 2004). Agriculture is, therefore, most important for households with less land. Livelihood strategies are complex and there is no one pathway out of poverty. Often, farm households first try to establish their food security, which usually involves renting-in more land and investing in new rice technology (Orr and Adolph, 2004). Rice is Bangladesh's most important crop, grown over almost 75% of the cultivated land area. Improved rice technologies help families feed themselves. Evidence from PETRRA showed that significant impacts on livelihoods can be made in a very short time by introducing new, pro-poor rice technologies. This is because: Returns from land are higher than for micro-enterprise or wage labour, especially where rice can be grown in two seasons The payback period is short, because a cropping season lasts only three months Costs are low because technologies exploit under-utilised resources such as water, fallow land and household labour. A study of interventions for healthy rice seed further confirmed the importance of simple technology in improving the livelihoods of poor households, particularly with regard to better entitlement and greater freedom of choice (Bayes, 2004). The technologies developed, validated and fine-tuned under PETRRA were pro-poor and aimed at increasing productivity and profitability of rice farming. Those covered in the book include high-yielding varieties (HYV), health of farm-saved seed, disease management, seed drying and storage, and integrated crop and soil management. But also more complex challenges are addressed such as the pioneering of integrated rice-duck husbandry, developing an aromatic rice value chain, and a pro-poor market for mobile pumps. For each of these technologies, appropriate support mechanisms are needed to reach large numbers of farmers quickly. During the first two decades of the Green Revolution in the 1970s and 1980s, the increase in yield came mostly from gradual replacement of low-yielding, traditional varieties with high-yielding, modern ones. But due to a lack of appropriate delivery mechanisms farmers have been slow in replacing these early modern varieties. In 2000, for example, only 12% of the Bangladeshi farmers had received information on the more recent modern varieties from public-sector extension officials. Farmers gathered little information from input providers or NGOs. Over the past 15 years, diffusion of modern varieties has mainly taken place through informal farmer-tofarmer exchange (Hossain et al., 2003). As poorer households are more active users of modern rice technology (Siddiqui et al., 1990), they will benefit most from new extension methods that specifically target this group. 6

New Road Map

PETRRA was committed to improving the well-being of 'resource-poor' farm households, including women, by identifying, developing and validating improved technology and service delivery methods. Close interaction with rural communities was a must. The definition of PETRRA's target group emerged from the stakeholder analyses: 'Households with three to eight months net food security from own rice production and where more than half of the household income is derived from own farm production' (Orr, 2002). Any definition, however, only served as a working guideline. Households with only one to two months rice food security, and who had their major income from non-agricultural activities, also participated. Villagers have a very clear picture of the economic position of their own households. Well-being analysis or self classification by households was introduced to identify poor households. About one-half (49.8%) of the population in Bangladesh is still living below the poverty line, the great majority of them in rural areas, and with a preponderance of women (Duncan et al., 2002). The official national literacy rate is 50% for men and 41% for women (BBS, 2004), with averages being far lower for the poorest people. These figures reflect global trends in gender and poverty (see Box 1.3). Among the poor households in Bangladesh, women are getting increasingly involved in agriculture as their male partners often temporarily migrate for wage labour or to non-farm activities. With this in mind, reaching large numbers of farmers, particularly women, became a key criterion for PETRRA. An estimated 1.3 billion people in the world, of which 70% are women, fall below the international poverty line of US$ 1 per day (DFID, 2000). In most societies, women are more disadvantaged than men due to gender inequalities in employment, education, literacy, technical knowledge and access to land. In fact, two thirds of the world's 876 million illiterates are women, and the number is not expected to decrease significantly in the next twenty years (UN, 2000).

Box 1.3 Gender and Poverty

Poverty is more complex than a lack of income, it also involves a lack of assets, skills and opportunities, along with greater vulnerability and insecurity. For the targeted households, women's agricultural knowledge is becoming more important to help the family make wise decisions in farming. Orr and Adolph (2004) showed that a shared vision by the husband and wife was at the heart of successful graduation out of poverty. This book unveils the challenges and potential of working with poor farmers, men and women, not merely as producers, but also as customers, sellers, marketing agents and agricultural extensionists. The case studies reveal processes and show that working holistically with the poor, as partners, is the only way to build pathways out of poverty. 7


CHANGING DIRECTIONS Diversification of service providers and innovations in extension are needed, not only in Bangladesh but across the world (Chowdhury and Gilbert, 1996; Rivera and Zijp, 2002). Continued economic liberalisation is likely to result in a growing number and greater diversity in service providers. While public sector funding is decreasing and recent discussions focus on cost-recovery and public-private partnerships (Rivera and Zijp, 2002; Anderson and Feder, 2003; Ramírez and Quarry, 2004), there is still a keen need to develop locally-adapted extension and farmer education methods that address the poor, especially women (Berdegué and Escobar, 2001; Jiggins et al., 1997; Kanji, 2003). Since the 1980s, farmer-centred education approaches such as farmer field schools (FFS) have blown a fresh breeze over the extension landscape, highlighting the need to train farmers in their own fields through experiential learning rather than through prescriptive skills development (Röling and Wagemakers, 1998). But to reach the millions of farmers, more innovations are needed. With this book, we present some broad principles for a new road map by illustrating the innovative complementarities that can be built between farmer education, farmer organisational development, extension and communication, and pro-poor business development.

STRUCTURE OF THE BOOK Innovations in Rural Extension presents the evolution of methods, outlines them in contextual detail, and presents the reader with the keys to success and some of the difficulties he or she may encounter while implementing them. The book has six parts, of which this short introduction is Part I. The next four parts are each introduced by an overview chapter, followed by a number of empirical cases. The last part summarises findings and offers some fresh perspectives. Thelma Paris and colleagues from PETRRA launch Part II on gender in agricultural extension. Poor women are eager to learn about all aspects in agriculture, not just on the roles they traditionally fulfil, and have proven to be innovators and committed extension agents. Cases are presented on women-led group extension, the family approach whereby husbands and wives receive training together, and the production of videos whereby women's knowledge and skills are presented alongside scientific information. The latter project won a prestigious international award for effective communication. Part III on learning with rural communities is introduced by Jeffery Bentley and Paul Van Mele. They discuss the specific role farmer-to-farmer extension has in learning about local knowledge and developing mass media communication messages. A first case 'Watch and Learn' shows that video, made by a multidisciplinary team and 8

New Road Map

involving peers, has a higher impact on rural women's seed health practices compared with farmer-to-farmer extension. The next case 'Village Soil Fertility Maps' gives impressive evidence of how within a short time and with limited financial resources, soil fertility management was improved in more than 200 villages by combining principles of soil fertility mapping, participatory research and farmer-tofarmer extension. The last two cases in this part build on old forms of reaching rural audiences, namely going to local markets or other public places and using folk songs. They have been revived into new agricultural extension methods called Going Public and picture songs, the latter bringing entertainment-education to rural women. Part IV on enterprise webs is introduced by Noel Magor, who stresses that complex linkages are essential for poor farmers to adopt certain technologies. It places extension in a broader context and looks at rural development from a business perspective. Tools such as enterprise webs can help organisations to analyse the weakest links of an enterprise and to make decisions about where vertical integration would be better than strategic partnerships or vice versa. Cases are presented on establishing integrated rice-duck farming, pro-poor markets for mobile pumps and a value chain for aromatic rice. In Part V, Solveig Danielsen, M. K. Bashar and Mark Holderness investigate the emergence of pro-poor rice seed systems in Bangladesh. The first two cases illustrate significant changes in mindset among senior scientists of the Bangladesh Rice Research Institute in validating technologies with NGOs and poor farmers, and in channelling multiple partnerships into a rice seed network. The last three cases shed light on the experiences from the three diverse NGOs in training farmers as seed producers and involving them in extension and marketing efforts. Part VI puts all experiences in a broader context, draws on transaction cost theory, and targets donors, policy makers and academics. The bulk of the book, however, merges quantitative impact assessments with more qualitative process analysis, and is written in a simple style to offer 'relaxed reading' for development workers, service providers and university students. We want others to enjoy reading what we have enjoyed doing.

REFERENCES Anderson, J. R. and Feder, G. (2003) Rural extension services. World Bank Policy Research Working Paper 2976. Ayas, K. (1999) Project design for learning and innovation: Lessons learned from action research in an aircraft manufacturing company. In: Easterby-Smith, M., Burgoyne, J. and Araujo, L. (eds) Organizational Learning and the Learning Organization: Developments in Theory and Practice. Sage Publications, London, pp. 176-193. Bayes, A. (2004) Technology, entitlements and freedom of choice. PETRRA Policy Brief 3. IRRI Dhaka, Bangladesh. Available at http:// www.petrra-irri.org 9


BBS (2004) Statistical Pocketbook of Bangladesh 2002. Bangladesh Bureau of Statistics (BBS), Ministry of Planning, Dhaka. Berdegué, J. A. and Escobar, G. (2001) Agricultural knowledge and information systems and poverty reduction. World Bank AKIS Discussion Paper. Biggs, S. D. and Matsaert, S. H. (2004) Strengthening poverty reduction programmes using an actor-oriented approach: examples from natural resources innovation systems. AgREN Network Paper 134. Available at http://www.odi.org.uk/agren/ Chambers, R. (1997) Whose Reality Counts? Putting the First Last. Intermediate Technology Publications, London. Chowdhury, M. K. and Gilbert, E. H. (1996) Reforming agricultural extension in Bangladesh: Blending greater participation and sustainability with institutional strengthening. AgREN Network Paper 61. Available at http://www.odi.org.uk/agren/ DFID (2000) Poverty Elimination and the Empowerment of Women. Department for International Development, London. Duncan, A., Sharif, I., Landell-Mills, P., Hulme, D. and Jayanta Roy (2002) Bangladesh: Supporting the Drivers of Pro-Poor Change. DFID, London. Available at http://www.lcgbangladesh.org Fernández-Armesto, F. (2003) The Americas: the history of a hemisphere. Weidenfeld & Nicolson, London. Hall, A. J., Yoganand, B., Sulaiman, R. V. and Clark, N. G. (eds) (2003) Post-Harvest Innovations in Innovation: Reflections on Partnership and Learning. DFID Crop Post-Harvest Programme, South Asia and NR International Limited, Kent, pp. 1-15. Hossain, M., Lewis, D., Bose Manik Lal and Chowdhury Alamgir (2003) Rice research, technological progress, and impact on the poor: the Bangladesh case (Summary Report). EPTD Discussion Paper International Food Policy Research Institute No. 110. Jiggins, J., Samanta, R. K. and Olawoye, J. E. (1997) Improving women farmers' access to extension services. In: Swanson, B. E., Bentz, R. P. and Sofranko, A. J. (eds) Improving Agricultural Extension. A Reference Manual. FAO, Rome. Kanji, N. (2003) Mind the Gap: Mainstreaming Gender and Participation in Development. IIED, London. Magor, N. P. and Salahuddin, A. (2004) Reaching resource-poor farmers, men and women through improved uptake methods and pathways. Proceedings of National Uptake Workshop. Department of Agriculture Extension, Dhaka held April 17-18, 2004. IRRI, Dhaka, Bangladesh. Orr, A. (2002) Project Strategy. PETRRA, IRRI Dhaka, Bangladesh. Available at http://www.petrra-irri.org

10

New Road Map

Orr, A. and Adolph, B. (2004) Pathways from poverty: Household-level processes of graduation. PETRRA Policy Brief 2. IRRI Dhaka, Bangladesh. Available at http://www.petrra-irri.org Ramírez, R. and Quarry, W. (2004) Communication strategies in the age of decentralisation and privatisation of rural services: lessons from two African experiences. AgREN Network Paper 136. Available at http://www.odi.org.uk/agren/ Rivera, W. M. and Zijp, W. (eds) (2002) Contracting for Agricultural Extension. International Case Studies and Emerging Practices. CAB International, Wallingford. Röling, N. G. and Wagemakers, M. A. E. (eds) (1998) Facilitating Sustainable Agriculture: Participatory Learning and Adaptive Management in Times of Environmental Uncertainty. Cambridge University Press, Cambridge. Siddiqui, M. R., Islam, R., Khan, A. H., Magor, N. P. and Ahmed, N. U. (1990) Towards equity: impact of modern rice technology in a rainfed lowland environment. In: Rhoades, R. E., Sandoval, V. N. and Bagalanon, C. P. (eds) Best Paper Awards 1990. CIP/UPWARD, pp. 53-80. Stroud, A. (2003) Transforming institutions to achieve innovation in research and development. In: Pound, B., Snapp, S., McDougall, C. and Braun, A. (eds) Managing Natural Resources for Sustainable Livelihoods: Uniting Science and Participation. Earthscan Publications, London, pp. 88-112. United Nations (2000) The World's Women 2000: Trends and Statistics. Available at http://unstats.un.org/unsd/

11

Part II: Gender

2 The Right to Learn

Women Want More Agricultural Advice

Thelma Paris Shaila Arifa Nabi Ahmad Salahuddin Noel P. Magor

INTRODUCTION Millions of poor farm families in Bangladesh grow rice as a way of life. Every adult household member, male and female, plays a vital role in seeing that the rice harvested from their small holding is enough to feed the family. Most of the women who work in field agriculture are from households with very small holdings (200 to 1,600 square meters), followed by women in landless, small and medium size farms (Safilios-Rothschild and Mahmud, 1989). Women also work more in the field in the tribal areas of Bangladesh (Sattar et al., 2004). Poor women grow vegetables on their farms, manage livestock, and work as wage labourers on other farms. In addition, in the present transitional rural society, temporary migration of men is common with women becoming de facto heads of households who make all agricultural decisions, hire labour, sell crops and control -at least partially- the farm income. It is more common for smallholder women to head households and make farming decisions than official statistics suggest (Safilios-Rothschild and Mahmud, 1989). Various non-governmental organisations (NGOs) empowered poor rural women by forming groups and lending them microcredit to generate income, but rice research and extension institutions have tended to exclude women. This has not been deliberate, but more an omission of not seeing. Women have tended to be included more frequently in topics recognised as women's work, like seed storage or crop processing. The idea that women also benefit from learning about field crop agriculture has been under-explored. This lack is associated with the widespread belief in Bangladesh that women do not do field agricultural work because of deeply embedded norms of patriarchy (a set of social relations with a material basis that enable men to dominate women) and purdah that prescribes seclusion and restricts women's movements outside their bari or homestead. We purposefully wrote this chapter broader than 'women and extension', because it 15


is one's attitude to women in agriculture that influences how we include them. We trace experiences from the DFID-funded Poverty Elimination Through Rice Research Assistance project (PETRRA) from 1999 to 2004 in engaging with women. It began with integrating women into identification of research issues, and then learning how to draw women more into training and dissemination and in this to appreciate their feedback. Including women, leading to a response by women themselves, has been the most interesting lesson for project staff and may pose the most significant shift for the future. Providing women with all sorts of information opens up the road to development.

INCORPORATING GENDER CONCERNS From the beginning, PETRRA's main philosophy was to put poor people first, not the technology. Over time, a value-based approach emerged and evolved along with the project that emphasised joint learning through action and reflection. Gender equity was a guiding principle for PETRRA. A gender strategy was developed that encompassed clients and research management. The gender strategy for PETRRA's sub-projects included guidelines for incorporating gender concerns in each stage of the project cycle, namely: needs assessment and constraints analysis; identification of technology options and opportunities to solve constraints; enhancing the capacity of various stakeholders (farmers, NGOs and government extension agents); farmer participatory on-farm experiments; monitoring and evaluation; testing innovative extension approaches; establishing support mechanisms to sustain adoption of innovations; and impact assessment (Paris, 2001). However, was this to be a case of including women 'because a project demanded it'? Needs assessment and constraints analysis To identify poor farmers' needs and constraints to increasing rice productivity, a participatory rural appraisal (PRA) approach was required. The PRA Promoters' Society of Bangladesh together with scientists of the Bangladesh Rice Research Institute (BRRI) and local NGO personnel facilitated stakeholder analyses in major rice growing regions in the country. After the first round, where mainly men interviewed men, a more gender balanced team of facilitators started to hold separate women's meetings, besides those for all community members. For example, the Integrated Crop Management sub-project gathered information on gender division of labour in rice production as well as on farm and household decision-making.

16

The Right to Learn

Call for concept notes and the PETRRA approach to gender The constraints analyses led to a call for concept notes and proposals to identify or develop technology for rice production and post-harvest and diversification beyond rice. Access to knowledge for men and women was a priority. The proforma had a specific section on including women. On reflection, we realised that the early project requirements were too general and too easily satisfied reviewers. In a later call we included the specific issue of women-to-women extension and the assessment form was more probing in assessing engagement with women. For example, we gave bonus points if the lead person was a woman. PETRRA provided training on gender and working with women in the village. It encouraged including women. Involving more women was a prerequisite for beginning to effectively work with them. By incorporating gender concerns into its sub-projects, PETRRA not only helped improve women's welfare, but also made the research programmes more relevant. For example, researchers had the idea for a rice seed drying table, but women and their husbands came up with a range of designs of inexpensive tables that also met other needs of the household, such as threshing rice or storing kitchen pots (Chapter 7, this volume; Van Mele and Zakaria, 2004). There were no early champions. The Seed Health Improvement sub-project (SHIP) focused on seed; even though preserving seed is a women's activity, only one of the 28 field researchers was a woman. In a SHIP workshop for farmers at the end of year one, no women clients attended. At that meeting some male farmers asked for women to be included in training. But other village men disagreed, even after the second year of training (see Box 2.1). Sometimes gender and community participation clash (Guijt and Shah, 1998).

To ensure that only poor women participated in the research subprojects, communities identified their own poverty criteria and classified households through well-being analysis.

The quarterly monitoring report required by PETRRA included a section on women, asking more or less, 'What work have you done with women in the last quarter?' Apparently sub-projects provided post-harvest training for women even if their focus was soil fertility management. Women were being included, but on the edge. The NGO Shushilan (see Chapter 10) provided a useful insight. Although only working with a few farmers, half their demonstration plots were managed by women. Although they previously did not work in agriculture but did focus on women, Shushilan took the initiative under PETRRA to give women training in rice growing, not just post-harvest.

17


Box 2.1 The Woman and the Rose

by Paul Van Mele It was May 2, 2001, my first field visit in Bangladesh. When we arrived in Sreepur, separate groups of men and women farmers were manually sorting seed. Under the Seed Health Improvement sub-project (SHIP), male farmers were trained for three seasons to check their seed grain-by-grain and remove the ones with spots and other irregularities. It took two years before the project put gender into practice. Despite the fact that women do the post-harvest work, this was the very first season researchers trained them on seed health. Because of the rain, the 30 men and 30 women had gathered in a rice mill. There was no electricity and I wondered how on earth they could possibly see the tiny spots on the seed in this darkness. One of the social scientists rounded the people up to say that the session would be cancelled, because of the rain. It struck me that she was only talking to the men; the women all sat quietly in the back, some hiding behind a corner of their saris, others looking after their children and babies. I urged the team to ask the women's opinion on training. After all, wasn't that what the project was supposed to do? After a while, one woman with heavy framed glasses spoke up: "We want training in separate groups." Some men agreed that their wives would benefit from training on seed health, but only in groups and jointly with the men. Then an older man fiercely raised his voice while pointing to the women, "You are not allowed to get any training!" I was shocked. "What is going on here?" I wondered. A heated discussion lasted for about 15 minutes, until the lunch packets arrived. The meeting dissolved; some ate on the spot, while others took their food home. When we sat under a tree, eating our chicken, the woman with the glasses suddenly appeared in front of me saying, "Please make sure that we get training." While she talked, she held my hand and gave me a rose.

A message began to emerge: 'Include women in all your activities, not just the ones in which they work directly.' This message became a guiding principle for PETRRA about mid way through the project. Enhancing men and women's knowledge and skills PETRRA sub-projects trained village women: (i) on technologies in which they are actively engaged such as post-harvest; and (ii) on overall rice production so that they can contribute more in decision-making. From 2000 to 2003, women in extension activities of PETRRA sub-projects increased from 10 to 41 percent. PETRRA reached male and female farmers through hands-on training, farmer participatory on-farm experiments, demo plots and farmer cross-site visits. All activities embraced the principles of learning by doing and 'seeing is believing'. The NGO Sustainable Agriculture and Farming Enterprise (SAFE) trained farmers in farmer field schools (FFS) to use herbicides for weed control safely. At first, men objected to women attending, but after discussions in small groups and then in large meetings, men agreed to include women as partners in learning about growing rice. 18

The Right to Learn

Seventy eight women joined the ten farmer field schools. When women spoke up in the sessions, men realised how much women knew about rice farming, and that women play an important role in herbicide safety and in household decision-making. Later on, the men even asked the project to invite the women to PETRRA's communication fair in Dhaka. Women said that attending the farmer field schools improved their knowledge of rice growing, and therefore their ability to contribute to family decisions. They also learnt how to wash the sprayer, away from ponds, and to bury empty herbicide containers. Chemical weed control saved time for more profitable activities. Since women no longer had to cook meals for hired labourers, they had more time to spend tending vegetable gardens, which women are allowed to do because vegetable plots are close to the homestead and not considered 'public'. Women said that they had more time to tend goats, chickens, cows and fish to supplement their family income. Testing and validating technologies with farmers Promising technologies, whether developed by scientists or farmers, were tested and validated by farmers on their own fields. In all parts of Bangladesh, drying rice seed in the rainy season is a new need, brought about by the introduction of irrigation pumps and modern varieties. Rice grown in the dry or boro season is harvested at the onset of the rains. So in early 2002, CABI Bioscience started several participatory activities in the SHIP sub-project. Multipurpose seed drying tables emerged as a pro-poor technology, developed by women in Maria village, Bogra (Van Mele and Zakaria, 2004). The technology rapidly spread to other areas. Through the Rural Development Academy (RDA) in Bogra, with help of the local government, 1,800 out of 4,800 households of Amrool union made drying tables, most of which also served for threshing rice. Large national NGOs such as the Bangladesh Rural Advancement Committee (BRAC) and Proshika included it in their programmes, and when the Agricultural Advisory Society (AAS) introduced it in their women-led group extension programme, more than 60% adopted the technology (see Chapter 3). The drying tables also featured in one of the videos on seed health (see Chapter 5), which were disseminated to multiple NGOs, governmental organisations and TV stations.

Children prepare a mud meal. Playing is one of the earliest and pleasantest styles of learning by doing.

But also local skills were validated scientifically. BRRI scientists learnt through a carefully designed trial that women accurately assessed seed moisture by cracking rice grains

19


between their teeth. This trial was developed to open scientists' eyes to the value of local skills, and to increase women's self esteem. The men watching the outdoor event from the sidelines shared in the climax when results were revealed. In another sub-project, for the first time ever in Bangladesh, women alongside their husbands rated unreleased varieties prior to harvest in a 'mother trial', established by scientists in a village. Scientists learnt that women prefer coarse grain for eating and fine grain for sale, while men are mostly interested in yield and plant type. Women also received seed of their selected varieties which they evaluated in 'baby trials' in their own field, using treatments they expressed a particular interest in (Gregorio et al., 2004; see also Paris, 2002; Snapp and Heong, 2003; Paris et al., 2004b). Testing different types of trainers and groups

CABI Bioscience helped the Seed Health Improvement subproject validate local skills. Women were invited to rate three batches of seed with different moisture content. Results of the voting test came mainly as a surprise to local scientists. "We are confident," laughed the women, "we were born in the rice."

To speed adoption of technologies that will directly benefit poor rural women, PETRRA encouraged partnerships with NGOs that were already working with women's groups. PETRRA tested several approaches to rapidly disseminate farmervalidated technologies and ensure that they respond to women's needs - not only to men (Table 2.1).

Women-led group extension All the extensionists and clients of this approach are women. It was tested to determine whether local women can be effective extension agents and under what conditions. The NGOs that successfully tested this approach were AAS, Rangpur Dinajpur Rural Service (RDRS) and Shushilan. Chapter 3 describes how AAS successfully disseminated post-harvest technologies to poor female farmers after forming and developing the leadership of 26 womenled groups in Kishoreganj and Habiganj districts. Initially, female staff members of 10 local partner organisations, covering 30 villages, were trained as trainers, but soon trained village women volunteered to spread the word to other villages. Both the low cost rice seed drying tables and use of plastic drum with naphthalene for rice seed storage are now being widely used by women of the project's and neighbouring communities. RDRS in collaboration with BRRI and a local NGO, Janani, established nine women-led extension groups to spread technologies to poor women on rice-potato-rice cropping. The recovery percentage of credit was higher for women-led groups than for non-women-led and maleoriented extension. The project strengthened the decisionmaking ability of women farmers, and the institutional

20

The Right to Learn

capacity of local organisations to access good quality services, and be responsive and accountable to the delivery of services.

Family approach in agricultural extension The Department of Agricultural Extension (DAE) tends to include mainly male farmers in their training activities and to deal only with new methods in rice production. Women's training focuses only on homestead activities, if they receive any training at all. However, crop management decisions affect the whole family and not only the male head of the household who is often referred to as the only farmer in the household. Table 2.1 PETRRA's research on extension: different types of trainers and groups EXTENSION APPROACH

WOMEN'S ROLE IN DISSEMINATING INNOVATIONS

OUTCOMES

Women-led group extension on seed drying and storage (AAS; Chapter 3)

Women promoted seed drying tables using pictures, village fairs and folk songs

Technologies were widely adopted and further promoted by village women extension agents, beyond the project area

Women-led group extension on ricepotato-rice cropping technologies (RDRS)

More than 90% of the participating women shared their experience from training and cross-site visits with other poor women

Recovery of credit is higher from women-led groups than from others

Family approach to extension on rice and seed production (AAS; Chapter 4)

Women trainers taught women-only or mixed groups; women shared learning more readily with the wider community than men did

Training husband and wife together improved decision-making, adoption of technologies and rice provisioning ability

Farmer-to-farmer extension on seed health improvement (SHIP)

Women and men trainers taught other farmers how to improve the quality of farm-saved seeds

Seed management improved from the field until post-harvest, resulting in 10-12% yield increase

Chapter 4 describes how trained female staff of partner organisations with the help of a female agronomist from AAS taught different kinds of groups of poor farmers: single sex groups, husband and wives, or whole families (including children). The training contents, materials and schedule were based on discussions with selected farmers. The main lesson of this project is that including women in training enhances learning within the family and wider community, for all topics. In the SHIP sub-project, both husbands and wives were invited to participate in training and extension. Focus group discussions revealed that the participation of 21


husbands and wives led to better gender relations after training. For example, they started sharing their opinions on different crop management aspects. Another sub-project used the family approach for training farmers about more efficient application of nitrogen fertiliser by making use of a visual tool, called the leaf colour chart (see photo). Trained women kept their leaf colour chart in a safe place and reminded their husbands to apply their newly acquired knowledge.

According to the NGO RDRS, in Northwest Bangladesh, women were more sincere in participating in extension and were more dedicated than men to follow the advice from the groups' weekly learning sessions.

Women shared their knowledge more readily with their neighbours. By applying less nitrogen fertiliser at more appropriate times, farmers said they were now able to buy a sari for their wives. Based on insights from these projects, other NGOs decided to use the family approach for up scaling technologies for horticultural crops, and integrated farming systems such as rice-fish and rice-duck. Testing learning tools with women Several learning tools were developed and tested to spread information to women: videos, photographs, and cultural shows (Table 2.2). While developing these tools, women's opinions were elicited and considered.

Colour photographs In the women-to-women extension approach described in Chapter 3, small groups of women were given A4-size colour photos, showing different drying scenes. Sessions were built around the photos, on the principles of adult education. Learning the basic ideas of seed drying triggered the participants' creative thinking.

Educational videos by women for women Chapter 5 discusses how four short training videos on caring for healthy seed were 22

The Right to Learn

developed. Village women without any previous training were invited to review a first rough edit of the videos. They made major improvements in the script, and asked to see all four programmes linked together. To ensure relevance to farmers, and scientific soundness, the team learnt to identify and involve both local innovators and technical experts early in the script writing.

Educational entertainment Chapter 10, in Part III on learning with rural communities, describes song-anddance shows to motivate social and technical change. Despite the advances in electronic communication, live shows are again becoming popular. Shushilan added educational themes to neo-traditional musical shows to share ideas about appropriate rice technology. The Shushilan troupe sings and dances in front of a painted, scroll backdrop to promote modern varieties, improved chemical and organic fertilisers, line transplanting, seed health and natural enemies of insect pests. They performed 24 shows in different areas of Southwest Bangladesh. More than 25,000 people, half of them women, attended the shows. These shows were fun and educational for women who have limited access to cinemas and little opportunity to attend conventional extension programmes. Noticing Shushilan's success, other PETRRA sub-projects got inspired. The women-led extension project by AAS held a village fair to demonstrate rice seed drying and storage, around which theme a local troupe made a song and performed (Chapter 3). What women learnt Women who participated in these projects said that they benefited from the training not only in terms of production and income, but also in recognition, saving money, making sound decisions, managing crops better, among other benefits.

The leaf colour chart helps farmers to make better decisions on amount and timing of nitrogen fertiliser, resulting in significant savings and better yields.

Recognition "Before only NGOs talked to us and loaned us money. But now for the first time scientists came to us, asked questions and listened to us. We did not realise that we have knowledge to share. We also learnt that how we manage seeds affects the rice yields, and that how rice is managed in the fields will affect the quality of the seeds we keep. Thus we realised how important our roles are if we practice what we learnt from the training." 23


Saving and earning money "Now we know more about fertiliser management particularly the proper time and amount of application so that we don't waste money. We also want to know how to manage pests without the use of too much chemicals. We have heard of IPM (integrated pest management), but we don't really know what this means." "By producing quality seeds, we can now sell good seedlings and earn money. We use our money for buying good clothes for our children, who are now motivated to go to school. We also use our money for buying mustard oil rather than soybean oil, bath soap, sandals, and also goats."

Making sound decisions "It has been our custom for women not to go to the field. But it will be useful for us to know how much and when to apply inputs so that when our husbands are away, we can also make sound decisions when we supervise the labourers we hire." "Now we know that kopal (fate) cannot control our production. It is the knowledge we gained and how we use this knowledge which makes the production good or bad."

Managing crops better "Before we were not allowed to go to the fields and did not have any knowledge about managing the rice plants in the fields. Now we know that we can play an important role by caring for the crops in the fields, by visiting them once in a while to see whether there are insects or diseases. Taking care of rice is just like taking Table 2.2 PETRRA's research on extension: learning tools for poor women

24

LEARNING TOOLS

WOMEN'S ROLE IN DISSEMINATING INNOVATIONS

OUTCOMES

Photographs on seed drying (AAS; Chapter 3)

Female project staff and later also village women extension agents used laminated photographs to support their group discussions

The visual support gave women confidence that the technology was easy; more than 60% of women made their own drying table

Videos on post-harvest (CABI and RDA; Chapters 5 and 7)

Village women provided inputs in the script research, development and refining the videos on rice postharvest innovations

There is high demand for the videos; video offered a way for village women to spread their learning beyond their village

Cultural shows on rice and seed production (Shushilan; Chapter

Women used popular mass culture (folk songs, drama and dance) to carry agricultural messages; women acted as demo farmers and extension agents

The NGO can hardly keep up with demand to perform shows with various messages; women became successful demo farmers

The Right to Learn

care of our children, whom we nurture from infancy until they grow up."

Better social status "The community used to undermine us. Now we have proven that we are in no way inferior, but rather superior to our male counterparts in the field and in the household. Even if we do not go to the field, we advise our husbands how to prepare the fields, sow the seeds, or transplant the seedlings better."

Breaking down cultural barriers "Before the men and our neighbours used to ridicule us when we worked in the fields. Now they have accepted this."

MAJOR LESSONS FROM WORKING WITH WOMEN 1) There must first be a commitment to include women. Partners need space to gain confidence in including women. 2) Partnerships can be an important means of increasing access to women. For example, TMSS is a women's NGO that formed a partnership with RDA and CABI for developing educational videos. The link has opened the eyes of TMSS to opportunities in agriculture for their women clients and also provided an entry point for RDA to work with women. Government extension agents also gained more confidence in working with women after partnering with NGOs and communitybased organisations. 3) Women should be given access to all agricultural information, not just on postharvest, even if they do not work in the field. Their knowledge and skills are vital in making sound decisions on growing rice and postharvest, particularly when wives become de facto heads of households after husbands migrate to jobs in the cities or abroad.

Men still like to casually slip in during women's group training sessions.

4) Learning by doing (through participatory on-farm experiments, hands-on training and managing their own demonstration plots) is important for disseminating technical knowhow to poor women with low levels of formal education and social restrictions on their mobility. Women demo farmers can overcome these barriers after proving that they can learn improved technologies and that they know as much as male farmers about field agriculture. 25


5) Women who are trained as local extension agents should have the credibility to disseminate the technologies to other women, either by testing the innovation themselves or by demonstrating its effectiveness. Potential adopters are aided in evaluating a new idea if they are able to observe it in use under conditions similar to their own. They should have the ability to work well with people and be willing to take training, train others and share information. 6) Innovative women farmers make excellent extension agents. For example, with capacity building and limited financial incentives from the NGO AAS, female farmer extension agents confidently developed new women's groups in neighbouring villages. 7) There is no single prescribed extension method to reach all poor women. However, in early stages, women can express themselves better if they are with a group of women only. Village meetings may start with mixed groups, followed by single sex groups to keep men from dominating the floor. More women have access to training if it is village-based rather than residential. Agricultural topics traditionally geared towards men only, should be given to husbands and wives jointly (family approach in extension). 8) Biological scientists and agronomists should listen to women's opinions, assess attitudes towards certain practices and consider their criteria in the design and dissemination of rice technologies. For example, women's criteria for a seed drying table are: simple, low cost, made from local materials, easily transported, and with multiple uses. Also extensionists can learn from women: women preferred photos to written messages and liked posters to be placed in their house or the village health centre. The video scripts for women-to-women extension were developed and refined based on rural women's inputs. 9) Learning methods and tools such as educational entertainment, videos and photographs are relatively cheap and can add value to face-to-face extension in disseminating knowledge-intensive technologies. 10) The dissemination and adoption of certain technologies may require sustained support. For example, Shushilan developed a mechanism to supply seed and inputs among the farmers per their need, and opened an agricultural information centre, managed by a woman. Women demo farmers need continued access to quality seeds for their demonstration plots.

CONCLUSION PETRRA sub-projects promoted gender equality by going beyond a concern to 'include women' by training women on rice farming, not only on post-harvest. Women-to-women and family approaches reach poor women effectively. Videos, photos and musical shows spark the audiences' interest and facilitate learning. The following case studies show that poor women, despite social barriers and low levels 26

The Right to Learn

of formal schooling can be effective change agents if given equal opportunities as men in getting training and access to support services and resources. In a policy dialogue held in Dhaka, Bangladesh, several gender-responsive policies were suggested, one of which was to channel agricultural credit through women rather than men (Paris et al., 2004a; Hossain et al., 2004). In 1987, Robert Chambers and Janice Jiggins outlined some hopeful ideas for participatory research, citing now familiar ideas such as scientists were not working with farmers; research was reductionist and could not easily handle complex interactions of resource-poor farming. They concluded that there were some worthy efforts, but no one was really doing research with farmers (Chambers and Jiggins, 1987a, 1987b). Ten years later, Jiggins and colleagues made specific recommendations to improve women's access to extension (Jiggins et al., 1997), the challenges remaining high. As this book shows, there are now mainstream agricultural R&D projects that have new methods for interacting with resource-poor women to develop and spread appropriate technology.

REFERENCES Chambers, R. and Jiggins, J. (1987a) Agricultural research for resource-poor farmers Part I: Transfer-of-technology and farming systems research. Agricultural Administration and Extension 27, 35-52. Chambers, R. and Jiggins, J. (1987b) Agricultural research for resource-poor farmers Part II: A parsimonious paradigm. Agricultural Administration and Extension 27, 102-128. Gregorio, G. B., Salam, M. A., Karim, N. H. and Seraj, Z. I. (2004) Final Evaluation Report of Sub-Project on Development of High-Yielding Varieties for Coastal Wetlands of Bangladesh. IRRI, Dhaka, Bangladesh. Guijt, I. and Shah, M. K. (1998) The Myth of Community: Gender Issues in Participatory Development. Intermediate Technology Publications, London. Hossain, M., Paris, T. R., Bose, M. L. and Chowdhury, A. (2004) Nature and impact of women's participation in economic activities in rural Bangladesh. CPD-IRRI Policy Brief 7. IRRI, Dhaka, Bangladesh. Jiggins, J., Samanta, R. K. and Olawoye, J. E. (1997) Improving women farmers' access to extension services. In: Swanson, B. E., Bentz, R. P. and Sofranko, A. J. (eds) Improving Agricultural Extension. A Reference Manual. FAO, Rome. Paris, T. (2001) Gender Strategy for the Poverty Elimination Through Rice Research Assistance (PETTRA). IRRI/BRRI/DFID, Dhaka, Bangladesh. Paris, T., Singh, R. K., Atlin, G. and NARS from India (2002) Farmer participatory breeding and participatory varietal selection in eastern India: Lessons learned. In: Whitcombe, J. and Atlin, G. (eds) Workshop on Breeding Rainfed Rice for Drought-Prone Environments: Integrating Conventional and Participatory Plant Breeding in South and Southeast Asia. IRRI, Los Baños, Philippines, April 2002. 27


Paris, T., Chowdhury, A. and Bose, M. L. (2004a) Changing women's roles in homestead management: mainstreaming women in rural development. Presented at the Policy Dialogue on Women's Contribution to Rural Economic Activities: Making the Invisible Visible. April 22, 2004. BRAC Centre Inn, Dhaka, Bangladesh. Paris, T., Singh, A., Singh, H. N., Luis, J. and Hossain, M. (2004b) Integrating a gender dimension on participatory research and development: a case experience in eastern India. In: Participatory Research and Development. Vol. 3A. CIP/UPWARD Philippines, in press. Safilios-Rothschild, C. and Mahmud, S. (1989) Women's Roles in Agriculture: Present Trends and Potential for Growth. United Nations Development Programme and UNIFEM, New York. Sattar, M. A., Biswas, J. C., Maniruzzaman, Md., Wahed Sarker, G., Neogi, M. G. and Ramjan Ali, M. (2004) Completion Report on Integrated Crop Management (ICM) in NorthWest Region of Bangladesh. IRRI, Dhaka, Bangladesh. Snapp, S. and Heong, K. L. (2003) Scaling up and out. In: Pound, B., Snapp, S., McDougall, C. and Braun, A. (eds) Managing Natural Resources for Sustainable Livelihoods: Uniting Science and Participation. Earthscan Publications Ltd, London, pp. 67-87. Van Mele, P. and Zakaria, A. K. M. (2004) From concept to impact: developing and communicating multipurpose seed drying tables in Bangladesh. In: Participatory Research and Development. Vol. 3 A. CIP/UPWARD Philippines, in press.

28

3 Breaking Down Barriers Village women spread the word

Helen Latifun Nessa Paul Van Mele

SUMMARY Women deserve access to affordable, appropriate and environmentally friendly technologies for drying and storing rice seed. To reach more women, the national Agricultural Advisory Society (AAS) trained women trainers of ten local organisations, covering 30 villages in two districts in Northeast Bangladesh. AAS taught them on organisational, communicational and technical aspects, after which they established groups of poor women. Organising the training sessions in the house of one of its members made participation easier and created a relaxed learning environment. In group discussions, women trainers used a set of photographs depicting real-life situations to stimulate creativity, which resulted in a range of multi-purpose drying tables made of local materials. The seed drying tables brought poor women together around a common issue and gave them a powerful experience, enhancing their self-esteem. For seed storage, AAS bought plastic drums at the nearest city market and sold them in the rural areas at wholesale price. With the improved technologies and training, yields improved within a season. Moreover, not only did women gain control in deciding how the extra money would be spend, the additional income also bought them respect. Early project successes spread first to other female family members of those women trained, followed by other group members and the wider community. After one season some trained women became effective extensionists themselves, each visiting three to five neighbouring villages and mobilising new groups. When given the opportunity and with limited financial support, poor women trainers were highly committed to alleviate poverty and displayed a high level of solidarity with poor women in other communities. 29


ACTORS AND NETWORKS "We do everything, particularly in processing and storing rice and rice seed, but get no recognition. Teach us something for which family and the society remember us," an appeal of one of the women during a focus group discussion conducted by AAS. AAS is a national non-government organisation (NGO) promoting appropriate agricultural technologies for the poor. Over the years, it has developed a network of more than 150 NGOs and community-based organisations (CBOs). By 2004, AAS employed 15-20 staff members who had trained about 7,500 and motivated another 15,000 farmers under various sub-projects of the DFID-funded Poverty Elimination Through Rice Research Assistance project (PETRRA). In the process, AAS established a good working relationship with national and international agricultural research and extension institutions. In developing and testing a women-led group extension method, AAS identified two local NGOs and eight community-based organisations in ten villages. These partners played a significant role, especially in selecting poor women farmers, forming groups, organising training courses, monitoring of progress and problems, disseminating information, and generating community support. They were also crucial in maintaining the linkage between AAS and the groups.

EVOLUTION OF THE METHOD In July 2002, AAS submitted the sub-project proposal to PETRRA for their approval. Mr. Harun-Ar-Rashid, executive director of AAS, already had had some success with the Farmseed project (see Chapter 18), but now wanted to specifically train the wives of these farmer seed producers on post-harvest issues. He was urged to get in touch with Paul Van Mele, who had been working on participatory technology development under the PETRRA Seed Health Improvement subproject. Possible ways to shortcut the participatory process were discussed when they met in Dhaka at a meeting facilitated by PETRRA. Paul had a written narrative of his first women's meeting in Maria village in which arose strong community engagement for the development of seed drying tables, which addressed a new need (see Box 3.1). This helped in understanding what really had boosted people's motivation and creativity to solve their own problems. He also selected and handed over 10-15 photographs, depicting underlying principles of Box 3.1 A New Need

30

Drying rice seed in the rainy season is a relatively new need, brought about by the introduction of irrigation pumps and modern varieties during the past decades. Rice grown in the dry or boro season is harvested at the onset of the rains. Drying tables emerged as a pro-poor technology, first developed by women in Maria village, Bogra (Van Mele and Zakaria, 2004).

Breaking Down Barriers

drying and a variety of designs of locally-made drying tables. The project proposal was further revised and approved in November 2002, after Harun had identified and appointed a woman (the senior author) to lead the project. Activities would be implemented in Kishoreganj and Habiganj districts in Northeast Bangladesh.

THE WOMEN-LED GROUP EXTENSION METHOD Build partnerships with local organisations As soon as the sub-project was approved, AAS officials visited the project areas, met with partner organisations, discussed the project objectives with their partners and agreed on their roles within the project. This went smoothly because AAS had worked in the area for many years and maintained good contacts with many organisations. Train women trainers Training courses were designed and imparted to the female field staff of AAS and partner organisations on rice seed drying and storage. Later on, women farmers joined the core of trainers, and group management and communication skills were emphasised. Select resource-poor women and form groups The next crucial task was to select women of farm families who: have land, but less than 0.4 ha consume rice of own production for 3-8 months

In Maria village, Sarifunnesa proudly shows us the seed drying table she made with her husband Hatem Ali: "Before we dried our seed on the ground, but this became a real problem when we started growing boro rice. We are proud of our table."

have experience in processing of rice and rice seed be physically fit, and are willing to undergo training, train others and disseminate information. Poor women were preselected by field staff of AAS and its partner organisations through home visits and a benchmark survey. Focus groups then decided on the final composition, with each group comprising 20-25 motivated members. Group members were either part of already existing NGO or community-based organisation groups, or had no prior exposure to rural development activities in their area at all. Despite members of existing groups being more likely to 31


become good women extension agents, AAS and its partners mostly preferred to have entirely new groups established for the project (Table 3.1). Each group selected a group leader in a democratic way while each partner organisation nominated a person to monitor and liase between them and the group. In some cases, the same person performed dual functions of group leader and liaison person. "What I have learnt from the project is very important and has given me a lot of benefits. Now I want other poor women to get the same benefits, so I decided to go to other villages myself," said Masuda, one of the women farmer extension agents in Haria village in Habiganj district. When Paul Van Mele asked her how she introduced herself in a new village, she replied "When I tell them my name and what I want to share with them, some people are suspicious, but when I tell them I have been trained by AAS, they get confidence. Those who are interested establish a group." Masuda is a convinced communicator and one out of many innovative farmers in Bangladesh. She takes care of two women's groups as part of the project, and out of women's solidarity and Islamic conviction she started activities in two more villages. Table 3.1 Differences between existing women groups and newly established ones for agricultural extension

32

VARIABLES

EXISTING GROUPS

NEW GROUPS

Gender-related barriers

Social and religious barriers have been overcome; female members already go to markets, work on the field, and attend local government meetings

Barriers still prevail, but decrease faster in villages where older groups are already in place

Intra-household decision-making

Women are more independent

Women depend on decision of male household members

Ease of using female group members as extension agents

High

Low

Openness to new ideas

Low. Members easily communicate and are in a socialisation process, but may be disinterested in new ideas

High. Projects can inject new ideas easier

Motivation of AAS and its partners to work with these groups

Low, because other organisations have already conveyed all their messages and inputs to these members, and little social capital remains to be built

High, because they can get new members, or help to overcome social barriers without having to cover the whole range of development activities


Strengthen group management and women's communication skills Several orientation and training courses on organisational and technical aspects such as group dynamics, leadership and information dissemination were organised once groups were formed. Especially the women farmer extension agents developed good insights in how to best select a group coordinator. "We observe the women's attitudes during the first meeting and select a group coordinator based on her talking skills, commitment, time consciousness and good leadership skills," said Masuda. Create a conducive learning environment Initially, Ms. Rokhsana Begum, agronomist from AAS and other women trainers from AAS and its partners facilitated the sessions. Later on, women farmer extension agents joined in. Training sessions were held in the house of one of the group members. The hospitality of the host towards the participants created a good learning environment. To further improve learning, groups were divided into two smaller ones during discussion sessions. In the beginning, the participants could not hide their anxiety and hesitation to attend these sessions, but soon after the discussions began, they found the topics very familiar and their participation became spontaneous. At first, participants were invited to narrate their experiences of drying rice and rice seed. Women treat their rice for home consumption and seed differently than rice for sale. The latter is dried directly on the courtyard floor instead of on bamboo mats or jute bags, and only for two instead of three days.

Photographs depicting daily life are used in women's group discussions and challenges them to look for creative solutions to improve their seed drying.

Introduce technologies by building on local knowledge and practices During these sessions the participants were given laminated A4size colour photographs, depicting different aspects of drying, and were encouraged to describe the advantages and disadvantages of each of these (Table 3.2). These sessions were held to stimulate the creative thinking process to improve seed drying. Explore locally available solutions After a short break, photographs of three different drying tables were shown to the participants. Women were again encouraged to express their opinion about the use, advantages and disadvantages. There was no real table on display, but the participants expressed their keen interest as if they were choosing a real one. Their decision criteria for a good drying table were as follows:

33


lower cost easy to make scope for multi-purpose use and easy to transport from one place to another. Table 3.2 Photograph-supported women's group discussions CONTENT OF PHOTOGRAPH

ADVANTAGE

DISADVANTAGE

REMARKS

Saris hanging to dry

Dries quickly under sunshine and open air

None

Dries quickly because saris are thin

Quilt hanging to dry

Dries, but slowly

Requires more time

Dries slowly because quilts are thick

Chillies and other spices drying on the roof

Out of reach of the children

Birds may eat if not kept an eye on

Dries quickly because hot roof produces more heat, but is too hot for seed

Various things drying on the machan (scaffold)

Could be protected from rats, poultry, rain; could be used for multiple purposes and as platform for sitting & gossiping

Poor farmers cannot afford it

Requires cash money at the time of construction

Rice seed drying on the yard/floor

Could be stirred by feet and hand as needed; all adult family members can contribute

More time consuming; clear sunshine required; children and poultry destroy rice

This is a traditional practice, familiar to local communities

The women were willing to make tables and some suggested that at least one table should be made in their presence by the project. Within a year, more than 60% of the project women made their own table. Selina Akhter, one of the women said: "I can motivate anybody to make a table, and if I would have a photo of my drying table, I would show it and discuss with others whenever I go out." This hasn't been explored so far, but surely opens up new routes for women-to-women extension. Improve access to outside technologies In the group discussion session, a plastic drum was exhibited before the participants and its benefits explained: it is airtight and moisture-proof; seed is protected from rats and insect infestation; the quality of seed remains good ensuring good 34


germination and better plant growth and subsequent yield. Although all participants were in favour of a plastic drum, some were a little disappointed, as it would be difficult for them to buy a drum at Tk 200 (US$ 3.50). However, after initial success of early adopters became obvious, the demand increased rapidly. Sufficient availability of plastic drums in the local markets of the project areas remained an acute problem for months. AAS, with financial assistance from PETRRA, then decided to organise and pay for transport of plastic drums, particularly for the poor. Others equally placed orders to buy plastic drums, but on a full cost recovery basis. Transportation cost per drum was Tk 25 (US$ 0.45) for Kishoreganj and up to Tk 35 (US$ 0.60) for Habiganj district. By early 2004, AAS had distributed about 600 plastic drums in the project area, the majority of which for the poor (Table 3.3), and for the following season other community members requested 200 more drums.

A proven cow: expose early success of innovators to group members In Bangladesh, the price of a cow is determined by her milk production. It means that one has to show the result first, and only then can start preaching if one wants to sell something. Initially, most poor women hesitated to buy a plastic drum and to build a drying table, because of the relatively high price of bamboo. But as the good results of those who were using these technologies became obvious, most got convinced about the value of the investment. Over 400 women made their own drying table, often with locally available materials cheaper than bamboo. Ms. Alia, 25 years old and having three children, explained how her family had received training from AAS on various topics. She now had enough of her own rice to feed her family for eight months per year, compared to only four months before Table 3.3 Percentage adoption of technologies by poor women in Kishoreganj and Habiganj districts TECHNOLOGY

PARTICIPATING WOMEN (N=570)

OTHER WOMEN IN PROJECT VILLAGES

WOMEN IN OTHER NEARBY VILLAGES

Seed drying tables

60-70

20-25

5-10

Plastic drum

80-85

20-35

10-15

35


the project. "If there is money, there is respect. Even my mother-in-law and my neighbours look up to me now," she said. Her group was considering to jointly save money to buy a power tiller. Stimulate informal communication and share with wider community The women's sisters, aunts and sisters-in-law received the news of achieved benefits first, after which it circulated among neighbours. Following initial success, the groups organised open sharing and demonstration sessions in their village. Pictorial brochures, a poster and a video documentary were produced to show the advantages of innovative rice seed drying and storage. For wide-scale information dissemination, a village fair was held in a festive mood at a suitable open space near the roadside in Pakundia, Kishoreganj. In February 2004, nearly 500 people, of whom 300 were women, gathered from 25 neighbouring villages. Local staff of AAS and representatives of the local partner organisations guided and briefed the visitors. Women farmers staffed the stalls, each designed for demonstrating various aspects of rice seed drying and storage, and proudly shared their experiences with others. Another attractive event in the fair was jari gaan, a folk song narrating advantages and disadvantages of the traditional and improved drying and storage methods, presented by a local folk singer group. Many PETRRA sub-projects adopted the use of folk songs in their activities, after this was brought to their attention by the NGO Shushilan in Southwest Bangladesh (see Chapter 10, Picture Songs).

KEYS FOR SUCCESS Good selection criteria and procedures All women were sincerely motivated to participate in the project. This was achieved by proper selection of partner organisations that helped in identifying resource-poor Village fairs and folk entertainment are a socially accepted way to disseminate agricultural information to women in rural Bangladesh.

36


women and forming groups. Carefully identified women farmer extension agents and group coordinators helped to reduce the communication gap. Better-off farmers, men and women, readily picked up on innovations that were triggered or introduced by the project to the poor. The reverse would probably not have been true. A culturally and socially sensitive method For training, women did not have to move outside the family boundary, which the society could have objected to. The organisation of functional groups of poor women helped to empower them, while training and exposure increased their confidence. Inter-personal communication processes between women ensured effective dissemination of information beyond project participants. Accessible and environmentally friendly technology There is no hi-tech or capital-intensive entrepreneurship required in this project. Bamboo or other materials required for making a drying table are available in every rural area and the farmers themselves can easily make a drying table. Confidence building and respect for the local knowledge system By validating traditional knowledge, women gained confidence in their problemsolving skills, as illustrated in Box 3.2.

RISKS, DIFFICULTIES AND ASSUMPTIONS People's expectations People, when approached by an NGO or community-based organisation, expect that they will get access to a broad range of rural development activities and inputs, such as credit. After all, that is what others got when they became a member of an NGO. Because this project only focused on post-harvest technologies, with training as a major input, women had little scope to actually strengthen their groups. Women's groups would likely be more motivated if a broader agricultural training package could be offered, including modern varieties, crop diversification and postharvest. Giving women access to a broad range of agricultural technologies is indeed the way forward, as discussed in Chapter 4 on the family approach by HarunAr-Rashid. Inadequate supply of plastic drums Plastic drums are hardly available at local village markets, only in towns. Moreover AAS could not accommodate increased demands for plastic drums after the project

37


Box 3.2 Teeth of Gold

by Helen Latifun Nessa At the onset of the session, I asked the women if they use any equipment to test seed dryness prior to storing their rice seed. All of them denied firmly, so I asked once more: "Are you sure that you don't have any equipment with you to test seed dryness?" "Yes, we are. We do not use any equipment," they replied. "Then how do you do the testing?" "We use our hand, feet, ear and teeth." "Don't you consider these as your equipment?" "Oh my God!" They all grasped the message and laughed loudly. When Ms. Rokhsana, the facilitator of the session, asked them to explain their traditional method, they replied: "A crackling sound comes when we shake seeds close to our ears." "A croaking sound comes when we bite rice with our teeth." "The seed feels slippery when we stir it with our feet." "The seed feels lighter when we test the weight of dried rice." Despite the range of diagnostic tools, all confirmed that using teeth is the best way. When I suggested they should take utmost care to keep their teeth well and strong, they all laughed again. Then the practical session began. Women were invited to rate three batches of seed with 11, 13 and 15% moisture content, respectively. They were all excited. One by one they came to the front and slowly started crunching the seed. All had a serious and attentive look on their face while Ms. Rokhsana noted down their opinion. Participants were still in for a surprise when Ms. Rokhsana brought out a scientific moisture meter to test the validity of their opinion. Once more doubt appeared on the participants' faces. "Do you have any objection?" she asked. Although a little confused, curiosity took over and they soon came forward accepting it as a challenge. Rokhsana explained that the moisture content of rice seed should be less than 12% and if it were higher, seed would require more drying. Rokhsana took the moisture meter and started testing. The results confirmed the women's opinion. The whole event made them very happy: they realised their teeth are worth gold.

finished, because of its financial and management constraints. Other means to improve storage could have been explored, such as painting earthen pots to make them airtight (see photo and Chapter 5). With strong community institutions in place, such as the federations in Northwest Bangladesh, communal storage also becomes an option, as described in Chapter 20. Shortage of technical staff Nowadays, highly trained people often face difficulties finding a job in Bangladesh. Yet at the same time there is a problem of getting qualified and motivated

38


agriculture graduates to work in rural areas, especially women. At least a core group of technically competent women trainers would be needed as master trainers to initiate the method elsewhere.

SCALING-UP Training women extension agents from rural communities had tremendous impact in mobilising other poor women. Bangladesh has many innovative women farmers, as will become apparent throughout the book, and surely it is a rich and so far under-utilised human resource in disseminating agricultural technologies. The method described in this chapter should be replicated in other areas of Bangladesh and countries where access of women to agricultural information is an issue. Since the implementation of this project and witnessing women's feedback, the agricultural NGO AAS started including women in all their extension activities.

CONCLUSION Farmers of all categories have accepted the women-led group extension method. Initially a core group of female trainers was needed, but as capacity was built within a village, innovative women farmers emerged as good extension agents. With some training and limited financial incentives, these women confidently developed new women's groups in neighbouring villages. This project has shown that poor women, when given the opportunity, display a high level of solidarity and commitment to alleviate poverty. Having an appropriate technology for the poor that is easy to extend, such as seed drying tables, gave women a fast and powerful experience of being successful extension agents and boosted their self-esteem.

Making earthen pots airtight by paint or old kitchen oil reduces people's dependency on outside technologies like plastic drums.

REFERENCES Van Mele, P. and Zakaria, A. K. M. (2004) From concept to impact: developing and communicating multipurpose seed drying tables in Bangladesh. In: Participatory Research and Development. Vol. 3 A. CIP/UPWARD Philippines, in press.

39

4 Family Approach in Extension Village women spread the word

Harun-Ar-Rashid

SUMMARY Rice cultivation in Bangladesh is a family activity as all household members contribute to it and have a say in decision-making. Therefore it is imperative to consider families as rice producing units rather than categorically divide training into gender-specific domains. The Agricultural Advisory Society (AAS), a small national non-government organisation, compared the impact of training either a whole family (with children), half a family (husband and wife), or husband and wife individually. Training husband and wife together gave equally good results compared to the whole family approach. Including women in training enhanced learning within the family and community overall, but particularly for insect and disease management. The latter seemed to be the most difficult topic to teach. Women, when interviewed separately, encouraged the project to give them more training, as men are often busy working the land. Women are eager to learn about all agricultural topics, not just post-harvest. 41


ACTORS AND NETWORKS We are often told that Bangladesh is a resource-poor country that is over-endowed with people and under-endowed with the resources to feed them all. But as executive director of AAS, I do not accept this. AAS has, from its earliest days, approached the challenges of agricultural development in Bangladesh from the perspective that Bangladesh, after all, is a rich country, blessed with abundant agricultural resources. We believe that Bangladesh's poverty is simply a reflection of its lack of capacity to effectively manage its rich endowment of resources. Accordingly, AAS has focused its energy on helping Bangladesh's farmers to become more productive in the context of their rich land, small plots, labour resources and abundant supplies of fresh water. AAS has developed a network of more than 150 non-government organisations (NGOs) and community-based organisations and is working hand in hand with them in sustainable agriculture and extension. From January 2003 to June 2004, we implemented a project on family training in three districts of Rajshahi division: Sirajganj, Pabna and Natore, in partnership with the NGOs Women Development Project, Pabna Protisruti and Lustre.

EVOLUTION OF THE METHOD

Providing training to women creates a window of opportunity

Traditionally, the Department of Agricultural Extension (DAE) targets mainly male farmers. Although in Bangladesh men are mostly engaged in transplanting and field work, while women play a major role in post-harvest, this does not imply that training should be restricted to these areas and players. Because all immediate family members participate in rice farming and are affected by decisions and results, it is important to understand the role of women in using and disseminating improved rice knowledge. Families have their own ways of allocating household labour, so it is essential to show respect for the family's internal dynamics by training households as rice producing units rather than dividing training into stereotyped gender-specific segments. The idea to test the family approach in extension arose when research in Bangladesh by the International Maize and Wheat Improvement Center (CIMMYT) found higher adoption rates of wheat technologies when all family members were trained together (Meisner, 1998; Meisner et al., 2004). We assumed the same would be true for rice technology, since the tasks of rice cultivation and wheat cultivation, in many ways, are pretty similar.

TESTING THE FAMILY APPROACH IN EXTENSION With financial support from the DFID-funded Poverty

42

Family Approach in Extension

Elimination Through Rice Research Assistance project (PETRRA), we compared the impact of training a whole family versus half a family (both husband and wife), husband alone, and wife alone. Setting the frame To test the approaches, we selected 12 villages, four from each district. In each district, one village was selected for studying a particular approach, so that each approach had three replicates. We identified 25 families in each village with the help of our partner organisations. Families needed to have less than 0.4 ha of land and three to eight months of rice provisioning ability. Where older children were in school and therefore unable to participate in the training, those families were classified under the half family group. Roadmap for research In each village we then organised a meeting to discuss with participants about the project's purpose and strategy. At the end of the meeting, groups were formed to make implementing the research activities easier. At first, we conducted a benchmark survey with selected research farmers and nonresearch farmers. Forty farmers joined in group discussions in each of the villages. A simple semi-structured questionnaire was used to gather information on farmers' knowledge of rice, crop and pest management, post-harvest, and quality seed technology. Finding out what people know laid the basis for developing our training curriculum and impact assessment. Needs-based training For two rice seasons, we taught farmers mainly in those areas where they needed improvement. Trained female field staff of our partner organisations and a female agronomist from AAS conducted the sessions, sometimes assisted by a male colleague.

Rogueing involves the removing of different varieties or off-types in order to get higher seed purity. Although normally done by men, women need training on these and other field activities as their husbands may be doing off-farm jobs at crucial times in the season.

From the various projects we implemented, we learnt that male trainers alone are unacceptable to women, whereas female trainers are accepted by men and women alike. The training module, contents, materials and schedule were not imposed, but negotiated between the farmers and trainers. The courses covered basic knowledge on rice plant, rice cultivation during aman (July - November) and 43


boro season (November - May), plant health management, post-harvest and modern rice seed production. Participants only received a light refreshment at Tk 15 (US$ 0.26) per head. Monitoring To assess the impact of the training, staff of AAS and its partner organisations held group discussions followed by individual field visits. A checklist was used to allow for comparisons between villages. Impact One and half years later, a two-week evaluation mission revealed that training half a family or a whole family had more advantages than the training of either spouse alone, particularly in gaining more knowledge, applying the new technologies, enhancing rice production, and influencing others within their community to adopt technologies. Differences between half family and whole family were negligible, so both can be termed as family approach. For assessing changes in knowledge, we asked twenty eight questions to each group of farmers. The score for each extension approach was determined as the percentage of correct answers based on the averages of the three villages. Groups were also interviewed in a control village (Table 4.1). Clearly people learnt a lot about agroecology, insect and disease management, and seed storage. Disease management scored lowest across groups indicating not only the complexity of the topic, but also the difficulty of teaching the topic. Lack of Table 4.1 Knowledge after receiving training in various modules. Control group did not receive training (average percentage of correct answers) TRAINING MODULES

WHOLE FAMILY

HALF 1 FAMILY

MEN GROUP

WOMEN GROUP

CONTROL GROUP

Morphology

74

87

77

66

44

Agroecology

75

79

64

40

42

Seedling production

95

93

92

91

54

Intercultural operation2

98

99

92

93

89

Insect management

80

79

64

63

37

Disease management

56

48

35

27

17

Post-harvest

97

98

92

90

84

Seed and seed storage

73

90

79

51

40

Average

81

84

74

65

51

1

44

Half family means husband and wife without children. 2Weed, water and soil fertility management.


good diagnostic skills and advice on plant disease management at the community level is a global issue. We also recognised that our staff lacked these skills and to address this, we embarked on a new project with CABI Bioscience (www.globalplantclinic.org) in 2003. That the women's group achieved relatively poor results on seed and seed storage came as a surprise. But in hindsight we learnt that some of the questions asked, such as 'what are the parts of a seed?' were not that relevant or important to change their seed storage practice. Table 4.2 indicates that groups where women were included stored their seed better than when men only were trained. But none of the farmers in any group applied all the new technologies that they learnt. In all groups people learnt that putting sticks in their field attracts birds and helps them manage insect pests, free of charge. Some started using a sweep net and destroying egg masses of stem borers, which can be easily seen as brownish clusters on rice leaf blades. What is important here is the concept of the family approach. Rice cultivation is entirely a family issue where all members make decisions together. This was confirmed during all focus group discussions. Yield increases of 20% and 27% were obtained when men only and husband and wife were trained, respectively. However, higher yields are only a portion of the measurement of technologies delivered and implemented that can increase farm incomes. Table 4.2 Percent of farmers adopting new rice technologies in their own fields NEW TECHNOLOGIES

WHOLE FAMILY

HALF 1 FAMILY

MEN'S GROUP

WOMEN'S GROUP

CONTROL GROUP

Modern rice variety

100

100

100

100

50

Seed bed preparation

80

86

87

75

0

Standard seed rate

69

67

75

33

0

Line sowing

52

57

72

67

25

Seedlings per hill

60

66

45

48

0

Balanced fertiliser

69

75

71

89

0

Insect management

67

71

65

56

0

Disease management

52

50

40

20

0

Water management

60

62

55

56

10

Quality rice seed

56

45

40

51

0

Rice seed storage

58

85

45

59

25

Average

66

69

63

59

10

1

Half family means husband and wife without children.

45


"Whenever my wife has to attend a training, I look after our baby," says Bulbuli's husband. Since his wife joined the federation in 1991 (a community institution, see Chapter 20), their life has improved significantly.

Training either the men or women alone would be cheaper in the short run, because with the same budget the project could train more families. But in the long run, the family approach may be more cost-effective as better results are obtained, there are more family members trained, and this knowledge is better shared with other families in the community. When we asked a random sample of 15 nonparticipating farmers in the villages where families had been trained, all of them had heard about the training and applied several new technologies in their field, compared to only 87% and 67% in the villages where women only and men only were trained, respectively. Besides, when both husband and wife were trained together, they gained more confidence as they applied the new knowledge in whole fields; when they were trained separately, they tried it out only in a portion of their field.

DIFFICULTIES, RISKS AND ASSUMPTIONS Children were occasionally distracting during training sessions, and either the husband or wife would then leave the session for a while in order not to disturb the group. If possible, husbands should take training with their wives, and if they do not have older children or grandparents to tend the little ones, arrangements could be made for some community members to tend the small children during the event. Overcoming social and cultural constraints for women's participation in training remains an issue in some areas, where villagers discourage women from leaving their homestead. Using women farmers as extension agents to train women's groups is one way to overcome this hurdle, as described in the previous chapter. Another way is to use video in women-to-women extension (see Chapters 5 and 7).

SCALING UP Although funded by USAID for the past 5 years, currently the Wheat Research Centre of the Bangladesh Agriculture Research Institute (BARI) is using government funding for whole family training. Additionally, they began using this approach for maize cultivation in 2002 and have trained over 5,000 families. A number of other projects funded by USDA and the Asian Development Bank will train growers in

46


maize cultivation using the whole family approach. But also the World Fish Centre of Bangladesh is applying this approach for fish cultivation, using many of the key concepts illustrated by CIMMYT (Meisner et al., 2004).

CONCLUSION The whole family training approach was developed by CIMMYT for disseminating post-harvest technologies on maize. AAS developed this concept further for rice and found that women should be given training on all aspects of farming, not just postharvest. Even if women do not work in the fields, they have a voice in household decision-making, and they can convey new ideas to their husbands, who may miss training if they are busy with fieldwork. Including women in training on all agricultural topics enhances learning within the family and community, and should be encouraged by all means. The project also showed that it is not necessary to have the whole family involved; training husband and wife together gave equally good results as the whole family approach. It should be left to the household to decide how many additional members should attend.

REFERENCES Meisner, C. A., Sufian, A., Baksh, E., Smith O'Donoghue, M., Razzaque, M. A. and Shaha, N. K. (2004) Whole family training and adoption of innovations in wheat-producing households in Bangladesh. Journal of Agricultural Education and Extension, in press. Meisner, C. A. (1998) Whole Family Training. VHS video cassette (15 min). CIMMYT, Dhaka, Bangladesh

47

5 Bringing Science to Life

Video development for women-to-women extension

Paul Van Mele A. K. M. Zakaria Rina Nasrin Basanti Chakroborty Md. Mojaherul Haque Josephine Rodgers

SUMMARY Over 18 months, the Women-to-Women Seed Health Video sub-project built capacity in national organisations, developed quality videos, and assessed their effectiveness. We formed two video teams based on organisational and individuals' comparative strengths and motivation. To break down communication and adoption barriers, we researched local knowledge, and involved rural women in developing and validating both the technologies and the video scripts. Video proved successful to catalyse local experimentation on a large scale. We ensured widespread dissemination of the videos by stimulating stakeholder ownership and pride in the project. In March 2004, the project received an award for effective communication from the prestigious International Visual Communication Association in London. The intricacies of the video development process are described in this chapter, whereas its impact, as compared to farmer-to-farmer extension, is presented in Chapter 7 called 'Watch and Learn'. 49


ACTORS AND NETWORKS An overview of the actors is given in Table 5.1. CABI Bioscience is the scientific division of CAB International (CABI) with Centres across the world, implementing more than 100 projects on sustainable agriculture. CABI helped in merging local and scientific knowledge and in ensuring participatory processes. Countrywise Communication, a UK-based private company specialising in video and multimedia training for agriculture and rural development, trained a local team in digital video production during two two-week sessions in January and March 2003 The Rural Development Academy (RDA) at Bogra has a mandate for training and action research. They coordinated the project locally, including the video production and impact study. TMSS, a national non-government organisation (NGO) with headquarters in Bogra, works with and for women in rural development. It addresses income-generating activities, agriculture, credit, agroforestry, fisheries and livestock among a number of other programmes. Apart from CABI Bioscience having worked with RDA since 2000 in the Seed Health Improvement sub-project (SHIP), none of the other partners had previously worked with one another. Trained by CABI Bioscience, RDA project staff had already acquired a solid understanding of participatory technology development. Both SHIP and this project are sub-projects of the DFID-funded Poverty Elimination Through Rice Research Assistance (PETRRA) project. When the Table 5.1 Process variables of the Women-to-Women Seed Health Video sub-project PROCESS VARIABLES

CABI BIOSCIENCE

RDA

TMSS

COUNTRYWISE COMMUNICATION

Type of organisation

Inter-governmental

Governmental

Non-governmental

Private sector

Expertise

Agricultural knowledge and information systems

Rural development

Women in development

Communication in agriculture and rural development

Input in project

Access funds; project coordination; participatory methods, script research and impact assessment

Audio-visual expertise; relationship with villagers with whom film was produced; local coordination, field research and video production

Gender expertise; interview skills; field research, video production and Bangla voice-over

Capacity building on digital video production; English voice-over

Motivation to get involved in project

To improve rural women's access to information; to diversify examples on how scientific and local knowledge can be integrated

To strengthen audio-visual unit as resource centre; to learn by doing about video production

To learn about the potential of video to reach rural women

To get involved in a project that measures the efficiency and impact of video as a communication tool

50

Bringing Science to Life

project leader (Paul Van Mele) first discussed the project with the executive director of TMSS, Dr. Hosne-Ara Begum, she immediately expressed her interest, despite the very small budget. Soon she appointed two senior staff members to work on the project. Equally, the director-general of RDA, Mr. S. M. Jahrul Islam, stressed the importance of the project during various staff meetings and public events, boosting the morale of the project. The method and training were highly relevant to both TMSS and RDA's ongoing programmes. All these factors contributed to each of the partners having strong feelings of ownership and accountability.

EVOLUTION OF THE METHOD The communication gap between research, extension and farmers has been under serious scrutiny over the past decades, resulting in a range of alternative approaches based on community participation in both research and extension, such as farmer field schools and local agricultural research committees (Braun et al., 2000). However, scaling up face-to-face extension is costly and the number of people reached is often limited, calling for a closer look at mass media (Snapp and Heong, 2003). Changing farmers' behaviour in rice pest management was possible in Vietnam through radio dramas, although the message was limited to one simple rule-of-thumb: do not spray insecticides for leaf folder control in the first 40 days after sowing (Heong et al., 1998). Video has been used extensively in rural development (Norrish, 1998; Coldevin, 2000; Bessette, 2001), even including topics such as soil fertility (Protz, 1998). But it is still under-explored as a means of merging scientific with local knowledge to get plant health messages across to the rural community. Based on discussions held over the past years with K. L. Heong from the International Rice Research Institute (IRRI) and Jeffery Bentley, an agricultural anthropologist and CABI associate, the senior author explored the idea of incorporating discovery learning by A. K. M. Zakaria Under another PETRRA sub-project, Paul from CABI and myself initiated participatory technology development activities in Maria village, Bogra. For me it was a new experience, and processes that build on local knowledge were gradually attracting me. Every day I was experiencing new avenues of farmers' empowerment and I felt changes within me. I could clearly see the right way to work with farmers. One day, while we were holding a village exhibition of locally-made drying tables for mid term evaluation, I wondered how a few activities could have had such an extremely valuable impact on women and stimulated their empowerment. Later that afternoon, we took the exhibition material to the crossroad of the village to look at the reaction of passers-by. At one moment, I told Paul that we should do something for sustaining and disseminating the idea. Paul smiled. There were a few boys with bicycles and all of a sudden Paul took one of the bicycles from a boy and challenged them for a race. Zooming away, he called out: "My friend, let's do it…" I now understand that this was the call for our video project journey.

Box 5.1 To the Crossroad

51


principles into mass media (TV) or small media (video). The opportunity to test this arose when PETRRA launched a new call for subprojects in 2002. Concept notes had to focus on women and uptake of post-harvest innovations. As CABI and RDA had good experiences working under SHIP, this provided an ideal opportunity to test the effectiveness of video as a communication tool to reach rural women (see Box 5.1). As we developed our concept note for a women-to-women extension project, we realised that the team had only one woman. The project needed women colleagues e.g. for preparing village women for video shots, attaching the microphone to saris, interviewing etc. Mr. Zakaria suggested collaborating with the national women's NGO, TMSS. In August 2002, the proposal was further discussed with Countrywise Communication and submitted to PETRRA, who approved it by November 2002.

PARTICIPATORY VIDEO PRODUCTION METHOD Develop a video team with complementary skills

Josephine Rogers from Countrywise Communication watches RDA's Kamrul edit a video. "Training people to use the video editing software requires tact, patience and practice," says Josephine.

We selected team members with complementary skills: familiarity of working with women in the village, interview skills, listening and talking skills, having a photographer's eye, subject knowledge, computer skills, among others. Two video teams were trained to stimulate competitiveness between teams. Each consisted of two men and two women which added to the diversity, richness and hence quality of the videos. The teams received training in video production techniques from Josephine Rodgers during two, two-week sessions. Each team member received training in all aspects of video production, so they were soon all familiar with the technical tasks and terms. At the house of Rina, one of the team members, her husband was complaining: "Whenever she watches TV, spontaneously she utters 'Mid-shot, long-shot, close-up, zooming in…'. Watching TV nowadays is just not the same anymore." By the end of the training each had found their niche, or preferred role. Kamrul, the RDA computer expert became a skilled editor: "Before I was blind, now I see with new eyes. Our thinking power has increased a lot." Conduct script research Assessing what people know and their attitudes toward certain practices is a first step in developing an adult learning programme. CABI Bioscience trained staff in the development of knowledge, attitude and practice (KAP) statements about seed health and post-harvest innovations. We agreed to produce four short training videos on seed

52


spots and sorting, seed flotation, drying and storage technologies (Table 5.2). Each would be a stand-alone video that could be used like a module in a training course. All statements were refined several times for relevance and clarity with about 25 women in 3 different locations in Bogra. Basanti from TMSS got back from a field visit one day saying: "I had a very difficult time trying to explain to women in the village about seed moisture content. We need to adjust that statement as women talk in terms of seed dryness, not seed moisture." Some statements dealing with seed pathogens or unknown technologies such as manual seed sorting were supported by live samples and short demonstrations. Once the KAP statements were completed, a benchmark survey in one village with one hundred female smallholders helped us determine the issues to address in the videos. By improving women's understanding of underlying principles (the life of pathogens and storage insects like rice moths, evaporation and ventilation during seed drying, air tightness of storage containers), we anticipated that women would be motivated to innovate and develop their own solutions, thus catalysing local experimentation based on global principles. Impact on learning is presented in Chapter 7.

Interviewing people requires specific skills, and having an eye for detail. Finding out what people know is a first step in developing a training programme.

All team members experienced the importance of identifying and involving both local innovators and technical experts in the early stages of script writing to ensure relevance, appropriateness and scientific soundness of the topics addressed. "We improved our skills of problem identification with farmers, both men and women, much better as we now have to pay more attention to details," said Parvin, who had already worked for three years in the Seed Health Improvement sub-project. Mojaherul Haque (Babu) added: "We can learn so many things from village women. They have many good ideas." Table 5.2 Post-harvest interventions addressed in videos SEED SORTING

SEED FLOTATION

DRYING

STORAGE

Brief description of technology

Manually remove diseased seed

Add salt or urea to a bucket of water until an egg floats; drop rice seed in the water and remove the bad ones that will float to the surface

Make a bamboo table or bench for drying rice; it can be quickly moved indoors in case of rain

Paint an earthen pot; fill it with rice seed and do not leave a dead air space; add leaves of neem or bishkatali and seal the pot. Store pot above the ground

Origin of knowledge and technology

Scientific principles; technology introduced by outsiders

Small modification of existing practice

Tables made through participatory technology development

Scientific and local knowledge and practice

Source: Adapted from Van Mele et al. (Chapter 7).

53


Interview innovators and respected individuals We identified women innovators who had good presentation skills. As some male farmers are selling cleaned seed and this was perceived as an important incentive to adopt seed health practices, we also interviewed a few men. Preparing the women for being interviewed such as briefing, positioning for interview, attaching the microphone is best done by a woman. Besides, having women rather than men interview village women makes them feel more relaxed. We also learnt that women in the village attribute higher credibility to the video when a scientist is also interviewed side by side with the local innovators. Illustrate improved household decision-making

"Women feel much more at ease when they are interviewed by other women," says Parvin while she clips the microphone on Parul's sari.

As male farmers have been equally involved in some of the post-harvest training sessions, we paid equal attention to showing activities where the whole household is engaged, such as manual seed sorting or developing seed drying tables. When the videos are shown in the village it is hard to keep men from participating. The video is more acceptable to the whole community if men and women appear on camera. Make clear and gender-sensitive voice-overs The men on the video teams realised the importance of having a woman in their team do the interviews and the voice-overs, to reduce the communication gap with the target audience. Validate and refine rough edits Five women from Magurgary village, who previously had not been exposed to any training, were invited for reviewing a first rough edit of the videos. We learnt that major improvements could be made to the script. The women first politely said they would immediately try out these new techniques. However, when we asked them how the videos could be improved, they spontaneously made a major contribution: a scene had to be included whereby one woman buys clean seed from another one while explaining how this improves her yield. From now on, the team would consider including role-plays in their future scripts. The women also mentioned that the messages were convincing, because they could see that other village women had also tried these new techniques and achieved good results. Besides, one of them said the techniques could easily be tested for their validity and that she

54


would get this message easily across to her husband. The women of Magurgary gave more credibility to things on video than to hearing people talk in real life. During the refreshers' training course and before the videos were finished, we displayed two of the videos for about 300 people in Nagar Tarun during Independence Day on March 26th, 2003 (see photo). The evening before, the video show had been announced by 'miking', where a rickshaw driver cycles through the village and announces messages over a microphone and amplifier. The local youth club organised the whole event. Show and discuss videos with village women Women said that the best time for showing videos in the village was at about 15.00 hours, after they had fed their families. They also said that it was important to see all the different aspects of seed health combined, not just as individual topics and that the videos should refer to one another. This advice was followed, so that each video addresses a specific topic in relation to the other post-harvest interventions. Each training video lasts only about 6-8 minutes. Making each one separately allows each programme to be shown at relevant times in the season, based on local needs. An entire show and group discussion can easily be held in an hour. In some cases, women asked to see the videos a second time. After each topic, our female project staff led a group discussion.

Video shows can be held for small or large audiences.

Bangladeshi villages typically consist of four to six paras or hamlets (see Appendix). Within each para, we invited small groups of 20-25 poor women to attend the video group learning sessions, held in the courtyard of one of the women. Since men and other interested women also showed up, we decided that in future the target group should get seats close to the screen. For research purposes, we rented a video player, TV and generator so that the team could easily show the programmes in all paras. How the videos will be used in future is discussed next. Assess impact and revisit communication strategy Each of the videos had a different response from the community. The seed drying and storage programmes were especially popular. The impact survey is discussed in Chapter 7. Based on women's feedback during the shows, we think that seed sorting needs supplementary training alongside the video. 55


Nowadays, nearly all villages have video players and TVs available for rental. NGOs such as CARE are using the videos in their training programmes. The government extension service (DAE) under the Ministry of Agriculture has well equipped audiovisual units at the district level, whereas eleven agricultural information offices across the country make use of mobile cinema vans to get messages across. Both could be used to further disseminate the videos.

KEYS FOR SUCCESS Teamwork and a shared feeling of ownership and pride in the project. All partners are motivated to promote the videos even after the life of the project. PETRRA, through a national communication fair and other forums, helped to announce the videos at the national level. Creative thinking and interaction with rural women. Experiences and innovations of resource-poor women fully incorporated to present an honest account. Both benefits and drawbacks of each technology are presented. New insights triggered by discovery and adult learning. At the end of every story, women are challenged to try out the new technology.

DIFFICULTIES, RISKS AND ASSUMPTIONS Camera sensitivity was one of the difficulties in taking shots and interviews. "Nearly all the women and men interviewed speak very well, but whenever they face a camera, they give opposite statements," says Basanti. To solve this problem, 'Practice makes perfect' was followed. Some trial shooting was made without tape and by cheering them up, the problem dissolved. Likewise, some women started talking in an artificial lingua in front of the camera. To avoid this, topics were discussed in their own words to make them free and to convince them that natural is better. Generally, women did not want to appear in front of the camera without applying makeup, putting on their best cloths etc., which sometimes took a long time. As the filming had to be as natural as possible, they needed to be convinced to wear their work clothes. Some of those not involved in the process tried to disturb the filming. They told the village women that showing their faces in a video programme would be reason for an unforgivable sin; that it is not allowed in our religion etc. These people were taken aside and reassured that they would also be in the film. Accordingly, in one afternoon the camera crew filmed them. They looked very happy; little did they know that most of the shots were taken without film. 56


Sometimes a huge crowd, especially children, made so much noise that recording interviews became impossible. 'Chocolate therapy', handing out sweets to keep them quiet, was a good solution. Although women were previously informed about the timing of the shooting, they could not always stick to the schedule, because of their high workload. In those cases, the filming had to be cancelled and gear packed up. Often other women could be interviewed or new shots taken so that the visit to the village was never in vain. The selection and arrangement of the location for shooting was not always an easy task. Sometimes, waiting for the right season to film was not an option. "Creating artificial rain on a sunny day, by sprinkling water from a bucket on the rooftop, remains a memorable event for film crew and village participants," says Babu, one of the video team members from RDA. Filming the different steps involved in a technology like seed flotation is relatively straightforward. It gets more difficult when showing how spotted (diseased) seed develop in poor seedlings, how the process of evaporation works, or how air humidity can 'migrate' through an earthen pot in other ways than via the lid. Traditionally, women in Bogra dry their seed on the floor, even during the rainy season. Visualising moisture absorption by seed that are being dried on the earthen floor was difficult, but creative thinking by the video team made the breakthrough. Now, a farmer who would agree to give away a comic shot had to be found. Showing his wet longi after having sat on the moist ground would give a powerful and funny message that people would talk about and remember. As the team knew the people in the village, one of the first potential candidates, Hatem, was approached. He immediately agreed to volunteer: "If I can contribute to the project in any way, it doesn't matter whether people will laugh with me". The risk of any video project is that the video programmes, once produced, remain on the shelf after the project has finished. With the strong feeling of ownership by all partners, it is anticipated that in this case the reverse will happen. Signs are already apparent that both the videos and acquired skills are being marketed extensively. This will be discussed in Part III on learning with rural communities.

Although we wanted women to wear their work clothes when being filmed, they often showed up in their best clothes.

SCALING UP Increasing demand During a communication fair, organised in 57


Dhaka by PETRRA in September 2003, more than 300 copies of the videos were distributed. "The demand was extremely high and most people even wanted to pay money for it. As it was an output of the project we didn't want to charge anybody, but it filled us with pride," says Zakaria. CARE Bangladesh requested RDA for at least five more videos to play in their project areas of Gaibandha, Naogoan and Dinajpur, mentioning: "We are regularly making use of these videos in our stakeholder workshops and in the meantime we got huge response from the villagers and local leaders." Helen Keller International is using the videos in their programme in Nepal. "Although our programme is based on vegetable production, it is of good use as the principles for post-harvest of rice and vegetable seeds are more or less the same," said Mr. Zaman, their country director. By May 2004, about 700 copies (both in Bangla and English) had been distributed. Developing own initiatives Apart from the four videos developed by the project, the team is now working on a fifth video about rogueing, which is a pre-harvest practice to improve seed health. Also a video was made from a 50-minute traditional song that was specifically written around seed health issues (see also Chapter 10). Promoting video expertise The training and feedback during different events boosted the team's confidence and video communication expertise is now offered to various projects and organisations working in the region. Addressing the nation Based on feedback from women in the village (see Box 5.2), one of the authors (Zakaria) took courage to approach a national cable TV station. One month later, the programmes started being broadcasted by Channel-i throughout the country. In May 2004, the programmes featured on ATN Bangla, another national TV station. Box 5.2 Thinking Big

"Both me and my husband enjoyed the video shows and we were the first to make a drying table in our village," Aklima claims. "Many women came to our house to see the table." Aklima turns to Munni, one of the women of the video team, saying: "If you could put the video on TV, it would be very useful for the farmers all over the country." A great idea was born.

Other spin-offs After seeing the success of this innovative project, CABI started adding video58


centred learning to its basket of farmer training methods in projects across the world. Various other results came from the project (Table 5.3). Spreading the word internationally In October 2003, the video project featured as the lead story in CABI Bio News, sent to over 3,000 stakeholders including donors. It was also hosted on CABI Bioscience's website (www.cabi-bioscience.org), which has around 150 visitors per day. Table 5.3 Results of the Women-to-Women Seed Health Video sub-project RESULTS

CABI BIOSCIENCE

RDA

TMSS

COUNTRYWISE COMMUNICATION

Enhanced interest in project outputs

Using videos as participatory training tool; videos are a good resource for CABI's Good Seed Initiative; raising awareness towards donors

Acting as video service provider; providing training on participatory video production in research and development.

Using videos as training tool in villages; using new skills and links to develop other videos

Broadening customer base; using videos for marketing purposes

Enhanced interest in working with project partners

CABI started working with RDA on a new project called the Plant Health Initiative; CABI is involving Countrywise in developing new project proposals

RDA is promoting its work with CABI to visitors; for future video productions that involve women, TMSS will be asked to partner

TMSS anticipates to produce videos with RDA on some of their own programmes (credit, agriculture, fisheries, poultry, community hospitals and others)

Countrywise is exploring new collaboration with CABI management

Countrywise Communication showed the video in the global partnerships pavilion at the Royal Agricultural Show 2003. At least 100 decision-makers from embassies, government agencies and the private sector saw them and were impressed by the work. OneWorld TV incorporated an overview of the project with short clips of the video programmes on their website (tv.oneworld.net). Award-winning On 26 March 2004, exactly one year after the first village video show was organised as a try out, the project received a prestigious award for effective communication from the International Visual Communications Association (IVCA), one of the main organisations responsible for promoting the non-broadcast video industry in the world. This award was given to us during an official ceremony in London with the following comments from the judges: "Simplicity and inclusiveness make this an efficient and relevant piece of communication. Support documentation for the project was well produced. The project showed a good understanding of simple objectives effectively portrayed."

59


CONCLUSION By involving rural women in developing and validating the videos and the technologies shown on them, communication and adoption barriers are reduced (see Chapter 7). Participatory video helps to build bridges between local and scientific knowledge, and should be further explored for the presentation of other technologies. The stakeholders were all proud of their work on the videos, which led to fast and widespread scaling up, but follow-up is needed to get the videos fully established within the system of national, regional and local service providers.

REFERENCES Bessette, G. (2001) Facilitating Stakeholder Participation Through Communication in Participatory Natural Resource Management Research. A Guide to Participatory Development Communication for Research Teams and Practitioners. IDRC, Ottawa. Braun, A. R., Thiele, G. and Fernandez, M. (2000) Farmer field schools and local agricultural research committees: complimentary platforms for integrated decision-making in sustainable agriculture. AgREN Network Paper 105. Coldevin, G. (2000) Participatory Communication and Adult Learning for Rural Development. FAO, Rome. Heong, K. L., Esacalada, M. M., Huan, N. H. and Vo Mai (1998) Use of communication media in changing rice farmers' pest management in the Mekong Delta, Vietnam. Crop Protection 17(5), 413-425. Norrish, P. (1998) Radio and video for development. In: Richardson, D. and Paisley, L. (eds) The First Mile of Connectivity. FAO, Rome. Protz, M. (1998) Video, gender and participatory development. In: Guijt, I. and Shah, M. K. (eds) The Myth of Community. Gender Issues in Participatory Development. Intermediate Technology Publications, London, pp. 165-177. Quizon, J., Feder, G. and Murgai, R. (2000) A Note on the Sustainability of Farmer Field Schools to Agricultural Extension. The World Bank, Washington. Snapp, S. and Heong, K. L. (2003) Scaling up and out. In: Pound, B., Snapp, S., McDougall,C. and Braun, A. (eds) Managing Natural Resources for Sustainable Livelihoods: Uniting Science and Participation. Earthscan Publications Ltd, London, pp. 67-87.

60

Part III: Learning with rural communities

6 Creative Learning Methods

Jeffery W. Bentley Paul Van Mele

INTRODUCTION About half of the population in Bangladesh lives below the poverty line, mainly in rural areas and with a majority of women (Duncan et al., 2002). But poverty is more than a lack of resources, it is also a lack of opportunities. To improve their livelihoods, the poor need to be able to benefit from information and services that are both production-oriented (e.g. agricultural technologies, credit and markets) and protection-oriented (e.g. pensions, health care and disaster preparedness) (Farrington et al., 2002). But often the poor cannot access services and information because they either lack time or money to seek it or the information providers are not perceived as socially or culturally approachable. Although from 1987 to 2000, men in Bangladesh have reduced the labour supply to agriculture by one third and women started to work more in agriculture, institutional services for development still by and large target men only (Hossain et al., 2004). Learning helps to transform information into knowledge. Even if extensionists sensu lato improve the poor's access to information, questions arise as to what extent farmers can apply this information. What does it help to learn about improved varieties if these are not available, or can only be obtained through great effort? Or to listen to a lecture or radio programme if the vocabulary is too pedantic or academic? And what is the best time and method to reach poor women, considering that the majority in rural Bangladesh is illiterate? Clearly, to assess the effectiveness of learning, one has to consider the education approach, the content and the way this has been developed, and the context. Creativity is needed in developing farmer education methods, but equally in engaging multiple service providers in pro-poor development. From 1999 to 2004, PETRRA focused on technology development, uptake and extension, and policy. At the start, since PETRRA was a research project, scientists from the Bangladesh Rice Research Institute (BRRI) had high expectations that

63


funds would be invested in their own activities. Instead, the project stimulated a wide range of actors and partnerships between NGOs, and the public and the private sectors to develop and test extension, communication and learning methods. All of the methods evolved during the project as PETRRA and its partners learnt by doing and exchanged experiences.

Open the door. Even when ideas for new technology are appropriate, they can be locked behind the doors of gender discrimination, marginalisation of the poor, or the exclusionary extension methods. PETRRA encouraged local and international partners to look behind the closed doors.

Under various donor-funded projects in Bangladesh in the 1990s, the Department of Agricultural Extension (DAE) was exposed to participatory approaches in farmers' needs assessment and experiential learning, but apart from project-induced activities their main mode of working was still meetings, and to a lesser extent demonstration plots and field days. Institutionalising changes in the largest national resource of extension staff was a major challenge, and low internal ownership was mainly blamed on donors driving the agenda (Pasteur, 2002). PETRRA's competitive bidding mechanism did not allow it to provide funds to DAE unconditionally, but the door was open to them to participate in the bidding. But DAE did not submit any concept notes, illustrating their initial resistance to competing with others for funds, or to joining partnerships. Over time, PETTRA or its partners often invited block supervisors to participate as resource persons, to give training, and attend field days or regional workshops. Through this involvement, DAE gradually realised their own strengths and weaknesses. The signing of a memorandum of understanding with a major NGO in August 2004 illustrates that an endorsement by the minister for agriculture can further push changes in mindset (see Box 21.1). The resistance of DAE to bid for sub-projects in a way created more opportunities for NGOs to emerge. Relatively few NGOs in Bangladesh focus on agriculture, and if they do, it is mostly not part of their main development programme. Many NGOs do not consider agriculture to be an entry point for reducing poverty, so they tend not to build agricultural expertise within the NGOs. This position is counter to the consistent findings of PETRRA that shows the impact of improved rice technology on poor households, and that agriculture offers a platform for livelihood diversification (Orr et al., 2004). Because NGOs recognised their limitations, many of those that bid for projects established formal and informal partnerships to link to agricultural expertise. The methods described in this part of the book were designed with three things in mind: reach, women, and ownership. By reach we mean that the PETRRA project challenged its partners to think about how their method could potentially

64

Creative Learning Methods

reach a large number of people. Some of them did manage to meet medium-sized groups, of say several hundred at a time, with a quality message, while others started experimenting with video and voluntary farmer extension agents. It is not a trivial matter; using conventional methods, extensionists have reached perhaps 10% of the world's farmers (Zijp, 1999). Few other sectors of the economy try to stay viable without new ideas; even such apparently humdrum businesses as leather handicrafts, cut flowers and plastic kitchen supplies fail without a fairly constant stream of new ideas (Fairbanks and Lindsay, 1997). Smallholders also need new ideas to keep the family farm viable. Women were important to PETRRA, and most of the methods made a special effort to reach them. Because of social norms, many women in Bangladesh hardly leave their village or even homestead, so ensuring their access to information and services was a major challenge (see also Part II on gender). Ownership: perhaps most importantly, many of PETRRA's extension methods were developed within the organisational culture of each partner, and were created or adapted locally through feedback from farmers. Encouraging local researchers and extensionists to think creatively and competitively about extension may have been as important in the long run as the new techniques they invented for growing rice.

BIDDING AND LEARNING WITH PETRRA PETRRA provided an opportunity for agencies to find new ways to work with the rural poor who relied on agriculture, but who had hardly enough rice for more than eight months of the year. Sub-projects learnt about new uptake and extension methods from each other, from PETRRA, and elsewhere, while PETRRA in turn built on experiences from its sub-projects. In an open tender in 1999, PETRRA launched a first round of calls for concept notes that focused on seed uptake methods; most of the projects that gained approval are described in Part V on pro-poor seed systems. In this call, the successful agencies were also selected to stimulate diversity in service providers, by involving national NGOs, regional NGOs, divisions of a government research institution, a local private company and a local NGO with little agriculture expertise. Each learnt unique lessons. Shushilan, for instance, was a local NGO in the southwest with a small geographic coverage, just two to three upazilas. It started in the mid 1990s with a strong social justice programme and a major emphasis on women. Leaders were from the local area. Shushilan had no prior experience in agriculture, but was like many local NGOs with a modest social programme and a deep commitment. PETRRA funded Shushilan for two reasons: first, it proposed links to local governmental agricultural expertise (extension services and regional research stations); second, it emphasised 65


women. Because of its clear poverty focus and lack of farm experience, Shushilan partnered with local experts to train women in rice production, not just in postharvest. Shushilan developed picture songs for motivating large audiences to adopt appropriate technologies, including integrated pest management. "This was a real breakthrough for us: not only was this the first time that a social development organisation capitalised on its own cultural expertise to deliver locally validated agricultural messages, but most of all their experiences highlighted that women should be given access to all agricultural information, not just on post-harvest," said Noel Magor, PETRRA project manager. To help poor farm families learn knowledge-intensive technologies, such as fertiliser management, PETRRA forged novel partnerships in its second call for proposals in 2001. The case study on soil fertility maps emerged from the synergy of two scientists: one a leading soil researcher at the Bangladesh Rice Research Institute (BRRI), the other an agronomist and executive director of the NGO Agricultural Advisory Services (AAS). Going Public is an extension method that teaches agricultural topics in public places. It was not proposed as a sub-project, but was introduced by CABI Bioscience in the Seed Health Improvement sub-project, based on CABI's experience in Bolivia (Bentley et al., 2003). PETRRA's flexible management helped to explore new frontiers. An external review by Gary Alex, extension specialist from the World Bank, led to a third and last call for concept notes in 2002. New proposals had to address womento-women extension, links with the private sector, or the extension of organisationally complex technologies. Women-to-women extension was specifically addressed by the award-winning video project and other case studies described in Part II on gender. To bridge the knowledge gap on herbicide use and safety, the NGO SAFE joined with the herbicide manufacturer Syngenta, BRRI and the UK Natural Resources Institute (NRI). They jointly designed training materials for use in farmer field schools (FFS) and for small-town seed and agrochemical dealers, the farmers' main source of information on herbicides (Chowhan et al., 2004; Riches, 2004). In another sub-project, IRRI partnered with the NGO AID Comilla who in turn linked up with various local NGOs. They used small-scale farmer experiments to convince farmers not to spray insecticides and to reduce nitrogen fertiliser application by making use of the leaf colour chart. The initial phase was followed by farmer-to-farmer extension (Jahn et al., 2004). Examples of sub-projects dealing with organisationally complex technologies are given in Part IV on enterprise webs. One of our strongest impressions of PETRRA is that the donors got their money's worth. The money spent on PETRRA went into technologies that the poor could use, into pro-poor seed systems, and into innovative extension and farmer education methods. PETRRA encouraged a real diversity of extension methods. Some were

66


created fresh for this project, such as village soil fertility maps and picture songs, or the prototypes came from elsewhere and PETRRA helped to shape them, such as Going Public. PETRRA also took some established methods and gave them a new angle, as making videos with women, for women. But competitive tender was not easy. Based on the concept note approval by the external technical advisory committee, PETRRA sometimes had high expectations about the performance of a partner, but in implementing the sub-project it became apparent that the partner had clearly different objectives and was not flexible to adjust its mode of working to align more closely to values set by PETRRA (see Box 1.2). In other cases, sub-projects were approved and started convincingly, but changes in project staff completely changed the dynamics of the partnership and modus operandi, and consequently required intensive counselling and monitoring. Some early champions under PETRRA started to attract the attention of donors and other organisations, got flooded in work and lost the focus of their sub-project. Donors have a responsibility in assessing an organisation's carrying capacity before funding new projects. The capacity of PETRRA management was also stretched to the limit by the diversity of organisational models, communication, education and extension methods.

PETRRA CASE STUDIES After briefly describing some of the case studies, we will discuss how face-to-face and mass media methods all have a role to play in triggering learning. We have assessed the methods for their poverty relevance. We call methods that reached Table 6.1 Key features of learning methods tested under PETRRA METHOD

CONTEXT

POVERTY RELEVANCE

Farmer field school (FFS)

Staff of the NGO SAFE had years of experience in FFS and could build on existing FFS groups

Inclusive, although poor may drop out due to time constraint

Videos (Chapter 7)

The project could tap into source of village women trained in other project and linked partners with multiple skills

Focused, poor women were specifically invited

Farmer-to-farmer (Chapter 8)

Neither the researcher nor the NGO had practical experience in participatory approaches, but they shared a vision of using them

Focused, poor farmer trainers who work the land establish new groups of poor farmers in other villages

Going Public (Chapter 9)

None of the partners had any experience in this method

Inclusive, although women were excluded when organised at markets

Picture songs (Chapter 10)

The method built on Shushilan's experience with social drama

Inclusive, there is little choice over who participates

67


both poor and non-poor 'inclusive'; while methods specifically designed to reach the poor are 'focused'. Watch and learn (Chapter 7) Videos are not only more cost-effective, but they may trigger farmers to experiment and change behaviour more than farmer-to-farmer extension. In other words, videos are cheaper for reaching an audience and they are better at getting ideas into people's heads. The counter-intuitive conclusion that videos may exceed the quality of face-to-face extension may be because videos, if well done, can be more convincing than real people. Well-written videos, with good photography and clever editing can hold a lot of compelling images and sharp, concise dialogue in a short time, and so may have more emotional impact than a person talking.

A simple technology, created with and for the poor. Suraton Bibi sits with her grandson Aktar on a seed drying table. "This is great," she says "after my family has finished threshing and drying the rice seed on this table, I can relax on it."

Although video is not a new method in rural development, the videos in the Bangladeshi case study use educational principles to teach women and invite them to experiment with appropriate low-cost technologies, informed by scientific agronomy and a respect for local farming. The technologies are simple new practices created with and for the poor of Bangladesh. The videos are made with village women and are especially suitable for audiences with low illiteracy rates. The authors recommend showing the videos in facilitated group sessions, especially for showing technologies that involve motor skills like sorting seed grain-by-grain. So an eight-minute video becomes a one-hour group discussion. Still, PETRRA's four videos on producing quality, farm-saved seed take up far less time than the original season-long participatory experiments, which aimed to test and fine-tune the technologies and learn about local perceptions and attitudes. A test for the future will be to broadcast the videos on network TV in Bangladesh, to really try them as mass media, and measure the impact. It will be important to learn how many people can adopt these appropriate technologies just by seeing them on TV. Nowadays, nearly all villages in Bangladesh have video players and TVs available for rent. Various pathways to reach the poor with videos are described in Chapter 5. The videos have been taken up by numerous NGOs and research organisations in Bangladesh and abroad, and various project partners are currently promoting the production of training videos as an efficient farmer education tool.

68


Village soil fertility maps (Chapter 8) Poor farmers, chosen and trained as extension agents by the sub-project, established and disseminated local fertiliser recommendations to at least 4,000 people in 216 villages. Farmer-trainers asked resource-poor farmers, who worked the land, to draw a map of their village, and to indicate the most fertile fields, the next most fertile ones, and so on. The farmers chose three to six fertility grades, gave each a number, and were asked to colour each on a map. Initially researchers confirmed each grade with laboratory tests which boosted credibility and morale among the farmer trainers. They then initiated experiments with the villagers to determine the right amount of fertiliser for each grade, based on crop requirements. Scholars have demonstrated either or both of these ideas of map-making with local knowledge of soils before (Behrens, 1989; Kanté and Defoer, 1994; Sandor and Furbee, 1996; Sillitoe, 1996; Defoer and Budelman, 2000; Herlihy and Knapp, 2003), but this was the first time they were used together in Bangladesh. The village soil fertility maps reached many people, for a face-to-face method. It reminds us of the 'promotor' method used in Latin America, where a full-time extensionist (e.g. an agronomist) trains farmers as extension agents. The farmer extension agents work part-time and are often paid for their time or rewarded with farm supplies. This method has been championed by Roland Bunch for many years (Bunch, 1982) and farmer extensionists are still working to good effect in Nicaragua, for example. It is too early days to say how this system of farmer-to-farmer extension will evolve in Bangladesh, but surely poor farmer trainers have gained respect in their communities.

Mozaffor Hossain demonstrates to other farmers how they can improve their farmsaved rice seed. When one of the passers-by notices the CDs with videos on seed health, he asks us how much one costs. He is obviously a better-off farmer; he said: "I just bought a TV and video player and would like to show these videos in my village."

Going Public (Chapter 9) As with the other case studies presented in this part of the book, Going Public addressed the quality-quantity dilemma: How to scale up farmer education? Most face-to-face extension methods reach audiences of 30 people or less. Going Public can reach several hundred people in one meeting, as the short sessions are repeated several times whenever the audience changes. With this method, participants can not be chosen. Because of the structural set up of the method, learning exercises need to be short and season-long field observations are impossible to do. The method is inclusive towards poor and not-so-poor, the facilitator has little choice over who attends a session. Going Public started in Bolivia, but PETRRA offered an opportunity to test it in Bangladesh. Mountainous and 69


sparsely populated Bolivia could hardly be more different from Bangladesh geographically. The Bolivian Going Public was designed for markets, where people come in from remote villages, hours away. Andean women not only go to markets, they play all the roles there, from consumers, to retailers, farmer-wholesalers and merchants. This is not the case in Bangladesh. Nash and colleagues soon realised they could reach women by Going Public in villages instead of in markets. Under PETRRA, the method was tested by two government institutions, BRRI and RDA, and the NGO AAS. After PETRRA, the NGO RDRS was convinced of its benefits and decided to start using the method in their agricultural programme from 2005 on. Picture songs (Chapter 10) The NGO Shushilan fought for women's rights, for the poor, and against poverty for years before starting to work with PETRRA in 2000, when for the first time they began using agriculture as a tool to improve human welfare. When Shushilan put agricultural science at the service of the people, the backbone of their effort were extension agents who met with community members in long-established, local clubs. They also used demonstration plots, with the added twist that most of the demo farmers were poor women, learning to work their own land, alongside their husbands and brothers, which helped the household keep down cash expenses for labour, while growing their own food. Picture songs elegantly combine music, dance and paintings by local artists. They perform upon request of local organisations.

Shushilan used village women extensionists and demo plots to good effect, but in Chapter 10 we chose to describe another of their extension methods, the off-beat and charming 'picture songs', little shows that travel upon demand from village to village, where women and their families can enjoy a bit of innocent and educational fun. The events are organised by local clubs or organisations. The troupe sings of new, appropriate technologies while a canvas with large paintings on a scroll illustrate the lyrics. The smallholders who adopt the song's technologies may double their rice yields while lowering their costs. The families were delighted with the results. "We have used folk songs and drama in agricultural extension for many years, but this is the first time that I see folk media being combined with large illustrative pictures," says Mr. Fazlul Haque Rikabder, director of the Agricultural Information Services at the Ministry of Agriculture.

70


The method is clearly inclusive; the troupe has no control over who attends these public gatherings. The authors recognise that the larger an audience gets, the harder it is to assess impact. Most NGOs working under PETRRA had no expertise in rigorously assessing impact of their extension methods, indicating an important area for capacity building. Under PETRRA, Shushilan performed 32 cultural shows with the agricultural pot song, and by October 2004, they had another 181 performances under other projects.

NEW DIRECTIONS IN FARMER TRAINING Through its value-based research, PETRRA guided sub-projects to target poor farmers, men and women. Technologies, extension and education methods were tested and validated by the poor. Farmer education differs from extension in that it is more farmer-centred than trainer-centred, with more emphasis on the learning process. In many cases it is hard to draw a clear line between the two as the level of learning depends by and large on the quality of facilitation. To avoid confusion we use the term learning methods here, as all methods allow for joint learning. In each sub-project staff learnt something from farmers Traditional top-down extension methods are designed to move information in one way: from researchers to extensionists to farmers. But working on extension methods allowed PETRRA's partners to learn from farmers, to everyone's benefit. Working together, the researchers, the NGO staff and the farmers modified the technologies in ways that made them more farmer-friendly. For example: Researchers had the idea for a rice seed drying table, but farmers taught them how to make a cheap table from local materials that would fit in the small rooms and porches of a Bangladeshi farm house (Chapter 7). While making village soil fertility maps, project staff learnt from farmers that rice plants did not respond to phosphorous as much as the chemical tests of soil suggested they would. This led to an improved lab test and more sensible fertiliser recommendations. Researchers realised that farmers knew so much about soil that they could draw soil maps, run experiments and work as extension agents (Chapter 8). Going Public was designed to learn from farmers, as well as teach. AAS learnt that farmers had many names for bakanae, without realising that it was a disease. Surprising as this may seem, it shows that farmers may have quite a different view of a problem compared to scientists. Taking farmers' perspectives into account helps to address them in a sensitive way (Chapter 9). Foundation seed, the basis for producing quality seed, is sold to resource-poor farmer seed producers by the NGO AAS in 3-kg instead of 10-kg bags (see Part V on pro-poor 71


Learning content: appropriate technologies In a World Bank review on research and extension, Purcell and Anderson (1997) stated that regardless of the extension system used, a supply of appropriate technology is essential if extension investments are to be worthwhile. By interacting with resource-poor farmers, men and women, PETRRA's partners validated and fine-tuned various second-generation technologies, ready for scaling-up, e.g. Using high-yielding BRRI varieties, bred for local conditions Improving farm-saved seed through training on seed health Saving labour with new generation herbicides Applying lower or better balanced doses of chemical fertiliser, based on a participatory assessment of the crops' needs for each particular soil type Complementing this with organic fertiliser, to improve soil texture and soil health Avoiding the use of insecticides. Methods: face-to-face and the media Besides the mass media like radio and TV, the 'small media' are becoming more important as new formats like internet and videos join newsletters, posters, pamphlets and older, folk media. While the media can be mass or small, face-to-face communication is not just for small audiences either; at conferences or political rallies a speaker may address thousands of people. Potential audience size is not strictly limited by the communication technology. Face-to-face extension takes place in many ways.

Many of the fresh tactics discussed in this book are aimed at medium-sized audiences, whether it is the narrowcasting of videos, Going Public or picture songs (see Figure 6.1). Each PETRRA partner fostered its own method, for various topics and depending on their own organisational strengths and interests. There is a place for both the media and face-to face extension. For instance, farmer-to-farmer extension worked quite well for the soil maps, but for training women on post-harvest we argue that video works better. This is because the soil information is eminently local, and must be taught in person, while conditioning and storing seed can be done in the same way across large areas of Bangladesh, so it lends itself better to the mass media. Marrying method to content Some methods are more appropriate for

72


Personal/Focused

Public/Diffuse

Small groups Farmer field school 1

Farmer-tofarmer

Medium-sized audiences Going Public

Picture songs

Videos

Large audiences Radio1

TV1

PETRRA did not use radio or TV, but these might be used in future projects, e.g. to show the seed health videos to more people.

Figure 6.1 Learning methods tested under PETRRA with audience size

Marrying method to content Some methods are more appropriate for certain topics than others, and could easily be used together, as we describe below. Table 6.2 presents an overview of how the methods tested under PETRRA can be used for a broad range of topics. Mixing methods Use intimate, face-to-face methods to learn about local knowledge, to develop and

Table 6.2 Versatility of learning methods METHOD

CONTENT

POTENTIAL FOR OTHER TOPICS

Farmer field school

Herbicides, planting density

Farmer field schools are highly versatile (see LEISA Magazine March 2003 at www.leisa.info); SAFE used it for teaching farmers about herbicide use and safety

Videos

Seed health (seed sorting, seed flotation, drying table, storage pot)

High potential for many topics; for certain topics, like seed sorting, videos can be used as small media, accompanied by demonstrations, while for other topics it can be used as mass media; is particularly useful to reach women in rural Bangladesh

Farmer-to-farmer

Getting the right balance of chemical fertiliser and encouraging farmyard manure

Soil mapping per se is closely linked to soil management, but other types of community mapping can be and have been used for other natural resource topics (e.g. irrigation, planting dates, weed management)

Going Public

Seed health (field sanitation, drying tables, seed sorting)

Has a high potential for other topics, especially ones that are visual, and can be demonstrated in a few minutes; is good for showing farmers how to diagnose a pest or disease where information like smell or size may be confusing if presented in photos or videos

Picture songs

Soil management, seed health, natural pest control, highyielding varieties (HYV)

May work best to introduce a topic, to motivate farmers or to make a technology more memorable; is not intended as a stand-alone method; can be used to encourage tolerance between religious communities

73


test the educational materials with their target audience, to get the words just right. For instance, while making the seed health videos, we learnt of a farmer who lit a candle in a seed storage pot as she closed it. This burned out the oxygen, and made the container even less insect-friendly. But then, once we have the right technology, thoroughly discussed and adapted by farmers, we should be able to teach it with small or mass media in most cases. Various methods can be used simultaneously to complement one another. It is a cliché that the further away we get from small, face-to-face methods, the more passive the audience becomes. It is not always so. For example, the PETRRA videos were made with farmers, with the same women and men who had planted the trials with researchers where the technologies were developed or validated. The videos can be shown in villages, so that an extensionist can help the villagers discuss what they have seen. But even if shown on TV, the videos should be able to invite the audience to experiment. Also the picture songs could and should be filmed and broadcasted on TV. Several extension methods combine well, such as the picture song with demonstrations and training with extensionists. Going Public and farmer field schools could interact with the media in interesting ways (see Chapter 9 and Bentley et al., 2003). Field schools could be turned into radio and video studios, where farmers could speak in their own words, to describe their experiences for the millions of farmers who will never be fortunate enough to attend a farmer field school. This is elaborated on in the concluding chapter. Miking, whereby a rickshaw or other small vehicle drives through the village and announces messages over a microphone and amplifier, is often used to announce political gatherings or public health issues in Bangladesh. Various PETRRA subprojects used miking to announce major events, such as the video shows, Going Public events, or mobile pump demonstration days (Chapter 13). Likewise, in Bolivia we recently saw a Going Public where agronomists set up their stall at a family-owned seed and chemical store. The shopkeeper announced the Going Public over local radio stations beforehand, which helped to draw in a crowd.

CONCLUSION Diversity, flexibility and creativity in extension, communication and farmer education methods is required: it allows service providers to tap into their own organisational strengths and use methods that are appropriate for the needs of their clients. To reach large numbers of farmers, men and women, there is a lot of scope for creating new methods of large face-to-face or small media. But irrespective of the method used, a good understanding of the local context, and the extent of building on adult education principles will determine its effectiveness.

74


REFERENCES AND RECOMMENDED READING Behrens, C. A. (1989) The scientific basis for Shipibo soil classification and land use: changes in soil-plant associations with cash cropping. American Anthropologist 91(1), 83-100. Bentley, J., Boa, E., Van Mele, P., Almanza, J., Vasquez, D. and Eguino, S. (2003) Going Public: a new extension method. International Journal of Agricultural Sustainability 1(2), 108-123. Bunch, R. (1982) Two Ears of Corn: A Guide to People-Centered Agricultural Improvement. World Neighbors, Oklahoma City. Chowhan, G., Orr. A. and Riches, C. (2004) Partnerships for information flow: Herbicides in Bangladesh. Proceedings of the National Uptake Workshop. Department of Agriculture Extension, Farmgate, Dhaka held April 17-18, 2004. IRRI, Dhaka, Bangladesh. Dagron, A. G. (2001) Making Waves - Stories of Participatory Communication for Social Change. Rockefeller Foundation, New York. Defoer, T. and Budelman, A. (eds) (2000) Managing Soil Fertility in the Tropics. A Resource Guide for Participatory Learning and Action Research. Royal Tropical Institute (KIT), Amsterdam. Duncan, A., Sharif, I., Landell-Mills, P., Hulme, D. and Jayanta Roy (2002) Bangladesh: Supporting the Drivers of Pro-Poor Change. DFID, London. Available at http://www.lcgbangladesh.org Fairbanks, M. and Lindsay, S. (1997) Plowing the Sea: Nurturing the Hidden Sources of Growth in the Developing World. Harvard Business School Press, Boston. Farrington, J., Christoplos, I., Kidd, A. D. and Beckman, M. (2002) Can extension contribute to rural poverty reduction? Synthesis of a six-country study. AgREN Network Paper 123. Herlihy, P. H. and Knapp, G. (2003) Maps of, by and for the peoples of Latin America. Human Organisation 62(4), 303-314. Hossain, M., Paris, T., Bose, M. L. and Chowdhury, A. (2004) Nature and impact of women's participation in economic activities in rural Bangladesh. CPD-IRRI Policy Brief 7. Centre for Policy Dialogue, Dhaka. Jahn, G. C., Kamal, N. Q., Azad, A. K., Dulu, N. I., Orsini, J. B., Morshed, M., Dhar, N. M. S., Kohinur, N. A. A., Barrion and Almazan, L. (2004) Completion Report on Livelihood Improvement Through Ecology (LITE). IRRI, Dhaka, Bangladesh. Kanté, S. and Defoer, T. (1994) How farmers classify and manage their land. IIED Issue Paper 51

75


Mundy, P. A. and Compton, J. L. (1995) Indigenous communication and indigenous knowledge. In: Warren, D. M., Slikkerveer, L. J. and Brokensha, D. (eds) The Cultural Dimension of Development: Indigenous Knowledge Systems. Intermediate Technology Publications, London, pp. 112-123. Norrish, P. (1998) Radio and video for development. In: Richardson, D. and Paisley, L. (eds) The First Mile of Connectivity. FAO, Rome. Orr, A., Seema, F., Nabi, S. A. and Peter, J. I. (2004) From technology to livelihoods. PETRRA Policy Brief 1. IRRI, Dhaka, Bangladesh. Available at http:// www.petrrairri.org Pasteur, K. (2002) Changing organisations for agricultural extension in Bangladesh: strategies for change. Lessons for Change in Policy and Organisations 3. Institute of Development Studies, Brighton. Available at: http://www.livelihoods.org/info/docs/Org_Chan_Bang.pdf Purcell, D. L. and Anderson, J. R. (1997) Agricultural Extension and Research, Achievements and Problems in National Systems. The World Bank, Washington D.C. Riches, C. (2004) Enhancing rural livelihoods need not cost the earth. PETRRA Policy Brief 2. IRRI, Dhaka, Bangladesh. Sandor, J. A. and Furbee, L. (1996) Indigenous knowledge and classification of soils in the Andes of southern Peru. Soil Science Society of America Journal 60(5), 1502-1512. Sillitoe, P. (1996) A Place Against Time: Land and Environment in the Papua New Guinea Highlands. Harwood Academic Publishers, London. Singhal A. and Rogers, E. (2003) Entertainment-Education: A Communication Strategy for Social Change. Lawrence Erlbaum Associates, London. Zijp, W. (1999) Keynote address of the 15th annual meeting of the Association for International Agricultural and Extension Education. Trinidad-Tobago, March 22-26, 1999. Journal of International Agricultural Extension Education 6 (2), 9-17.

76

7 Watch and Learn

Video education for appropriate technology

Paul Van Mele A. K. M. Zakaria Jeffery W. Bentley

SUMMARY To improve seed quality on-farm, four videos were developed with village women on seed sorting, flotation, drying and storage. The video production is described in Chapter 5 of this volume. In this chapter, the effectiveness and impact of this adult education approach is assessed. Besides being more cost-effective than farmer-tofarmer extension, video has the power to better explain underlying biological or physical processes and allows learning about local innovations. The seed drying and storage programmes created new knowledge on evaporation and porosity and presented it alongside local innovations, resulting in a high level of experimentation. After seeing the videos about 40% of the women changed their seed drying practices. The use of botanicals such as neem, bishkatali and tobacco leaves in storage containers increased from 5 to 75% with video and from 25 to 55% with farmer-to-farmer extension. Presenting people with a wide range of new ideas stimulates experimentation. After watching the videos, 85% of the women tried out different storage containers, mainly smaller and more airtight ones. Most of the women already practiced seed flotation in some form, but after watching the video, a few women added salt or urea to the water, which helps remove more bad seed. Seed sorting, going through it one grain at a time and removing all spotted and discoloured seed, was not widely adopted. But those who watched the videos learnt about seed borne diseases. Seed sorting demands motor skill, the way typing or driving does. For technologies that rely on motor skills, video training sessions must be complemented by hands-on training. 77


ACTORS AND NETWORKS CABI Bioscience is the scientific division of CAB International. Under the Womento-Women Seed Health Video sub-project, which started in November 2002, CABI helped in assessing local knowledge, developing an impact assessment survey, and ensuring farmer participation in the development of the technologies and the script (see Box 7.1). Box 7.1 How Video Can Make a Difference

To break down communication and adoption barriers, we learnt about local knowledge, and involved rural women in developing and validating both the technologies and the videos. Adoption of technologies was higher with women who saw the videos compared to face-to-face extension. See also Chapter 5 to learn about the video production process.

The Rural Development Academy (RDA) at Bogra coordinated the project locally, including the video production and impact study. RDA and CABI had a mutual understanding and shared vision as a result of collaboration under PETRRA since 2001. TMSS is a national NGO that focuses entirely on women in rural development. Two of their staff, Rina Nasrin and Basanti Chakroborty, helped film the videos and do the impact surveys. In May 2004, additional qualitative data were gathered by Dr. Jeffery Bentley, agricultural anthropologist, during two field visits, with help from Md. Mojaherul "Babu" Haque from RDA and Laila Arzumand Banu.

THE VIDEO EDUCATION METHOD On four award-winning videos, village women show how to sort, float, dry and store seed (Van Mele and Zakaria, 2003). Each video covers a specific topic (Table 7.1) and lasts only about 6-8 minutes. In the village, an entire show and group discussion can be easily conducted within an hour. We will describe how the same four technologies were conveyed using different methods in 11 villages in Bogra district: 1. Village video show followed by discussion 2. Farmer-to-farmer extension 3. Village video show followed by farmer-to-farmer extension. Video plus discussion In Husnabad village, 100 poor women were identified through well-being analysis and their knowledge, attitudes and practices assessed prior to the training. We 78

Watch and Learn

Table 7.1 Post-harvest innovations shown in video programmes SEED SORTING

SEED FLOTATION DRYING

STORAGE

Brief description of technology

Manually remove diseased seed

Add salt or urea to a bucket of water until an egg floats; drop rice seed in the water and remove the bad ones that will float to the surface

Make a bamboo table or bench for drying rice; it can be quickly moved indoors in case of rain

Paint an earthen pot; fill it with rice seed and do not leave a dead air space; add leaves of neem or bishkatali; seal pot and place it off the ground

Learning messages

Spotted and discoloured seeds are unhealthy; these cannot be removed by winnowing or seed flotation; seed sorting improves yield

Winnowing does not remove all insectdamaged and partially filled seed

Seeds absorb moisture from soil; wind helps in drying seed; a drying table has many other uses than just drying rice seed

Pots absorb moisture, which paint prevents; completely filled pots are dryer than half empty ones; some kinds of leaves repel storage insects

Local knowledge and innovations

Women have little knowledge about seed borne pathogens

Women already soak seed in water prior to sowing; flotation with salt or urea is a small modification of existing practice

Drying tables were designed with the full participation of local women and men

Traditionally some people sealed pores of earthen pots with used oil; only a few people use botanicals

divided them in small groups of 20-25 women and invited them to attend the video group learning sessions, held in one of their courtyards. Farmer-to-farmer In Telihara village, 100 poor households, also identified through well-being analysis, were trained by 30 families who had received training on seed health for the last four years, including some who had helped make the videos. These farmer trainers came from Maria village, in another union called Amrool. Video plus farmer-to-farmer In Narchi and eight other villages in Amrool union - including Maria village, where the techniques were invented or validated by farmers working with researchers - we tried a combination of both approaches. Large groups of people watched the videos followed by demonstrations by three male farmer trainers from Maria village. Impact assessment In Narchi village, we interviewed a small sample of the community. In Husnabad and Telihara village, we organised a quantitative survey to assess knowledge, attitude and practices before and after the intervention. 79


Although the videos target women, they generally bring their children along, while men gather at the back out of curiosity.

Each of the approaches has given us new insights into strategies for adult education, i.e. for finding the best and cheapest way to change farmers' behaviour and decisionmaking capacities.

LEARNING Almost everyone who saw the videos retained most of the key concepts. When we asked people to tell us what they had learnt in the video, they could all describe something. At first, one woman said she hadn't learnt anything, because she had watched through the window of the packed schoolroom, but then she reflected a minute and recalled learning about seed flotation. One man hesitated before answering, and then in a neat reversal of gender roles, his wife jumped in and answered for him; she remembered a great deal from the video.

Box 7.2 Technologies That Are Easier to Adopt

In some cases, people learnt more from videos than from farmer-to-farmer extension (Table 7.2). For example most people who watched the video learnt that winnowing does not remove all insect damaged seed, but only few people who received farmer-to-farmer training remembered this. Women's attitudes towards certain practices such as Address urgent needs seed sorting also improved more after they saw the Build on local knowledge video than in farmer-to-farmer training. and capacities Cost little and have low opportunity cost Result in perceived high economic or social benefits.

80

Learning is all well and good, but our real question was if the new ideas helped the women to improve their post-harvest practices. Factors enhancing adoption of technologies are summarised in Box 7.2. The level of adoption was

Watch and Learn

of the following order: seed sorting < seed flotation < seed drying < seed storing. We will now discuss each of these in the light of both the video and farmer-tofarmer extension. Sorting seed is tedious The local people already do a fair amount of sorting and cleaning seed before they put it up, but few understood that we meant 'sorting' to mean going through it one grain at a time and removing all spotted and discoloured seed. To clarify what we meant by seed sorting during the survey, we used a small sample and demonstrated it to the women before asking the question. After seeing the videos, all agreed this

Table 7.2 Changes in knowledge and attitude of women with farmer-to-farmer versus video-supported extension (percentage of correct answers) SURVEY QUESTION

TARGET ANSWER

FARMER-TOFARMER (N=100)

VIDEO (N=100)

Before

Before

After

After

Example where people learnt more with farmer-to-farmer extension Wind can dry seed?

Yes

8

27

0

4

Winnowing removes all seed with holes? (i.e. insect damaged seed)

No

8

17

21

72

Manual seed sorting is tedious?1

No

5

10

11

33

Manual seed sorting takes too much time?

No

4

20

10

56

Seed placed on earthen floor can absorb water?

Yes

42

34

43

79

Do you know what causes holes in seed?

Yes2

88

97

84

97

Do you know what causes spots in seed?

Yes2

47

70

47

75

Winnowing removes all seed with spots?

No

91

99

99

100

Air can pass through earthen pot in other way than via lid?

Yes

27

50

31

43

Keeping air from passing through my container is difficult?

No

78

96

69

97

Example where people learnt more with videos

Example where people learnt about the same with videos

1 2

Actually, it is tedious. The question is intended to ask if people find sorting too tedious to do, in spite of its benefits. These were not yes-&-no questions on the questionnaire, but we have simplified them here.

81


practice would increase yield, and only half still stuck to their opinion that it was too tedious and time-consuming. The video proved to be a better means of changing attitude than the farmer-to-farmer extension. Nevertheless, for reasons we will see below hardly any women actually started sorting. Sorting seed is needed Research in Bangladesh by PETRRA's Seed Health Improvement sub-project (SHIP) showed yield increases of 10-15% after proper seed sorting. But showing this technology through either video or farmer-to-farmer extension did not convince farmers to sort their seed. Because it takes a lot of time to sort seed, people will probably not sort it grain by grain, but may take the time to pick out the seed with the most conspicuous spots (Table 7.3). Golenur Begum, one of the women from Narchi, said that sorting thirty kg of seed took half a day with two people. However, she was probably doing a rough-&-ready seed sorting, and not exactly the seed sorting that the SHIP project recommends, because tests under SHIP indicate that a newly trained person requires up to eight hours to properly sort just one kg of seed, and as much as two hours after three seasons of training. Learning to quickly spot spots takes a keen eye and practice. Therefore adoption is higher if women watch the video and then have hands-on training, as was the case in Narchi village where at least two of the nine women interviewed actually went to some effort to manually sort their seed. However, Narchi village is also influenced by other seed health activities such as village contests organised by RDA and the Union Parishad in Amrool union. As confounding activities influences impact assessment in extension research, we will mainly draw lessons from work in Husnabad and Telihara village, where no SHIP project activities took place. Begum, the wife of Dulu Mia in Narchi village, proudly shows her rice seed storage pots she painted after watching the video. One is wrapped in plastic, and filled with seed, which she sorted. She is not sure how long it took her to sort, because she sorted it in between her other household tasks.

82

Seed flotation Of the 200 women interviewed in Husnabad and Telihara, more than 50% said they practiced seed flotation, but only four actually added salt to the water to float out more seed. Poor women think it is a waste to add salt, as they often cannot even afford to have salt with their meal. During the semi-formal interview a couple of months later, few people said they did seed flotation, but several women described it in detail, even mentioning putting salt in the water until an egg will float in it, so that the water will float out more of the partially filled and insect damaged seed. They said they would do it when they planted the rice they have stored now. This would indeed make sense, since most women already soak the seed just before planting it, to trigger pre-germination.

Watch and Learn

Table 7.3 Post-harvest innovations and their scaling-up potential SEED SORTING

SEED FLOTATION DRYING

STORAGE

Required capacity building

Initial training following the video session is needed to teach motor and observation skills to recognise the spots

Slight, mainly the concept that adding salt to water floats out more of the bad seed

Idea that seed absorbs moisture from soil and some simple skills for building tables to help dry seed in the rainy season

Several related ideas on porosity and keeping storage pots dryer e.g. painting them

Monetary cost

None

Only salt or urea need to be bought

Low to high; a light drying tables costs on average Tk 60 (US$ 1.1), and a heavy bamboo one about Tk 300 (US$ 5.3)

Low to high; a small pot of paint costs Tk 12, and a plastic drum costs around Tk 300; insecticidal leaves cost nothing

Opportunity cost1

High; a lot of time is required to properly sort seed

Low; activity can be easily done prior to soaking seed for pre-germination

Moderate; materials need to be collected and time is spent making the table

Low to moderate; used cooking oil and botanicals are readily available; paint or drum need to be purchased

Perceived need

Moderate; impact will mainly be high when seed quality is very poor

Low; women prevent storage pest outbreaks by improving drying and storage, and already remove bad seed when they soak them to pre-germinate

High; since irrigated rice was introduced, seed drying in the rainy season is a new and urgent need

High; women know that moist seed is more heavily attacked by insects, and women already spend much effort re-drying their seed monthly

Adoption

Low and partial adoption, i.e. only heavily infested seed are removed

Moderate

High

High

Scaling-up potential

In areas with known poor quality seed

Throughout the country

Throughout the country wherever irrigated dry season or boro rice is grown

Throughout the country; more local innovations could be included

1

I.e. the household's other options for spending the money and time needed to invest in this technology.

When we asked 'Do you know what causes holes in the seed?' people said: insects, poor seed drying and air in the container. People are well aware that there will be less insect damage if the seed is dried properly and the container is airtight. Their understanding comes from experience rather than from learning the underlying principles of insect ecology. No wonder women adopted drying and storage to 83


prevent insect damage more readily than floating out insect damaged seed with salt water: proper drying is easier to do, it is consistent with what they already do and know, and if done well may save them the tedium of drying seed each month. Drying is a new need In the past 30 years, Bangladesh has grown much more irrigated rice during the dry or boro season (November - May). This means that there was little existing technology for drying rice in the wet season. "During the rainy season, after the rain is over, we will dry. But if it continues to rain, there is nothing to do but put our seeds on the floor and stir them with our legs, to remove excess moisture," sighs Kancham, one of the women in Husnabad village who has not seen the video yet. Before seeing the videos, 43% of the women interviewed agreed to the statement that 'seeds placed on the earthen floor can absorb water', but that went up to 79% after they watched the video. Fifteen had made a seed drying table and 28 dried on thatch or jute bags instead of drying directly on the ground. We were surprised to observe no changes in the village where farmer-to-farmer extension had taken place. The education approach clearly influences the extent people start experimenting (see Box 7.3). Box 7.3 Igniting Experimentation

Video gives you the guarantee that the learning content you intend to get across will be conveyed in People experiment more readily when they are the same way over and over. When farmers have to provided with information train others on several topics, as in this case, you that helps them understand lose control about what will be taught. Although the the underlying principles of farmer-extensionists showed several seed drying that technology. tables, they were mostly covered up, used to display other materials, so farmers could not see what they were for. Similar observations were made when the tables were used in Going Public sessions (see Chapter 9 by Nash and Van Mele). Storing seed airtight is important After watching the video, Rabeya from Husnabad village speaks out in the group: "Every time I dry my seed very well and keep it in an earthen pot, but after I open

Spot the difference. Video can easily show how a technology is used in its natural environment (left). The function of the drying table is entirely masked during demonstrations when cluttered with displays of other materials (right).

84

Watch and Learn

the pot some months later I always find mouldy seeds and insects. Now I understand why it happened." People know from experience that airtight seed storage is important, but often do not have the money to buy a plastic or steel drum. Most poor households store their rice seed in jute bags or earthen pots and spend considerable time re-drying it every two to six weeks. While watching the storage video some women were surprised to learn that earthen pots can absorb moisture through their bottom and sides (although a fourth of them already knew that). Video allows presenting people with a wide range of new ideas to stimulate innovation. After watching the videos, 85% tried out different storage pots the next season, mainly smaller and more airtight structures. Several people mentioned painting their earthen storage pots. They find the idea not just practical, but a charming way to add a bit of colour to the home, since the pots are stored on bamboo platforms right inside the house. It was easy to find people who had sealed the pots with clay, and who had put bags in the pots to fill up the dead air space. This technology seems well on its way to adoption. The use of botanicals such as neem, bishkatali and tobacco leaves in storage containers increased from 5 to 75% with video and from 25 to 55% with farmer-tofarmer extension.

Saiphal Islam inspects his drum filled with rice seed. Note the blue pot, neatly sealed with clay and covered with a sheet of clear plastic. After learning about seed health technologies, Mr. Islam decided to try painting an earthen pot to store rice seed. He will carefully plant the seed from the pot and from the drum, and compare the yield from each one.

KEYS FOR SUCCESS Seeing is believing In 1938, thousands of people in the U.S. believed they were being attacked by Martians, after hearing the H.G. Wells' novel War of the Worlds in a radio broadcast. For many people, media is more convincing than being told by a person. When invited to give feedback on the first rough edit of the video programmes, one of the women of Magurgary village in Bogra district mentioned: "If you talk by mouth, people in our village will not be convinced, but we have a lot of faith when we see it on TV." Using appropriate language and symbols Preparing and storing seed is largely women's work in Bangladesh (Hartmann and Boyce, 1983). The women from Maria village who appeared on-camera were authentic. The fact that they had worked with rice seed all their lives no doubt helped win the sympathy of their audience.

85


When the NGO Agricultural Advisory Society later used the video in more than 30 villages in Sylhet district, none of the participants complained they couldn't understand the accent of the women from Bogra district. However, a fifth video, showing a local troupe from Bogra performing songs specifically designed around seed health issues, was hard to understand when shown in Sylhet. Many things influence whether or not people understand the dialect of another region. It helps if they are motivated in an interesting topic, if the speech is clear, and if there are visual clues like painted pots and drying tables. These factors need to be considered when assessing the scaling-up potential of both education and entertainment videos, which involved local communities in the production process. Hatem shows his wet lungi after sitting on the earthen floor. This powerful image in the video evokes laughter, and opens up the discussion on drying seed on tables instead of on the floor during the rainy season.

Training or education videos can easily show how, for instance, a drying table is made, as in the Do-It-Yourself programmes shown on European and American TV. Images can be more universal than language. Video can bring messages across in a visual way that is hard to achieve through face-to-face extension. How can one teach that seed absorbs moisture when dried on the floor? Having one of the farmers sit on the earthen floor and showing his wet lungi was a memorable image and provoked hilarious laughter. Surely this is something people will talk about. Hatem shows his wet lungi after sitting on the earthen floor. This powerful image in the video evokes laughter, and opens up the discussion on drying seed on tables instead of on the floor during the rainy season.

DIFFICULTIES, RISKS AND ASSUMPTIONS

It is difficult to measure how much people learnt. A survey may be too long, ask leading questions that give the answers away, and yield spurious results. For example, in the quantitative survey when we asked villagers if seeds with spots reduce yield, the number of correct answers actually went down after the video. The more abstract the learning topic, the harder to assess impact. A qualitative study is often more revealing, but it also takes time, and some practice to analyse the answers. But the qualitative study helped us to understand the reasons why some things were being adopted, for example we could see the proud look on people's faces when they showed us the painted pots; they clearly liked to paint the pots because they thought they were pretty. People often chose a sky blue, instead of a dark or drab colour. 86

Watch and Learn

SHIP suggested that farmers sort seed to remove diseased ones, which would raise yields by 15%. All well and good, except that it may take eight hours to clean a kilo of seed, especially if one goes through it rice-grain-by-rice-grain looking for the ones with little spots on them. From the semi-structured interview we learnt that at least some farmers are modifying the technology, e.g. by cleaning rice seed more quickly just to remove the really bad grains, or cleaning a small amount of seed and planting it to verify that the harvest really is better.

SCALING UP A challenge of any participatory method is to maintain high quality when scaling-up. Feder et al. (1999) suggested that the generic problem of scaling-up in extension could be partly overcome through mobilising other players in extension, empowering farmers and their organisations, decentralisation and use of appropriate media. Videos allow one to demonstrate underlying biological and physical processes. Video animation could be used in the future to explain insect and disease life cycles and be used, for instance, in farmer field schools or Going Public. However, for techniques based on motor skills, like seed sorting, video must be complemented by practical sessions. Videos are an excellent means of showing how technologies, such as seed drying tables, have been developed locally, and presenting many local innovations in an orderly way. For technologies that are newly introduced and perceived as time consuming, such as seed sorting, a prior selection of areas where they may have the highest impact could be part of developing the communication strategy. Video allows one to bring together and draw on a broad range of training modules to develop locally appropriate training curricula. By adding a short, locally made entertainment video, local ownership could be increased.

CONCLUSION We were surprised that adoption of technologies and experimentation was as high or higher among people who watched the videos as it was among people who had been trained by other farmers. For example, people only adopted drying tables if they had seen the video, but not after getting training from other farmers. The farmer trainers were, by the way, excellent; they were often the same farmers who had invented the technologies and appeared on the videos. What we did see suggests that a well-made video, showing functional technologies and their underlying principles, can help a good part of the audience to adopt these technologies, and it can do so much easier and cheaper than face-to-face extension. A good video gets its ideas into the heads of some community members, who will 87


experiment with them. By then, the video has already played its role, and adoption will depend on how functional and profitable the technologies are.

REFERENCES Feder, G., Willett, A. and Zijp, W. (1999) Agricultural extension: Generic challenges and some ingredients for solutions. World Bank Policy Research Working Paper 2129. Hartmann, B. and Boyce, J. (1983) A Quiet Violence: View from a Bangladesh Village. Zed Books, London. Van Mele, P. and Zakaria, A. K. M. (2003) Women-to-Women Extension of Post-Harvest Innovations. Video/CD. IRRI, Dhaka, Bangladesh.

88

8 Village Soil Fertility Maps

Mohammad Abu Saleque Harun-Ar-Rashid Paul Van Mele Jeffery W. Bentley

SUMMARY Poor farmers in Northeast Bangladesh can readily tell which of their small plots are the most fertile, and which are the least. When we asked them to, they deftly drew maps showing three to six different levels of fertility. Building on this local knowledge, the Agricultural Advisory Society (AAS) and the Bangladesh Rice Research Institute (BRRI) identified farmer extension agents and gave them a twoday practical course on how to apply fertiliser in their rice fields, in a way that took into account farmers' own knowledge, and scientists'. We helped them set up a farmer-to-farmer extension programme for the poor. Poor farmers were comfortable working with other poor farmers in their own or neighbouring village. In each village, one farmer extension agent helped groups of about twenty people draw a village soil fertility map. Then with some help from AAS staff, the farmer extension agents used the soil fertility maps to locate small field experiments, in which farmers learnt to adjust fertiliser applications based on crop performance. Once they had the results from those trials, farmers held a result-sharing village meeting, and made recommendations for organic and chemical fertilisers, for each type of soil and each rice cropping season. In four seasons, volunteer farmer extension agents trained about 4,000 other poor farmers to apply balanced doses of nutrients in 216 villages. To further scale up the method, we held a three-day, handson training course for field staff of the Department of Agricultural Extension (DAE); experienced farmer extension agents acted as trainers. The government extension staff was shocked, but pleased to realise how much farmers knew, and had learnt, about soil fertility. In the future, if paid a small stipend, some farmers would make excellent extension agents and local coordinators. 89


ACTORS AND NETWORKS Recommending the right fertiliser dose for a village in Bangladesh is complicated. Even within a single village soil fertility is a mosaic of soil types, each of which respond best to a different dose of fertilisers. Farmers with less than 0.5 ha of land scattered in two to five plots may need a different recommendation for each tiny plot. Obviously, no one can test the soil in each field, but BRRI and AAS recently developed a method that allows them to adjust specific recommendations for each field, without actually testing them all. It combines elements of farmer knowledge, farmer participatory research, selective soil sampling, farmer-to-farmer extension and participatory rural appraisal (PRA) exercises. Although people elsewhere reported participatory exercises for ecological mapping (Gupta and IDS Workshop, 1989) and soil fertility management (Kanté and Defoer, 1994; Defoer and Budelman, 2000), none of these publications were available to the team in Bangladesh, illustrating the vacuum many researchers and NGOs in developing countries have to operate in. The BRRI Soil Science Division analysis soil fertility and develops fertiliser management packages. Through an exchange visit organised by the PETRRA project, one of the authors (Saleque) enrolled in a participatory crop improvement training course at the University of Wales in the UK. This course and literature on local knowledge (e.g. Murage et al., 2000) triggered the idea of mapping soil fertility with farmer colleagues, to develop their own nutrient management strategies. A partner organisation had to be found to help put these ideas into practice in the field. AAS, a small national NGO described in previous chapters, works on about ten different rice projects with men and women farmers. AAS had previous experiences with agriculture, but not with participatory research in soil fertility. The director of AAS, Harun-Ar-Rashid, is an agronomist who once worked at BRRI and is on good terms with their scientists, so collaboration was natural. Twelve local NGOs and community-based organisations (CBOs)1, who have been working with AAS in other projects in Habiganj and Moulvibazar districts, in northeastern Bangladesh, helped identify and organise poor farmers. Although Bangladesh has competent soil labs, in general, farmers have limited access to soil-testing services and depend on their own experience, advice from neighbours and DAE field staff (block supervisors), who have local soil maps, but not information at the individual field level.

EVOLUTION OF THE METHOD The first year (2002), AAS and their colleagues drew soil fertility maps with 12 villages. AAS started the meetings by asking the local people to draw a map showing the most fertile fields, the least fertile ones, and the in-between. Most of the 90

Village Soil Fertility Maps

communities had made simple maps before, in other meetings with outsiders. The local people intuitively grasped the idea of the different grades of soil fertility and sketched maps from scratch with markers on white boards, showing the grades of soil in their village. During the first three seasons, the two senior authors and other staff took several soil samples from each grade of soil that farmers identified in each village (see Table 8.1). Back in the lab, we were delighted to see that time after time, the soil that farmers said was the most fertile, actually was. Farmers had no trouble ranking local soils by fertility, and our scientific tests backed them up. In time, we began to trust the farmers more, and took fewer samples, only in those villages that had clearly different landscape and soil features from the ones we worked in previously. All villages had piedmont soil. We learnt that farmers think that the most fertile soil is black, deep, clayey, with lots of earthworms; it retains water well. They think of the least fertile soil as thinner, whiter, sandy, with fewer earthworms, holding less water. Some farmers say that fertile soils have a distinct smell. On the other hand, we researchers perceived fertile soil as being high in organic matter and nutrients, especially nitrogen, potassium and phosphorous. In spite of our different perspectives, the results of our laboratory soil tests for organic carbon, total nitrogen, available phosphorous and exchangeable potassium correlated well with farmers' perception of soil fertility. But there was one problem (see Box 8.1). In the villages, we learnt how farmers managed soil for each fertility grade. For example, farmers tended not to apply phosphorous or potassium, only urea, and not enough of that. By pooling scientific and farmer knowledge, we reached a consensus on which improved soil fertility management practices to test in each soil grade. This included the use of gobor cow manure, which everyone agreed was a good thing, higher doses of nitrogen than the farmers were using, and potassium. But because of what we had learnt with the farmers and laboratory tests, less phosphorous was suggested than we would have initially. Fertilisers were applied in split doses, each test plot receiving the same basal amount. At first, the soil tests for phosphorous puzzled us. The tests showed that phosphorous was low, but when we asked farmers, they told us that in the past their crops hardly responded to phosphorous. Actually, our results turned out to be an artefact of the testing method. When we followed standard lab practice - drying soil and grinding it before measuring nutrientsphosphorous readings were low. But when we measured the phosphorous on waterlogged soil, which is where rice grows, there was a higher pH (6 instead of 5) and so more available phosphorous. This was a good lesson for us: we scientists need to be careful in validating local knowledge, as sometimes our tests are inappropriate. The new laboratory technique, which emerged from working with farmers, analyses soil samples in a way that is much closer to the actual field conditions (Saleque, 2004).

Box 8.1 Farmers Lay Basis for New Lab Technique

91


Then, we tested the newly negotiated fertiliser applications in two or three fields per grade, comparing them side by side with farmer practices, in each of the 12 villages (Table 8.1). Farmers helped design the experiments, select the fields, decide on plot sizes and which rice varieties to plant. They did the trials on their own and took care of the fields like researchers, helped by field coordinators from AAS. Project staff recorded plant height and tiller number every 15 days. At the active growing stage, staff collected rice leaves to analyse them for nutrients. By observing crop colour and growth at different stages, farmers adjusted the second and third doses, so that the total amount deviated slightly from what we had planned initially in the group meeting. From the training farmers had learnt that it was better to apply the last top dressing of urea at about seven days before flower initiation, rather than any time hereafter, and also that too vigorous crop growth could be slowed down with small amounts of potassium. As our confidence in local knowledge grew, we took fewer soil samples. "As an agronomist, I felt very uncertain in the beginning of the project, but now I believe in my heart that farmers have the knowledge to correctly assess soil fertility and make a village map," says Mr. Ferdous, one of the enthusiastic young staff members from AAS. Table 8.1 Evolution of farmer-to-farmer extension for soil fertility management

92

SEASON

VILLAGES (n)

FARMER EXTENSION AGENTS

SOIL TESTING

COMMUNITY LEARNING

Aus 2002

40

0

Randomly 2-3 samples from each fertility grade, in each village

3-4 field trials per fertility grade, comparing local practices with our recommended fertilisation

Aman 2002

80

0

Same as above

2-3 field trials per fertility grade

Boro 2003

10

17

Same as above

1 field trial per fertility grade

Aus 2003

33

35

Randomly 2-3 samples from each fertility grade, but only in villages that had distinct soil features

In 12 villages we compared local practices with our recommendations, and in the others we only did demonstrations of improved fertilisation

Aman 2003

79

79

Same as above

Demonstration plot only

Boro 2004

82

81

Same as above

Demonstration plot only

Total

216

212


The farmers harvested the rice when it was ripe, and project staff recorded yield data. The plots with our negotiated fertiliser recommendations averaged higher yields than the ones that followed farmer practices, ranging from 12 to 49% in the minor rain-fed or aus season (April - August), 10 to 38% in the monsoon or aman season (July - November) and 35 to 40% in the dry boro season (November - May) (Table 8.2). Application of cow dung differed for each village and household, but farmers never applied during aman. After each harvest we held a village training workshop. According to the farmers even the lowest yield increase of 9.5% in aman season still justified the additional fertiliser cost of Tk 880 (US$ 15) per ha, their net profit being Tk 1,820 (US$ 32) per ha. Based on the soil fertility maps and results of the field experiments, we refined our nutrient management packages for each soil fertility grade (Table 8.3). Both the village map preparation and encouraging results of the soil fertility management experiments created great interest among the farmers. The ground had Table 8.2 Response of rice to different soil fertility management packages in Habiganj and Moulvibazar, 2002-2003 Fertility grade1

Grain yield (ton per ha)2 Farmer practice

% increase

Improved practice

Aus season (variety 539) I

3.07

3.50

14.0

II

3.41

3.81

11.6

III

3.16

3.77

19.4

IV

2.88

4.02

39.6

V

2.53

3.77

48.8

Aman season (BR11) I

3.04

3.58

17.6

II

3.08

4.23

37.3

III

3.80

4.16

9.5

Boro season (BRRI dhan 28 & 29)

1 2

I

4.55

6.13

34.7

II

4.01

5.54

38.0

III

3.90

5.46

39.9

Soil fertility grade according to farmers, I indicating the most fertile soils FP = farmers' fertiliser dose, IP = improved fertiliser dose

93


been prepared to introduce farmer-to-farmer extension. When we realised that the method was scientifically sound and farmers were keen on it, the project team brainstormed on how to scale it up with limited funding and staff time. AAS pioneered farmer-to-farmer extension in various other projects, such as the one on women-led group extension, described in Chapter 3. The following extension method can be used not only for soil fertility management, but for any topic on which farmers, men or women, have developed a deep local knowledge.

THE FARMER-TO-FARMER EXTENSION METHOD Explore strengths in local knowledge In this project, local knowledge on soil fertility was the starting point to develop training and field research. Identify innovative resource-poor farmer extension agents Once farmers had drawn a village soil fertility map and set up field experiments or demonstration plots for comparison, it became easy to find and train three to four volunteer farmer extension agents from each village. This was especially true for women. Some poor women extension agents in the Women-led Group Extension subTable 8.3 Fertiliser use in farmers' practice (FP) and improved practice (IP) plots in Srimongal upazila, Moulvibazar, 2002-2003 Fertility grade1

N (kg per ha)

P (kg per ha)

K (kg per ha)

S (kg per ha)

FP

FP

FP

IP

FP

IP

IP

IP

Aus (April - August) I

15

56

4

15

5

40

0

9

II

35

52

9

17

10

46

1

10

III

41

43

8

15

35

47

0

8

Aman (July - November) I

37

80

3

22

5

69

0

9

II

35

75

2

21

10

66

0

9

III

47

66

5

18

28

47

1

9

Boro (November - May)

1

94

I

49

70

4

19

7

58

2

9

II

32

90

4

16

8

59

0

8

III

57

92

5

18

16

62

0

10

Soil fertility grade according to farmers, I indicating the most fertile soils


project, described in Chapter 3, were even going to more villages than those required by the project. They appreciated the benefits of the project so much that they wanted other poor women to get the same benefits. Solidarity and a desire to serve the community are often the motivation for farmer-to-farmer extension. Farmer extension agents should be willing to work in their own or a neighbouring village to develop a village map and soil fertility management packages. During the last year of the project (2003-2004), we trained 212 farmer extension agents who were selected for their educational level, leadership ability, and enthusiasm. Young, illiterate farmers also emerged as excellent extension agents, being able to mobilise poor farmers quickly. "I can't read," admits 18-year-old Ramiz Ali from Mirzapur, "but I have many good friends in my village and they help me to write and read whenever I organise a group meeting." As AAS implemented six PETRRA sub-projects on seed, fertiliser and crop management in northeastern Bangladesh, many farmers had already received training on various topics. Especially when seed producers under Farmseed (see Chapter 18) were trained to become farmer extension agents, they became increasingly recognised as a reliable source of quality seed and agricultural knowledge. The principle of reducing transaction costs by combining multiple services by the same people is further discussed in the final chapter of this volume. Train extension agents All farmer extension agents received a two-day training in the AAS office, in which the first day dealt with soil fertility and the second day with the actual extension method. During their first attempt to prepare a village map, the project staff supported these new extension agents.

Farmer extension agents in Bondaue village learnt about soil fertility and helped the project by drawing soil maps in neighbouring villages. They also learnt other farmers to adjust their fertiliser application based on crop colour and growth. Once local suggestions for fertilisation were validated, they kept on visiting farmers about once a month, often in the teashop.

Training becomes easier once farmer-extensionists have undergone the whole process in their own village. By organising training-of-trainer courses, an exponential growth in number of villages trained can be achieved. Later on in the project, DAE block supervisors were also trained. This time the training had to convince people who were entirely new to the concept. The training was essential for institutionalising the method among the government's extension service, who have the country's greatest human resource. AAS developed a three-day curriculum, with experienced farmer extension agents as trainers. Classroom exercises 95


were supplemented by village group discussions and drawing soil fertility maps. Form farmer groups As they are poor themselves, farmer extension agents face few problems in selecting and motivating other poor participants in their own or neighbouring village. In general, the project selected farmers with less than eight months of rice provision ability. For block supervisors it may be more difficult to keep away the wealthier farmers, who hire labour instead of working the land, and who inhibit poor farmers from speaking up. Each group, should consist of about 20 real farmers, who plough the land and know it intimately. Women farmers should be included in the group if possible. Visualise local knowledge "Visualisation is key to learning and is very useful in group decision-making, especially when farmers are illiterate," says Ranadhir Datta, one of the first farmer extension agents. A good map can indeed explain a lot, and when developed by farmers it creates pride and ownership. Village soil fertility map drawn by farmers and extension agents.

Village soil fertility maps have to show village borders, main roads, rivers and crop fields. Different fields need to be identified and named with their local place names on the map by participating farmers, who then identify the most fertile fields and mark these as fertility grade I, the next as fertility grade II and the least fertile fields as grade III or higher. Elicit farmers' own terms for soil In the future, it would be interesting to elicit farmers' own terms for soil, e.g. black soil, sandy soil, asking them to rank these ethno-soil types by fertility, test them, make broad ranges of recommendations. These local categories of soil would have to be measured periodically, since they may vary somewhat geographically, but extensionists could then suggest other farmers "apply between such-and-such in black soil, so much in sandy soil" (see also Box 8.2). Of course, unravelling local terms is easier said than done. In his book Indigenous Knowledge Development in Bangladesh, Paul Sillitoe (2000) mentions several characteristics farmers use to describe soil. Farmers talk about jore or strength of a soil. Strong soils give good yields, but 'strength' is complex and is influenced by several pedological factors. Folk terms may have several related meanings. For example, farmers talk about ras or what is left behind after flood waters recede. But ras also comes from rain

96


and groundwater. For some farmers ras means soil moisture, for others what is taken up by plants. A soil that expresses the benefit of organic matter has quyat, which is not visible but adds to the jore, the 'strength' or fertility of the soil. Explore ways of improving local practices For each fertility grade, we identified the farmer's soil fertility management practice that gave the best yield, recorded the other farmers' practices and discussed the yield difference with them. In Chiapas, southern Mexico, researchers found that smallholder farmers had their own names for five local soil types: tierra negra (black earth), tierra baya (yellowish white earth), tierra colorada (red earth), tierra colorada arenosa (red sandy earth), and tierra cascajosa (earth that resembles subsoil or exposed bedrock). Researchers sampled 104 fields. An analysis of variance using the soil classes as the grouping factor indicated that farmers' soil taxonomy discriminated among the objective properties in the soil (organic matter, pH, % sand, % clay) and that objective properties were consistent with farmers' perceptions. However, ethno-soil names and their physical properties vary from village to village, and researchers must check them anew in each community where they work.

Box 8.2 Mexican Smallholders also Understand Soil

Source: Adapted from Bellon (2001)

Through group consensus building, an optimum soil fertility management package for each fertility grade and each season is developed. In each grade, a collaborating farmer tries the recommendation in a whole field (about 0.1 to 0.2 ha), adjusts applications based on crop colour and growth, and the group compares the results with three neighbouring farmers' fields. After harvesting the crop, the recommendation for each soil fertility grade is discussed and fine-tuned in a village workshop. In three years, we conducted 1,177 field trials of which about 1,000 were led by farmer extension agents. Many of the farmers who were initially reluctant to apply phosphorous, potassium and sulphur to their rice field, became motivated by observing farmer-led demonstration plots. Share findings with wider farming community While fine-tuning the fertiliser recommendations in the village workshop, farmers get motivated by their peers. With the map, each family can see to which fertility grades each of their plots belong, and how to apply plot-specific fertilisers. Later, we helped farmers summarise the recommendations and copy the soil maps onto an A4-sheet, and make ten photocopies for the village. Chatting with farmers, we came up with some ideas for improving the method. Farmers often discuss political, social and agriculture-related issues with their colleagues in the village teashop. "As I often go to Kashipur village, about 1 km from

97


my home, and discuss with other farmers in the teashop, it would be very good to have a copy of their village soil fertility map in there," said Nonishil, a farmer extension agent from Nischintapur who visits his extension village each month. This idea was suggested by one of the authors (Harun-Ar-Rashid) and Nonishil immediately agreed. Later on, we thought of giving laminated village soil maps with fertiliser recommendations to local fertiliser retailers, to display in their shops. Create incentives for farmer extension agents

Getting the balance right. By observing farmer-led demonstration plots in Northeast Bangladesh, many farmers started to apply phosphorous, potassium and sulphur to their rice field, not just urea. In more lowland areas, characterised by excessive fertiliser use, experiments led to reduced inputs.

Training and capacity building are the initial incentives for farmer extension agents, while later increased social recognition becomes more important. However, recognition by peers is harder to achieve for farmer extension agents who work in a neighbouring village. Nonishil stressed that his social recognition would be enhanced by writing his name and address on the village map that is displayed in the teashop of his extension village. In this way, everybody would know him and know where to seek advice. Despite this, farmer extension agents would need a financial incentive to initiate activities in new villages. If all farmer extension agents could be trained as seed producers, they could also use their enhanced social standing to make a bit of money, by selling quality seed, as is already taking place in those villages where AAS trained farmer seed producers.

KEYS FOR SUCCESS Farmers learnt to better observe their crop to assess crop fertiliser requirements. These acquired skills also helped them to monitor soil fertility over time. Preparing village soil fertility maps with poor farmers created great confidence among them and scientists. This was further enhanced when we realised that their knowledge of soil fertility was confirmed by laboratory tests and the results of field experiments. Their confidence got another boost when, in a village workshop, they presented their soil fertility maps, yield data of their experimental plots, and fertiliser advice in front of researchers, DAE and NGO staff, and private sector input suppliers. Pride, social recognition, the desire to escape poverty, and solidarity make this farmer-tofarmer extension work. Poor farmers more readily accept new information from their peers, working as extension agents.

98


Block supervisors can gain a lot of credit from the community by listening to them and building on their local knowledge. Training block supervisors through hands-on exercises helps to institutionalise this innovative approach in DAE. Block supervisors can easily apply this method if they collaborate with farmers who work the land and know the 'language of their soil'. With these local experts, and by building on previous experiments and topographical similarities, soil tests are not needed in each new village.

DIFFICULTIES, RISKS AND ASSUMPTIONS Initially, the participatory experiments were divided into two plots, one that followed the farmer's practice and the other with the suggestions based on the soil samples and consensus building exercise. Some farmers took it as a competition and started adding extra fertiliser to their 'farmer' plot, masking potential impact of the experiment. From then onwards, we decided to have a whole field under improved practice and compare this with neighbouring fields of non-participating farmers. So non-participants became part of the experiment without knowing it. It is an important lesson. As Graham Thiele and colleagues point out, a farmer field school or any other method should not try to force through a pre-determined result (Thiele et al., 2001). Because the researcher's technology may not always 'win', especially if farmers treat the experiment like a contest, and stack the deck in their own favour. We believe that those farmer extension agents who have been equally trained as seed producers have a vested interest in providing multiple services to the community. Others will need a small stipend in future to keep up the motivation to train new villages, organise meetings and field days, and monitor field activities.

SCALING UP In Habiganj and Moulvibazar districts in northeastern Bangladesh, up to 4,000 poor farmers are now applying balanced doses of nutrients in 216 villages. After attending a farmers' workshop, the Integrated Crop Management sub-project under PETRRA started preparing village soil fertility maps in Kurigram district in northwestern Bangladesh. Although the method was developed in piedmont soils, it can be replicated in any soil type after initial backstopping by soil tests. Nationwide adoption of the method will depend largely on the understanding of the method and acceptance by the senior DAE management. Many DAE block supervisors, upazila and district level officials already appreciate this new approach and have implemented it in 19 more villages in Srimongal and Habiganj districts. Potentially, each block supervisor, once trained by AAS and BRRI, could supervise 1099


15 farmer extension agents, to act as local group coordinators and resource persons. After the training course organised for block supervisors, one of them said: "Poor farmers rarely adopt our fertiliser recommendations, so it is very important for us to learn this new method. With some initial training, it is not so difficult to do and the maps can be drawn in a new village within only two to three hours. Once recommendations are known for a given village, only one trial in each fertility grade is needed to validate or fine-tune them in neighbouring villages."

How green is green? Integrating village soil fertility maps with other visual tools such as the leaf colour chart would further improve accuracy of timing and reduce dosage of fertiliser applications. In some villages, AAS initiated both approaches side-byside and farmers are eagerly observing each others' experiments.

100

When asked how the new method could benefit national agencies, Mr. Ferdous from AAS said: "At the national level a lot of money can be saved and soil testing can become more accurate if use is made of participatory soil fertility mapping." The soil testing service provided at national level by the Soil Resource Development Institute, the National Agricultural Research Services, and some NGOs have gone to a great effort to make site-specific fertiliser recommendations, but they can reach only a few farms. Both for logistic and technical reasons Hugh Brammer, long-term soil expert in Bangladesh, also invariably advices against the use of soil test kits and suggests to rely on simple fertiliser trials in farmers' fields (Brammer, 2002). The methods described in this chapter add to a number of other methods being tested across the world, and after having read this case study, scientists from the International Centre of Tropical Agriculture (CIAT) expressed their intention to explore its potential for scaling-up site-specific soil fertility management in East Africa (Vanlauwe, personal communication).

CONCLUSION Soil fertility knowledge and management is determined by a complex set of ecological dynamics, socio-cultural factors, institutional arrangements and policies of various sorts (Kerven et al., 1995; Scoones, 2001; Brammer, 2002). Responding to the non-adoption of blue-print recommendations by farmers, Defoer and Budelman (2000) compiled a resource guide for participatory learning and action research on soil fertility in the tropics. But the senior author of this work also admitted that resource flow maps and nutrient balance calculations are complex, and as such limiting their scope for being scaled up (Defoer, 2000). The general lack of ready to use methods for teaching many beneficiaries about soil fertility management was further emphasised by Snapp and Heong (2003).


The adaptive method described in this chapter builds on farmers' soil fertility knowledge to grade soils, and is relatively simple and easy to learn. Farmers draw a village soil fertility map as a starting point to develop and test improved locallyspecific fertiliser guidelines. Farmer extension agents help other farmers to understand the recommendations, to test these in their own fields and modify them if needed. The method also offers the government extension system a great opportunity to appreciate local knowledge more. Coordinated by DAE staff, farmer extension agents may work on many topics and increase the overall impact at the community level.

REFERENCES Bellon, M. R. (2001) Participatory Research Methods for Technology Evaluation: A Manual for Scientists Working with Farmers. CIMMYT, Mexico DF. Brammer, H. (2002) How to Help Small Farmers in Bangladesh. The University Press Limited, Dhaka. Defoer, T. (2000) Methodology on the move: case studies from Mali and Kenya on methodology development for improved soil fertility management.. In: Guijt, I., Berdegue, J. A., Loevinsohn, M. and Hall, F. (eds) Deepening the Basis of Rural Resource Management. RIMISP and ISNAR, 56-72. Defoer, T. and Budelman, A. (eds) (2000) Managing Soil Fertility in the Tropics. A Resource Guide for Participatory Learning and Action Research. Royal Tropical Institute (KIT), Amsterdam. Gupta, A. K. and IDS Workshop (1989) Maps drawn by farmers and extensionists. In: Chambers, R., Pacey, A. and Thrupp, L. A. (eds) Farmer First. Farmer Innovation and Agricultural Research. Intermediate Technology Publications, London, pp. 86-92. Kanté, S. and Defoer, T. (1994) How farmers classify and manage their land. IIED Issue Paper 51 Kerven, C., Dolva, H. and Renna, R. (1995) Indigenous soil classification systems in northern Zambia. In: Warren, D. M., Slikkerveer, L. J. and Brokensha, D. (eds) The Cultural Dimension of Development: Indigenous Knowledge Systems. Intermediate Technology Publications, London, pp. 82-87. Murage, E. W., Karanja, N. K., Smithson, P. C. and Woomer, P. L. (2000) Diagnostic indicators of soil quality in productive and non-productive smallholders' fields of Kenya's Central Highlands. Agriculture, Ecosystems & Environment 79, pp. 1-8. Saleque, M. A. (2004) An alternative method of soil phosphorus analysis for rice growing acid soils. BRRI Seminar Paper, April 1 2004. Gazipur, Bangladesh.

101


Scoones, I. (ed.) (2001) Dynamics and Diversity: Soil Fertility and Farming Livelihoods in Africa. Earthscan Publications, London. Sillitoe, P. (2000) Cultivating indigenous knowledge on Bangladesh soil: An essay in definition. In: Sillitoe, P. (ed) Indigenous Knowledge Development in Bangladesh. Present and Future. Intermediate Technology Publications, London, pp. 145-160. Snapp, S. and Heong, K. L. (2003) Scaling up and out. In: Pound, B., Snapp, S., McDougall, C. and Braun, A. (eds) Managing Natural Resources for Sustainable Livelihoods: Uniting Science and Participation. Earthscan Publications, London, pp. 67-87. Thiele, G., Nelson, R., Ortiz, O. and Sherwood, S. (2001) Participatory research and training: Ten lessons from the farmer field schools (FFS) in the Andes. Currents 27, pp. 4-11. 1The partner organisations involved in this study were Mac Bangladesh, Prantik, Nishchitapur Krishak Samabai Samity (NKSS), Uttar Varaura Bahumukhi Krishok Samabay Samity (UVBKSS), Association for Socio- Economic Development (ASED), Social welfare Advancement Brilliant Association (SABA), Pragoti Samaj Unnayan Sangstha (PSUS), Madhabpur Bahumukhi Nari Mukti Sangstha (MBNMS), Bahubal Agrani Samaj-Kalyan Sangstha (BASS), Bangladesh Association For Social Advancement (BASA), Rural Agriculture and Social Development of Bangladesh (RASD) and Manab Kalyan Sangstha (MKS).

102

9 Going Public

A quick way to interact with communities

Paula Nash Paul Van Mele

SUMMARY Going Public is an extension method that trains farmers in public places. Because people are already gathered, at a market or bus stop for example, Going Public uses less staff time and transport costs than a trip to a remote village. Many who attend sessions have never had contact with an extension worker before - and never will because of the impracticality of visiting every individual farm family. The increased access of extensionists to farmers is the essence of Going Public. It is a novel approach to complement other extension methods, not replace them. It is not a substitute for field-based learning and consultations, but it can make other extension efforts more effective and provides a unique entry point to rural communities. With simple, advanced planning, Going Public can be implemented by a confident facilitator who is familiar with local conditions. Planning begins by choosing a simple and short learning topic that is relevant to a general audience. A Going Public session might last only two hours, although each one is unique; the topic is repeated as new batches of passers-by arrive. Going Public allows one to learn about local farm knowledge, triangulate farmers' needs assessments and identify villages where services are needed most urgently. In this chapter, we describe how Going Public was used to learn with farmers about seed health, and pay special attention to facilitation. 103


ACTORS AND NETWORKS The experiences presented here are based on Paula Nash's field work for her MSc at Reading University. Paul Van Mele, who tested Going Public previously in Bangladesh, would help to select suitable partners and advise if needed. Both authors are staff from CABI Bioscience, the scientific division of CAB International or CABI, an intergovernmental organisation with expertise in sustainable agriculture, knowledge and information systems, and farmer education. We chose hard working organisations with different organisational styles, that were involved in PETTRA sub-projects and were enthusiastic about taking part (Nash, 2003). The Rural Development Academy (RDA) in Bogra had previous experience with Going Public, under the Seed Health Improvement sub-project (SHIP), when it was used at a village road junction to get community feedback on multi-purpose seed drying tables. The second time in 2002, Going Public addressed the rice brown spot disease Bipolaris oryzae, its management through seed sorting, and improved drying. The event took place at a weekly hat or market (Van Mele and Zakaria, 2004). The second organisation selected was the Bangladesh Rice Research Institute (BRRI). Since 2000, SHIP had conducted on-farm research and developed technologies with farmers to improve the quality of their farm-saved seed. As time went on, more emphasis was put on participatory learning. Therefore, Dr. Taher Mia, head of the Department of Plant Pathology at BRRI and national coordinator of SHIP, was enthusiastic about testing the Going Public method. In 2001, he had learnt about participatory research during a 3-week study visit to the UK, organised by CABI Bioscience. Working closely with farmers has been on his agenda ever since, so he fully supported his project staff to help implement these activities. The third organisation selected was an NGO, namely the Agricultural Advisory Society (AAS), with a staff of dynamic young men and women. AAS was unfamiliar with Going Public, but wanted to be involved. Their organisation and working philosophy is described in Chapters 3, 4, 8 and 18.

EVOLUTION OF THE METHOD Conventional agricultural training methods rely on face-to-face communication, specifically for teaching about pest and disease management. Nevertheless, the majority of farmers still have not been reached. To reach many farmers, but maintain the quality of face-to-face communication, we explored novel ways of applying these principles. Going Public started in Bolivia where CABI was developing extension exercises for farmer field schools (Bentley et al., 2003). Politicians, preachers and salesmen have attracted audiences in public places for centuries, so Bentley and colleagues wondered if extensionists could communicate plant health messages to crowds in 104

Going Public

market places. To test the idea, extensionist Juan Almanza led the first Going Public at Tiraque market, Cochabamba, Bolivia, in December 2001. People gathered as Mr. Almanza set up his 'market stall', unloading microscopes and simple tools for showing how to extract nematodes from the soil. Mr. Almanza told farmers that nematodes existed, showed a simple method to extract nematode cysts from the soil with a glass of water and a piece of newspaper, and gave advice about controlling nematodes. Since the first Going Public session in December 2001, other events have been held in Asia and Africa (Table 9.1).

THE GOING PUBLIC METHOD Figure 9.1 is a guideline for organising a Going Public event. We use plant health as the learning topic, but obviously this can be interchanged with any other topic. Further advice is given by Bentley et al. (2003). Identify the problems Identify the main plant health problems in the region. Local extension agents are good sources of information as they have regular contact with farming communities (Boa et al., 2001). Using topics that farmers care about engages their interest and stimulates discussion. Select key learning topics Once extensionists and farmers have been consulted, the plant health problem needs to be diagnosed and a short, accurate learning topic prepared. Because of their past experience, RDA and BRRI chose rice seed sorting, drying and storage as topics to address at the Going Public.

Discovery learning exercises for integrated pest and crop management are mostly developed for farmer field schools, but can be equally used in videos or Going Public sessions.

As two partners had selected post-harvest issues, our first concern was how to host a Going Public for women. "You may get many men at the market to come and see Going Public, but women don't come to the market place. We also have no prior experience of Going Public, so how will you bring women to the market place? My worry is that Going Public may not be able to address such issues as storage [of rice seed]," said Mr. Harun-Ar-Rashid, executive director of AAS. Because seed drying and storage is done by women, AAS considered this an inappropriate topic to take to the market place. Instead, they chose to address rice bakanae disease, caused by Fusarium moniliforme, because this would be of more relevance for men coming to the market. Later, we learnt to Go Public in places where we could reach women.

105


Table 9.1 Overview of Going Public events in various CABI projects DATE

PLACE

CROP

FOCUS OF ACTIVITIES

December 2001

Tiraque market, Bolivia

Potato

How to identify nematode symptoms on roots and test soil for presence of cysts

December 2001

Sucre public truck stop, Bolivia

Peach

How to distinguish peach leaf curl from aphid damage, and discuss integrated pest management

January 2002

Maria village road intersection, Bangladesh

Rice

Demonstrate seed drying tables developed with one community, and obtain comments on their wider appeal

April 2002

Noi Mile market, Bogra, Bangladesh

Rice

How to recognise brown spot disease, seed sorting and improved drying

September 2002

Kalimantan, Indonesia

Pepper

Describe general diseases, introduce a new problem recently occurring and how to manage it

February 2003

Ferry stop and markets in My Tho, South Vietnam

Fruit trees

Demonstrate diseases of mango and control options

February 2003

Thai Nguyen, North Vietnam

Pineapple

Demonstrate diseases, discuss control options

May 2003

Sylhet, Bangladesh

Rice

Bakanae disease in rice; what it is, how to recognise and how to control it.

May 2003

Mouna market, Gazipur, and Maria market, Bogra

Rice

Improvement to rice production. Rice seed sorting, drying and storage

September 2003

Tiraque, Bolivia

Potato

Andean potato weevil, seed health and other topics

March 2004

Kamuskono market, Sironko, Uganda

Banana

Raise awareness about bacterial wilt disease by discussing symptoms and control; gather local knowledge on disease and learn local names

April 2004

Weekly fair in Los Negros, Bolivia

High value horticultural produce

Hand out flyers, show samples of key diseases, collect samples for laboratory referral and provide advice on control

The right person to lead the event The extensionists, researchers, or local people leading the Going Public should have a good agronomic background to be able to explain the plant health problem, besides attracting a crowd and maintain a lively dialogue. The three partner organisations had different personalities heading the events. AAS used researchers to host Going Public events, whereas RDA chose farmers trained under the SHIP project. Both types of leaders were appropriate. The person leading the event will affect the types of people that attend Going Public. A lively character can be more 106

Going Public

Identification of problem Informal chats and interviews with farmers and extensionists to identify plant health problems Selection of topic Determine the focus of the Going Public learning topic Preparation and planning Facilitator Who will facilitate session and why are they being selected? How will they influence what type of people attend? What advice to give and what to do if you can't answer questions? Audience Who is expected to attend and will that make a difference to content of learning topic? Location Does permission need to be sought to use public area? How location may affect attendees- minority groups etc.? Demonstration materials Are they relevant, useful, easily transported? Can location provide any useful materials? Collect and prepare plant samples, leaflets, colour photographs Going Public event What types and categories of people are present? How many people are present? How far have they come? What questions are being asked? If there isn't an immediate solution, what can I do? What other crop problems have been identified? Reflection on the event Has the correct site been chosen? How would I do things differently? What was the reaction of the crowd? How was the crowd interacting with each other? How did the crowd interact with the person presenting?

Figure 9.1 Steps of Going Public

107


effective at attracting people, maintaining their attention, and starting dialogue. Yet shy members of the public may feel more comfortable listening to a relaxed person. How to choose local innovators is described by IIRR (1996) and Van Mele and Zakaria (2002), but anyone who is comfortable speaking in public, and who has something to say can Go Public. "We engaged farmers for Going Public, as they can do a much better job than researchers. Their conversation is much more natural," said Mr. A. K. M. Zakaria from RDA. Using farmers to conduct Going Public events is likely to be easier when a working relationship has already been established. Whoever hosts the Going Public session should have an idea of the likely questions that the farmers may ask and the questions that should be asked to the farmers. Find the right location Choose a venue in advance and ask permission from the relevant authorities. Selecting the right place is important. For example, in Bolivia, markets are open to women and men, but villages are less receptive to outsiders, the houses are dispersed and it is difficult to attract a crowd in a village. On the other hand, in Bangladesh, women rarely attend markets, but villages are densely populated and it is easy to draw a crowd there. So villages are good sites for Going Public in Bangladesh, while market towns are better in Bolivia. As women are by and large excluded from markets in Bangladesh, and we wanted their opinions on rice seed health, BRRI agronomists set up stalls in a village area where some women had already gathered. As word spread through the village, more women arrived to join in the discussion on rice seed drying and storing. Paula Nash noticed manure drying on a platform, and asked the women how they dried other

Photos of seed drying tables, which were made by families under the SHIP project, are exhibited in the village. Neighbouring women can see and comment on the technology without having to leave their village.

108

Going Public

materials, including rice. Once their curiosity was raised, we introduced the concept of drying tables. The agronomists showed seed flotation and women asked many questions. They shared with us their traditional seed storage practices. We spent about an hour and a half with women in the village and shared many ideas using few resources. Prepare demonstration materials Going Public can be made impractical by using large demonstration boards and a lot of poorly thought out demonstration materials that are heavy, require a car to carry them, or serve no purpose. To keep costs low, use portable, practical materials. For AAS's first Going Public in Sundarpur market, they took tables and benches so Nash could interview farmers for her thesis (Nash, 2003), although the experience taught us that taking clipboards and sitting on nearby walls for the interviews works just as well. BRRI took a plastic bucket to the Going Public session, filled it with water and demonstrated seed flotation to remove insect-infested and partially filled rice seeds. They also took a seed drying table to Mauna market, but then the extensionists realised that the table was a handy place to show off other demonstration materials. The table became so cluttered with other things that the audience could hardly see that it was a drying table. Project farmers who had already made drying tables tried to explain their many uses, but as they were hidden beneath irrelevant items, people could not see for themselves what the tables were good for. The farmers could have spread rice seed on the tables, and lifted them up to show people how the table makes it easy to bring seed in out of the rain. The farmers could have shown the audience the tables' other uses, like storing blankets or pans; they could have sat or lain on one of the solid drying-threshing tables to show people that it can serve as furniture. This would have stimulated the audience to ask questions. Videos can be better than farmer-to-farmer extension just for these reasons (see Chapter 7).

During a Going Public session, Shafique from RDA and Zabed Ali, one of the trained farmers, use a calendar to teach passers-by when to do which activity in order to improve the quality of their farm-saved seed.

Apart from showing three drying tables of different designs, cluttered up with other demonstration aids, RDA used other props to good effect, including different types of storage pots for rice, trays of rice seedlings grown from both hand sorted and unsorted rice seeds. The staff and project farmers of RDA used lots of information boards at Maria village. Some of the boards were placed at ground level so people could not see them well, especially when people crowded around. Agronomist Shafiul Karim (Shafique) 109


explained the rice calendar, showing the best time to practise improved rice tasks. The calendar, which was well placed for the crowd to see, provided a condensed view of complex information in an easy to understand form. Use materials that can be found locally and arrange them in ways that allow many people to see them at once, or that can be easily handed around to the crowd. Photographs or drawings help to stimulate questions and allow more people to take part. But real live materials are even better than photos. Samples of healthy and diseased rice seed have been shown beside healthy and infected seedlings. The Going Public event The facilitator(s) interact with the crowd and ensures that all topics are addressed and that education materials are wisely used. They should try to engage people on the outskirts of the crowd. When the event is in full swing, record some key data to give an estimate of where people are from, what types and roughly how many people are attending. Assign one person from the team to record these data and write down the questions people ask or interesting discussions that emerge as in Box 9.1. Asking the farmers in the crowd about their pests and diseases helps learn about local problems and can help choose crop health topics for future Going Public exercises. It also allows to triangulate information gathered through other means. BRRI staff learnt that farmers in Gazipur mainly referred to bakanae as sat, meaning off type or mixture of different varieties (as bakanae infested plants are clearly elongated), or as pata sada, meaning white leaf (leaves of infested plants turn pale). AAS staff learnt of many different local names that farmers used to describe bakanae disease (see Table 9.2) and as Mr. Ferdous said: "Although farmers did not know it was a disease, they had a good knowledge of the symptoms." Farmers also mentioned that they had had the disease for a number of years; some commented that their yields were steadily decreasing. Although the names and interpretation given by farmers has potential, it would need more work to unveil their local knowledge on diseases. Reflection Analyse what went well at the Going Public, and what could have been better, i.e. Box 9.1 Once Upon a Time in a Market

110

April 25, 2002. "These seeds look very bright and give such strong seedlings, which new variety is this?" wonders one of the farmers looking at the materials on display at noi mile hat, the weekly market near Maria village in Bogra. Mr. Zabed Ali, one of the farmers trained under SHIP, tells him it is one and the same variety. Those who have gathered look puzzled; some even think that a machine has polished the seed. "Come to my village, we can provide some healthy seed to you, or we can show you how to produce them yourselves. And let me assure you, we do not have a machine in our house," reassures Zabed. (See Chapters 16-20).

Going Public

Table 9.2 Local names for bakanae given by farmers when shown infested plants at a Going Public session, Sundarpur, 2003 (n=23) LOCAL NAMES

INTERPRETATION

NO. OF RESPONSES

Sat

Off type, a mixture of different varieties

19

Hinnua/Hinno

Taller and yellow plant in a hill

6

Laim/lam

Off type, a mixture of different varieties

5

Baoni

Rice plant with leaf rot at top that goes downward

2

Bao

Bad air. Also given to human diseases thought to be caused by bao; sometimes related to evil spirits

2

Monna

A kind of weed

2

Hinnua dhan

A rice plant which has nodes like chingri shrimp

2

Dhan vat kai

Pale or yellowish rice

1

Noa dhan

New rice

1

was it set up in the right spot, was it too noisy, were there too many distractions, how was the learning environment affected by the location? Learn lessons from documented Going Public events (Bentley et al., 2003; Nash, 2003; Van Mele and Zakaria, 2004). We encouraged the facilitators to write their experience in a lively narrative immediately after the event. This is a relaxed way to let people document their views and reactions. Short quotes should be recorded to document reactions, local knowledge and comments of people in the crowd. These could be used to clarify future action points, or be combined with the narratives into short stories, as we did in Box 9.1.

KEYS TO SUCCESS Keep it simple. Have a well-planned Going Public with a clear learning topic, the right amount of demonstration materials, and a facilitator who knows the topic and is comfortable speaking in public. Be creative. Think of possible problems and concerns before doing a Going Public, and come up with answers to solve them before they occur. Few resources are needed. One or two extensionists, transport, some demonstration materials, a few good ideas and some time are all it takes to Go Public. Farmers' time is not disrupted. One farmer remarked: "This system is very good, and we can learn a lot without any fixed schedule." He appreciated not being taken from his farm work by an extension programme with a lengthy agenda. 111


DIFFICULTIES, RISKS AND ASSUMPTIONS A main initial concern of partners was that women would not be reached using this method, as people only thought about markets. Go Public at places where women do gather: schools, NGO meetings, health centres and so on, as this reduces transaction costs (see Chapter 21). An agronomist raised the point that "Going Public is no substitute for field work". Indeed, some topics like village soil fertility mapping, as described in Chapter 8, are better done in the field, or in more controlled meetings than in a Going Public session. One characteristic of Going Public is that the facilitator has no control over who comes and who does not. Effectiveness of this new extension method is hard to measure as the audience may come from various places and differ every time. As with any other extension method, researchers will need to measure the impact of Going Public, which poses extra challenges due to the fluidity of the audience. See Nash (2003) for suggestions on how to study adoption rates. Research on extension methods requires detailed record keeping during events. This will not only tell us the type of questions farmers asked or what names they used to describe certain insect pests or diseases, it will also serve as a basis for future reflection.

SCALING UP Going Public is a flexible method that has been used in several countries, for various topics, and with diverse institutions, facilitators and audiences, making it a proven, transferable method. With ingenuity and planning even complex ideas can be communicated. After Van Mele presented the method during a national workshop on uptake in April 2004, organised at the premises of the Department of Agricultural Extension (DAE) in Dhaka, Syngenta expressed interest in using Going Public to promote its latest herbicides in Bangladesh, but so far hasn't undertaken action. Also the NGO Rangpur Dinajpur Rural Service (RDRS) showed interest, and some months later their management decided to start using Going Public in their agricultural programme from 2005 onwards. RDRS has a formal agreement with the DAE through the Northwest focal area forum, so it is likely the government extension service will also gradually learn about the potential of Going Public. In the future, Going Public could be combined with mass media and face-to-face extension in the following ways: Advise farmers where and when people can meet extensionists to ask questions and get

112

Going Public

materials like seed, leaf colour charts or leaflets with information on safe herbicide use. This can be done over the radio, TV, newspaper or through miking (see photo) Invite men and women farmers who have been trained by a project or a farmer field school (FFS) to interact with the audience at fairs, markets, village meetings, and other events Validate short, five-minute discovery learning exercises in a farmer field school and then use them at a Going Public.

CONCLUSION People attended the Going Public sessions in the markets from distances as far away as seven kilometres. Each organisation approached Going Public with different views, ideas and styles. Sessions differed in style of presentation: one was like a demonstration, and others were more like facilitated discussions. Either way, farmers shared information with each other, asked questions, picked up new ideas and learnt something about plant health management. The sessions also opened the eyes of those organising the event: they learnt about the names farmers use to describe certain plant health problems and how to better communicate with farmers in the future. One of Going Public's major advantages is that it is a quick and flexible extension method that can be used by anybody. Extensionists who want to use it, will develop their own style, noting the two-way flow of ideas.

Miking is a popular way to get messages across in rural Bangladesh. Rickshaws move from village to village and announce events through a microphone. It is an excellent means to announce Going Public sessions, field days or the promotion of new technologies, such as mobile pumps (see Chapter 13).

REFERENCES Bentley, J., Boa, E., Van Mele, P., Almanza, J., Vasquez, D. and Eguino, S. (2003) Going Public: a new extension method. International Journal of Agricultural Sustainability 1(2), 108-123. Boa, E., Bentley, J. W. and Stonehouse, J. (2001) Standing on all three legs: the técnico as a cross-cultural occupational group. Economic Botany 55(3), 363-369. IIRR (1996) Recording and Using Indigenous Knowledge: A Manual. IIRR, Cavite, Philippines. Nash, P. (2003) Comparing Conventional and Novel Extension Activities for Seed Health Improvement in Bangladesh. MSc thesis, Reading University, UK.

113


Van Mele, P. and Zakaria, A. K. M. (2002) The Innovation Tree: a new PRA tool to reveal the innovation adoption and diffusion process. PLA Notes 45, 54-58. Van Mele, P. and Zakaria, A. K. M. (2004) From concept to impact: developing and communicating multipurpose seed drying tables in Bangladesh. In: Participatory Research and Development for Sustainable Agriculture and Natural Resource Management. A Sourcebook. Vol. 3 A. CIP/UPWARD Philippines, in press.

114

10 Picture Songs

Jeffery W. Bentley Mostafa Nuruzzaman Qazi Wadud Nawaz Md. Rafiqul Haque

SUMMARY Picture songs were invented by a Bangladeshi NGO, Shushilan, during the PETRRA project. Although rooted in traditional song, dance and painting, the songs are new genre, an innovative way of teaching. Combined with conventional extension, the songs motivate people to adopt new technologies such as modern rice varieties, fewer agro-chemicals and more organic fertiliser, which some farmers said helped them to double their yields from about 2 ½ ton per hectare to over 5, while lowering their costs. The songs invite people to bring in samples of pests, diseases, soil and water to the plant health lab. Shushilan encourages women to speak in meetings, to run demonstration farms and to work alongside men on their own land. New information reaches large numbers of women through cultural shows that are a socially accepted form of entertainment. Songs are one of the few ways that people will learn and repeat a message by heart, and enjoy it. If you want people to remember something for the rest of their lives, teach them to sing it. 115


ACTORS AND NETWORKS Shushilan means 'good practice', and it is the name of an NGO in the far southwest of Bangladesh, just east of the Indian border and north of the Sundarbans, the 'beautiful forest' of mangroves, shrimp and royal Bengal tigers. Shushilan teaches communities to respect the rights of women, the poor, and the marginalised. Shushilan has a long and rich tradition of using drama, but until they started to work with the PETRRA project in 2000, they had no agricultural programme, and no experience in agricultural extension. Shushilan now does a lot of work in agriculture and has 130 staff members, up from 20 in 1997. They now have projects with CARE Bangladesh, Concern, and other large, international NGOs. Shushilan organised five farmer co-operatives and works with 150 farmers' clubs in Kaliganj and Shyamnagor upazilas, in Satkhira district. The clubs were started in the 1970s by villagers to deal with sports and local problems. The clubs are deeply rooted in rural society and are receptive to agricultural extension. Recently, the Department of Agricultural Extension (DAE) started working with smallholder farmers around Kaliganj to produce quality seed as part of the rice seed network (see Chapter 17). The Bangladesh Rice Research Institute (BRRI) breeds the new rice varieties and develops many of the technologies that Shushilan extends. Shushilan is proud of its relationship with BRRI and the DAE.

EVOLUTION OF THE METHOD A touch of romance Troupes of folk actors once roamed the Old World countries from Britain to China, putting on musical plays in small towns and villages. Bangladesh had several forms of travelling shows. Jari gan was singing and dancing, performed by a small troupe of four to six people. Leto was a musical by a lone troubadour, and jatra was a drama play, long and complex like a movie, with a cast of players. In the twentieth century, radio, movies, TV and videos killed these live, performing art forms.

Before Shushilan and PETRRA, perhaps no one had thought of using dance and picture songs in agricultural extension.

116

Wadud Nawaz, advisor of Shushilan, wanted to bring back the jari gan, while some people still remembered how to do it. There is perhaps a touch of romance in trying to save the knowledge of a complex task, of a thing that is too big and too alive to be kept in books. Performing a jari gan is like building a Fijian deep-sea canoe (Balick and Cox, 1996), or sailing from Hawaii to Tahiti by the stars (Finney, 1991). Knowledge of it must live in the heads of a crew, several

Picture Songs

people, who know how to act together. Once the skills are lost, they will be irretrievably gone. Nostalgia aside, there was another reason to try to revive the jari gan; it could be a way of teaching people, instead of just amusing them. Live musicals could reach women, because in Bangladesh the men often go to cinemas in town, but women rarely leave their own hamlet (see also Chapters 7 and 9 on video and Going Public). When the travelling shows stopped coming to the villages, a bit of innocent fun went out of the lives of the women. Art for teaching Ujir Hossain, Shushilan's cultural director, says he wanted to use folk songs for teaching since the mid 1990s. In July 2000, Shushilan bid for, and won, one of the first PETRRA sub-projects, which used folk songs as an extension method. A year after they had been using songs to teach new rice technologies, Mr. Hossain recalled some of the peep shows he had seen as a kid. A man would bring around a box and people would peer through an opening in it at amusing pictures inside. Mr. Hossain realised that if the pictures were much larger, a whole audience could see them at once. So he drew on another piece of Bengali artistic heritage: paintings on cloth. In 2001, the singers commissioned a local artist to paint them a pot. The BengaliEnglish Dictionary defines pot as: 1. cloth, screen, veil, canvas, garment. 2. painting, canvas, painted piece of cloth, picture (Ali et al., 1994). Shushilan calls their new genre a 'pot song', but here we call it a 'picture song' to avoid confusion with the English word 'pot'.

THE PICTURE SONG METHOD The singers on stage The barefoot dancers sing boisterously as they swagger into the room, circling the stage. They wear dark yellow uniforms, but some of them have a bright red sash tied around their heads. They play simple music on traditional instruments: an organ that slings over the neck, a drum and some cymbals. It is a catchy tune that the audience might find themselves humming as they walk home from the show. Two of the dancers are women, and one of them is dressed in a floor-length skirt and pants. Her eyes sparkle with enthusiasm as she sings and whirls around, stopping to reach out her arms to her audience, engaging them. Lija Hossain's voice has a good-natured urgency. Two of the dancers hold a large canvas, scrolled over two poles. They keep singing 117


even as they set the scroll upright. Smiling and swaying with the beat, they unroll a picture larger than a giant screen TV. The performers are part of Shushilan's Cultural Department. Some of them are full-time employees, but some of them are part time. One of the singers is a barber and another has a village shop. The stage is a meeting room with a cement floor, under a roof, but open to the breeze on three sides. It is part of Shushilan's rural training centre, a teaching farm with cows and chickens, tree nurseries, a library and lab near the small town of Kaliganj. When they hear the music, the neighbours drift in to see the show that they must have all seen before. And that's the point: it's fun. Handsome people belting out a contagious tune, rolling out a scroll of bright drawings, and lively dancing: it's supposed to draw people in and hold their attention, and it does. The troupe often performs upon demand of local organisations, their show taking place in the open air, or in large meeting rooms. Although Shushilan calls the performance a song, some parts are more like an opera. The picture song lasts for 45 minutes, half as long as a movie. And like a movie, the picture song has enough time to convey a lot of information. The message Shushilan promotes modern, high-yielding varieties (HYV) developed by BRRI and promoted by DAE. But the ballad urges listeners to use organic fertiliser such as cow dung and balanced chemical fertilisers (see also Chapter 8). The song tells farmers not to abuse insecticides, that insect pests have natural enemies: toads, ladybird beetles, and spiders. Later on, Liza Parvin sings about improved methods for preserving seed in a pot (see also Chapters 3 and 7). The multiple topics addressed indicate the cross-fertilisation that has taken place between various sub-projects. The song ends by inviting farmers to come to Shushilan with samples of crop pests,

118

Picture Songs

diseases, soil and water to see if their pond water is right for rearing shrimp. Shushilan will analyse the samples and give people a written recommendation on fertiliser, water, or how to control their rice pests. Shushilan keeps a database of the samples, the results, and the written recommendations. Plant health lab Shushilan never tried to use the picture songs as their only extension method. They combine songs with other genres. A picture song may motivate a farmer to take

The picture song encourages farmers to apply cow manure to their fields (left), and to apply urea three times and compound fertiliser once during the ricegrowing season (right).

These names look Bengali, but they are actually English. Among these beneficial animals, the insects are labelled (clockwise, from upper left) carabid beetle, ladybird beetle and mirid bug. Farmers have their own names for many insects, but learning these and their meaning is harder than it seems.

This detail urges the audience to thresh rice seed separately, and to store it in a plastic pot, or an earthen one, treated to absorb less moisture, and covered with plastic.

119


training courses from an extensionist in her village, visit demonstration plots, or Shushilan's agricultural service centre in Mahadpur village, where she has access to quality agricultural inputs and advice, take a soil, water, or plant health problem to their laboratory. Shushilan analysed 842 samples this year (221 of soil, 216 of water and 405 pests and diseases). The main rice plant health problems farmers brought in were badami dag (brown spot), khol pora ('sheath burn' i.e. sheath blight), pata pora ('leaf burn', bacterial leaf blight), mazra poka (stem borer) and gandhi poka (stink bug). Suriya Sultana (left) and Shahina Parvin (right) run a soil sample in the small lab in Shushilan's Agricltural Service Centre in Kaliganj district. Farmers want not only their soil and water analysed, but also want to learn which pests and diseases affect their crop and what to do about it.

While at the lab, farmers can read in the library. Despite it being a new initiative, the librarian Suriya Sultana says she has about 25 visitors per week. Bentley and Boa (2004) tell a story about another community-based laboratory or plant health clinic in Bolivia. Face-to-face method Picture songs are like Going Public (see Chapter 9) a face-to-face method for reaching large audiences, but the scope to interact with the audience is smaller. The troupe has performed its agricultural opera hundreds of times, and has created some on other topics: Agricultural technologies Gender Human rights and good governance Natural calamities including climate change Alternative dispute resolution Natural resource management Environment Shushilan emphasises three things: 1) reaching women farmers, 2) organised in authentic farmer clubs established years ago by the communities themselves, and 3) culturally appropriate extension. Writing, rehearsing, validating Shushilan decides the themes for their songs based on consultation with extension agents, farmers' co-ops and clubs and women's organisations. Shushilan's singer-

120

Picture Songs

songwriter Uzir Hossain grew up on a small farm, and although he went to college he is still in touch with his roots. After he writes a song, he teaches it to the troupe. They rehearse it and put it on for the leaders of Shushilan, some of the staff and friends and neighbours from nearby villages. The audience comments on the show; Mr. Hossain edits it, and the troupe starts performing it for the communities. In Chapter 5, also Van Mele and colleagues stress the need to thoroughly pre-test videos. Impact About 25,000 people have seen the picture songs. Shushilan has planted 501 demonstration farms with local people, including 347 women. We met with four farmers' clubs and asked the members what they had learnt from Shushilan. They said they had learnt to space their rice systematically, 18 by 20 cm, in lines, so they use about half as much seed as they used to. By analysing their soil they knew how much chemical fertiliser to use, so now they harvest more rice, using less fertiliser. They plant higher yielding rice varieties, especially BRRI dhan 28. They use less insecticide. So their yields have sometimes doubled and their costs have gone down. One group said that they harvested 8 or 10 mon per bhiga (between 2.2 and 2.9 ton per ha) before the project, but afterwards their rice yield rose to 18 or 20 mon per bhiga (between 5.2 and 5.7 ton per ha). When we asked them how they learnt these things, first they mentioned training in the villages by extension agents, followed by the picture songs in second or third place. The picture songs do not replace conventional extension, but they help to capture attention and reinforce messages. An extension programme could probably not be based only on dance numbers. But song and dance opens people's minds to a message that they can then learn in courses, demonstration plots, field days, videos and other methods.

One must be able to hold the public's attention: it can be jokes, riveting speeches, demonstrations or even a certain tone of tenderness and sincerity. Agriculture is performance (Richards, 1989), and so is extension.

In two of our meetings, men listened politely while thoughtful, articulate women did most of the talking. They said that until recently, women did not go out at all. Now they start going to the village market, and working in the rice fields, so the household saves money that would have been spent hiring labourers (see Box 10.1).

KEYS FOR SUCCESS Be entertaining

121


Have something to say Blend the old and the new Use realistic, naturalistic drawings. Box 10.1 Teach Your Husband Well

Box 10.1 Teach Your Husband WellFazila, wife of Rafiqul Islam of Gobindapur Chashi Samity, Kaliganj says, "I received training from Shushilan in early 2001. I tried to apply the technique in our 30 decimal (1,200 square meters) plot during boro (dry season). The variety was BRRI dhan 28. I planted 1-2 seedlings per hill instead of 5-8, which was what we used to plant. I also used the recommended dose of fertiliser, far less than what we were using. Influenced by some neighbours, my husband came home furious and started beating me, 'you _______, how dare you keep our whole family fasting all year long! You take the risk of planting so few seedlings and use less fertiliser; everybody says the crop will be a damn failure'. He drove me out of the house. I went straight to Shushilan and came back with Mr. Tapan Kumar Biswas, the agriculture supervisor and coordinator, Md. Mostafa Akhteruzzaman. Both of them tried to convince my husband, but he was unyielding. At last, he agreed to keep half of the plot under the new cropping system and the rest his way. After 20 days the story changed, and after harvest the yield turned out to be more than double: the production was 250 kg in the traditional system, but 550 kg in the improved one. My husband was regretful and he accepted the new techniques. Now both of us cultivate our land together, under the improved system." Gobindapur, Kaliganj, November 2003

DIFFICULTIES, RISKS AND ASSUMPTIONS Live entertainment in extension needs a message that matters and an institution that supports it with materials, transportation and salaries. The Shushilan songs have the right mix of message and music. The songs, dances and drawings have to be entertaining, but they have to have a message every step of the way, otherwise the singers are just competing with the TV. People everywhere inherit their past and then reinvent it. Shushilan took folk tunes, and old-time road shows, but added educational lyrics and big pictures. They haven't brought the jari gan back to life, but used it as inspiration for something new. Adding agriculture to their existing cultural expertise could be done at low cost.

SCALING UP Shushilan will keep working with farmers' clubs and co-operatives (see Box 10.2). Although Shushilan developed the picture song for PETRRA's Women-Led Cultural Extension sub-project, they now also perform picture songs for various other projects, such as fish and shrimp projects, supported by the World Fish Centre and others. The troupe is often asked to sing for NGOs like UTTARAN, Concern,

122

Picture Songs

PETRRA has made a lasting impact on Shushilan that goes way beyond the picture songs. Before PETRRA, Shushilan did not work in agriculture or with extension. Both of these are now major parts of their work. PETRRA activities helped them to get exposure to other donors and attract projects. In 2004, Shushilan employed 130 staff members, up from 20 in 1997.

Box 10.2 A Lasting Impact

IDEAL and Nokshikatha, with audiences of 1,500 or more in each show. The DAE also invites them to farm communities in Kaliganj. Shushilan now considers agriculture as an integral part of community development. Whenever they perform a show on social issues, they include the picture song with the theme on agricultural technologies.

CONCLUSION The first rule of extension is 'Thou shalt not be boring'. All extension must be interesting, even if it need not all be in dance numbers. A recent article in Natural History suggested that humans sing instinctively. People in all countries and all cultures sing: at celebrations, at sporting matches, in the shower, after waking up in a good mood. Learning is about repetition, like reciting the multiplication table over and over until we know it. But the more we repeat a message, the more boring it gets. Songs are one of the few ways that people will learn and repeat a message by heart, and enjoy it. If you want people to remember something for the rest of their lives, teach them to sing it.

REFERENCES Ali, M., Moniruzzaman, M., Tarque, J. and Rahman, L. (1994) Bengali-English Dictionary. Bangla Academy, Dhaka, Bangladesh. Balick, M. J. and Cox, P. A. (1996) Plants, People and Culture: The Science of Ethnobotany. Scientific American Library, New York. Bentley, J. W. and Boa, E. (2004) Community Plant Health Clinic: An Original Concept for Agriculture and Farm Families. CABI Bioscience, Egham, UK. Available at: http://www.jefferybentley.com/Report.htm Finney, B. (1991) Myth, experiment, and the reinvention of Polynesian voyaging. American Anthropologist 93(2), 383-404. Richards, P. (1989) Agriculture as a performance. In: Chambers, R., Pacey, A. and Thrupp, L. A. (eds) Farmer First: Farmer Innovation and Agricultural Research. Intermediate Technology Publications, London, pp. 39-43. 123

Part IV: Enterprise webs

11 The Enterprise Web Noel P. Magor

"We had better search for what the market can do for the poor, giving appropriate support for them to compete in these markets." Wood, 1994

INTRODUCTION In the research-extension continuum many good agricultural technologies are not adopted by farmers. At times, this may be due to inappropriateness of the technology, but often a misplaced emphasis on the process of developing the technology hinders an examination of the broader social and organisational reasons behind farmers' so-called rejection. There is more to extending a technology than giving the technical specifications to the extension department or other service providers. Drucker, a lead thinker in business organisation presents a more holistic view of a technology in his book Innovation and Entrepreneurship: Practice and Principles. He stresses that too often, the developer of an innovation only considers the specific technical requirements and ignores factors that are social, organisational, economic or perceptual, resulting in a lack of adoption or a failed enterprise (Drucker, 1985). In this section, we start with the assumption that a technology has satisfied the criteria of acceptability and suitability for farmers with limited resources, and then zoom in on the organisational aspects of technology delivery. This chapter introduces a visual analytical tool, the enterprise web, that can be used by service providers to unpack the many discrete activities that are undertaken for the introduction and dissemination of an innovation (Magor, 1996). It is followed by three case study chapters that illustrate the use of the enterprise web for three contrasting technologies. It may help readers in developing dissemination strategies for other innovations, whether agricultural or non-agricultural. 127


THE ENTERPRISE WEB: A HOLISTIC TOOL A generic example of an enterprise web Figure 11.1 illustrates an enterprise web, for which all discrete activities and causal links between activities have been identified. Developing one is an iterative process. The centre stage is given to the principal client. In our case, the clients are resourcepoor farmers, but could equally be a processing mill. All activities for a farmer to receive necessary knowledge and inputs to effectively adopt the technology are drawn above and all activities utilising the products are drawn below. The latter may also involve a new client for which a follow-on set of discrete activities is identified. The enterprise web is technology specific and adapted to the governance structure of the principle disseminating agency. The potential discrete activities, as depicted in Figure 11.1, are given in the paragraphs below.

Ecosystem identification Agriculture technologies are generally quite ecosystem specific. The suitability of a rice variety can be plot specific and depend on the cropping sequence. This activity involves researcher input and possibly discussion with local extension service experts. The service provider must have sufficient expertise to interpret the suitability of an ecosystem.

Selection of poor households Under the present political economy of Bangladesh, effort must be given to specifically identify poor households. The governmental Department of Agriculture Extension, for instance, could ask local NGOs to complete this task. For a large NGO, it may be the social development section that manages this activity and not the technical persons themselves.

Technical activities In a seed variety enterprise web the discrete activities may be: i) a government agency supplies breeder seed to a seed company; ii) this company grows foundation seed; iii) farmers grow truthfully labelled seed; iv) the company markets to local dealer; or v) a local dealer supplies seed to resource-poor farmers. For any enterprise, all activities can be listed and linked. In the generic example, two series of technical activities are given as 1a, 1b, 1c and 2a, 2b. Each of these is potentially performed by a different organisation.

Microfinance supply If a technology cannot be adopted without microfinance support, a series of activities related to this will need to be shown.

128

The Enterprise Web

Activities for input supply

Selection of suitable ecosystem

Selection of poor households

Technical Activity 1a

Technical Activity 2a

Technical Activity 1b

Technical Activity 2b

Technical Activity 1c

Efficiency boundary for hazard minimisation or transaction cost minimising

Provision of credit

Resource-poor men and women (client beneficiaries)

Activities for product utilisation and if needed link to next client

Home consumption

Sales of goods

Figure 11.1 Enterprise web with potential generic activities

Utilising the products This may simply be family consumption or a number of linked activities for marketing to urban or even international markets. As farmers participate more and more in the market economy, linking with a marketing client may be essential for the successful adoption of the technology. After visualising the discrete activities, an efficiency boundary can be drawn. All activities within the boundary are integrated and performed by the service agency. Activities outside the boundary are performed by another organisation.

129


The enterprise web as an explanatory and micro-analytical tool The enterprise web is relevant to understanding, planning and implementing initiatives to promote rural innovations in the following ways:

Beyond professional compartmentalism As a visual tool, the enterprise web is holistic. It enables professionals to step outside their disciplines. Soil scientists, agronomists, plant breeders, agronomists, microcredit specialists, development generalists and so on compartmentalise their knowledge, and they often lack capacity to incorporate non-technical knowledge. The glasses of knowledge of a technical profession restrict a person from seeing non-technical organisational and social issues as essential considerations in the extension of a new innovation. Even in research that strongly engages with farmers, subtleties in testing the technology have to be recognised in the future dissemination of that technology. For example, the scientist may not be aware of the types of support that are given in the very act of testing the technology with farmers. The scientist may hand-carry the seed of a promising new variety from his or her research station for testing. If the variety is found promising, the spread to neighbouring villages is still not guaranteed. The local seed dealer needs knowledge of the new variety and needs to know from where the seed can be procured. The enterprise web provides a focus that clearly shows all activities and links. It can act as a bridge for seeing all critical actors.

Identifying all activities and highlighting any potential weak links The very process of naming all activities provides opportunity for discussion. Drawing out visually each discrete activity parallels the activity chain used by Porter (1985) in the analysis of competitive advantage in business. A linchpin activity may be identified and special steps taken to ensure that it is managed effectively.

Analysis of organisational structure for managing weak links and minimising costs The enterprise web allows one to draw on business theory such as identifying core competencies, joint ventures or strategic alliances for dissemination (Jarillo, 1988, 1995; Thorelli, 1986; Prahalad and Hamel, 1990). In Bangladesh, discussion about which organisation does what can create conflict. The major service provider is the Department of Agriculture Extension and, more recently, donor-supported NGOs and to a minor extent the private sector. Plurality is increasing. Organisations need to look closely at their core competencies to determine exactly what they are best at doing and also what activities it may be better to implement in partnership. Organisational experience in this area is limited. The enterprise web helps to structure the decision-making as to who could best implement which activity. This application in micro level planning is illustrated in the case studies in Chapters 12 to 14.

130

The Enterprise Web

The enterprise web also allows the tools of transaction cost theory and governance structure to be used. See Box 11.1 for a layman's explanation of transaction cost theory. Transaction costs are different from production costs and measurable service delivery costs. For service providers, transaction costs include: time and expense to get up to date knowledge about a technology, to identify and establish links with farmers, to negotiate terms of service delivery with clients, and so on. Transaction costs have implications for the way service provision is organised. Depending on these costs, an actor may decide whether or not to deliver certain services to a particular client. High transaction costs resulting from limited access to inputs, markets or knowledge may also impede a poor farmer from adopting a technology.

Box 11.1 Transaction Cost Theory (drawing on Williamson 1979, 1981 1985, 1991 1996).

The NGO organisational culture in Bangladesh is to integrate activities and to do all things themselves. This approach has arisen from efforts to minimise risk in service delivery. Contracting between organisations and particularly between NGO and government has tended to be unreliable and inconsistent. Dependence on someone else may mean seed arrives too late or not at all. Governance is related to 'to do or not to do a particular activity'. If an organisation chooses to do an activity itself, this activity is controlled; hazard is minimised through vertical integration. If the organisation chooses not to perform the activity but to have it provided by another service provider, then it must give attention to the contractual relationship between them. The transaction cost for vertically integrating as opposed to a contract with another organisation must be analysed. By having all discrete activities visually shown in the enterprise web, the choice can be highlighted and critically discussed. In Figure 11.1, the service provider decided it could cover activities 1a, 1b and 1c itself, but that another organisation was better skilled to complete activities 2a and 2b. The boundary between what is and what isn't done by the organisation itself is indicated by the dotted line and is called the efficiency boundary. According to Williamson (1981), the arrangement minimises risk and is an economising decision for a business. It is also possible to use the enterprise web to identify the transaction costs faced by a farmer in adopting a new technology. For each activity that is not coordinated by the service provider, the farmer faces a corresponding risk and increase in transaction costs. If he or she needs to travel to several sources of knowledge and inputs to have all the necessary inputs for implementing the technology, risks of non-adoption go up rapidly. From this perspective, the one-stop shop has appeal for a farmer and can be visualised with the enterprise web. In Figure 11.1, the service provider decided it could cover activities 1a, 1b and 1c itself, but that another organisation was better skilled to complete activities 2a and 2b. The boundary between what is and what isn't done by the organisation itself is 131


indicated by the dotted line and is called the efficiency boundary. According to Williamson (1981), the arrangement minimises risk and is an economising decision for a business. It is also possible to use the enterprise web to identify the transaction costs faced by a farmer in adopting a new technology. For each activity that is not coordinated by the service provider, the farmer faces a corresponding risk and increase in transaction costs. If he or she needs to travel to several sources of knowledge and inputs to have all the necessary inputs for implementing the technology, risks of non-adoption go up rapidly. From this perspective, the one-stop shop has appeal for a farmer and can be visualised with the enterprise web.

The time farmers spend in training sessions, going to the market to sell their produce, and negotiating prices with the middleman all add up to their transaction costs. To get involved in the aromatic rice value chain, the number of separate transactions may be too great. A service provider, by consolidating some of the activities, effectively reduces the transaction costs of farmers.

Visualising complexity An innovation may be quite simple with a minimum of activities and no contracts between actors or it may be very complex. Figure 11.1 shows an activity 'sales of goods'. This could be expanded to link to a processing plant for which there are also a set of discrete activities. In turn, this may link to an independent export house with another set of discrete activities (Chapter 14). As the complexity of organisations increases, risk increases. A failure by the export house will jeopardise the adoption by farmers, not because the technology was inappropriate but due to market failure.

SELECTION AND DOCUMENTATION OF CASE STUDIES This book covers a wide range of extension methods and organisational models that emerged under the PETRRA project. Of the case studies presented in this book, three illustrate the versatility of the enterprise web. We present the practice of integrated rice-duck production, the manufacturing and marketing of the mobile pump and the production, processing and marketing of aromatic rice. 132

The Enterprise Web

The technologies and organisational requirements in the three case studies were quite different. Integrated rice-duck farming (Chapter 12), which was a new concept in Bangladesh, required training and changes in people's perception on the joint cultivation of rice and ducks. Critical linkages were needed to ensure ducklings at the right age and on time, and access to vaccines through government supply channels. For the mobile pump (Chapter 13), the development of local capacity to produce and market the pump was key to getting the pump to the farmers. For the production, processing and marketing of aromatic rice (Chapter 14), the emphasis was more on grower contracts, improving rice processing facilities to export standards, and marketing. Grower contracts required purposeful links with resourcepoor farmers; processing to standard necessitated mill modifications, and training; whereas marketing required drawing different millers together, analysing markets and ensuring quality. This last case study links to market chain analysis. A common activity for all three enterprises was the identification of resource-poor farm households, as per PETRRA's criteria.

Promoting rice-duck farming, mobile pump markets or an aromatic rice value chain has complex organisational requirements. To better understand these, along with the weak links in the dissemination model, a visual analytical tool called the enterprise web has been developed.

Various steps were used to develop the case studies: The editors of this book held a workshop with the three sub-projects to list discrete activities, link activities and identify indispensable steps. This did not occur spontaneously. Specifying all the linkages usually took a great deal of time. Principal investigators of the technology development sub-projects often did not distinguish between activities which were only required during the research phase and those required in the dissemination phase. We prepared a first draft enterprise web diagram, while one of us questioned and tested the validity of each activity. Based on the workshop outputs, we developed guidelines for authors that differed slightly from those used for other case studies in the book: sections would deal with actors, the enterprise web, keys for success, pitfalls, and a conclusion. The principal investigators prepared their first draft case studies, but still tended to 133


focus on the technical aspects of the innovation; hardly any reference was made to the enterprise web diagrams. The concept was foreign. The enterprise web was an add-on that made little sense to them. We rewrote the guidelines. One section was reserved for simply writing about the technology. The case study authors were then encouraged to write about each discrete activity presented in their enterprise web. This was then edited and further enriched through discussion and collecting additional information from farmers or other stakeholders. Only at this stage the principal investigators responded to the enterprise web as a useful tool for clarifying the steps required to extend their specific innovation. Visualisation coupled with writing was important for institutional learning. Integrated rice-duck production Travelling through the countryside you may occasionally see ducks in rice paddies, but it is relatively uncommon; the systematic practice of integrated rice-duck is new to Bangladesh and was introduced by some keen scientists who saw it in Japan. A specific output in the logical framework was on articulating the organisational requirements for extending the technology. Three organisations were involved in the technology development sub-project: the Bangladesh Rice Research Institute (BRRI) that provided leadership overall and expertise in rice; the NGO Friends in Village Development (FIVDB) that gave duck management expertise and had an extensive village network in the Northeast; and the NGO Bangladesh Development Services (BDS) that had a network of farmers in the tidal South-central region. The technology proved promising and generated a lot of enthusiasm concerning its potential. BRRI scientists and the duck technologist of FIVDB conducted good research with farmers in two potential ecosystems. However, both the BRRI and the FIVDB staff were very technically orientated and struggled to unpack the essential steps required for a service provider to extend the technology. The enterprise web came in as a useful tool around which to analyse discrete and essential activities (see Figure 12.1). Clearly, a link to the government livestock services is required, as they are the only agency in Bangladesh supplying the vaccine. Timely supply of ducklings of the right age also underpins rice-duck technology, opening up the discussion on who does what. Here, the NGO may be considered to be like a firm that needs to make strategic decisions (governance structure; Williamson, 1996 and Reve, 1990). To illustrate the different efficiency boundaries which may apply to the same technology, let us consider the two different NGOs involved in the project. The first is BDS that had no prior experience with ducks and only limited agriculture expertise, but it had a strong village network of poor households. Through BDS the households have access to microcredit. Should BDS establish its own hatchery? Should it be dependent on a government poultry farm for the supply of ducklings?

134

The Enterprise Web

Should it encourage a business man to establish a large duckling production unit? Or should it develop local hatcheries with farmers? Each of these questions relates to the boundaries that BDS sets for its own activities. BDS decided not to set up its own hatchery: the cost to establish and maintain one was too high and would make the overall cost to promote rice-duck unacceptable. On the other hand, the government poultry farm was unreliable in supplying ducklings and these inadequacies might have resulted in farmers' rejecting the technology. Helping local farmers to establish small-scale hatcheries seemed the most efficient method. It placed no extra transaction costs on the potential rice-duck farmer to access the technology, and being organisationally more efficient it reduced the costs for BDS. As they focus mainly on education, health and credit, they chose not to vertically integrate, but to outsource to local entrepreneurs, a choice relating to its core competencies (Prahalad and Hamel, 1990). For the second NGO FIVDB, the efficiency boundary was completely different. FIVDB pioneered the introduction of ducks as a viable stand-alone enterprise in Northeast Bangladesh in the mid 1970s. It has its own hatchery and has established numerous village level hatcheries in its target area. FIVDB can incorporate the extension of the rice-duck technology into its own programme with minimal transaction costs as it is building on existing expertise coupled with village networks. In fact, FIVDB has the opportunity to further expand the activities in the enterprise web: they could provide training and advice to other NGOs that, like BDS, may wish to take up the technology. The discussion around the enterprise web for integrated rice-duck shows that the extension strategy for a given technology is organisation dependent. Dr. Gazi Jashim Uddin Ahmed, sub-project leader and head of the Agronomy Division for BRRI suggested scaling up to a level that would link producers to Dhaka markets. This would introduce a new scenario of a joint venture between a commercial hatchery, an NGO working in a suitable ecosystem with poor farmers, and a marketing outlet. A joint venture may have the advantage of bringing together the social and organisational expertise of the NGO, the equity aspect of targeting marginal families, and the technical capacity of the business partner (Hennart, 1988; Kogut, 1988). From the farmers' perspective, anybody who wants to try out this new practice must have access to a knowledge source, ducklings of the right age at the right time, vaccine at the right time from the government and possibly a source of credit. Accessing each source independently results in high transaction costs for the farmer and may be so high that he or she does not take up the technology. For those members who were interested in rice-duck farming, BDS and FIVDB provided training in all aspects of the technology and removed the need for them to approach the government poultry farm and vaccine service, as such reducing the transaction costs for farmers.

135


Manufacturing and marketing mobile pumps International Development Enterprises (IDE) is a unique NGO in Bangladesh specialised in the marketing of mechanical, agricultural inputs. IDE had extensive experience in the marketing of the foot operated treadle pump, of which the mobile pump is a modification. A key objective of this sub-project was to fine-tune and document the extension method that included building public awareness, drawing in local NGOs for microcredit and for farmer selection, and developing a manufacturer-distributor-mechanic market chain. Producing, processing and marketing aromatic rice APEX is an NGO with private sector expertise and strong linkages to business leaders and senior public sector bureaucrats. In this sub-project they linked to Mark Industries, a private engineering firm and the Bangladesh Rice Exporters Association. The group of individuals, including supporting consultants, had a specific interest in seeing the marketing and exporting of aromatic rice become a reality. Aromatic rice is a traditional rice with new varieties now becoming available, and fetches a higher market price than common rice. Milling to export standard is a problem and affects export potential. The three principal clients for this series of linked technologies are farmers, millers and exporters, who each appears sequentially in the enterprise web. Figure 14.1 clearly shows that the aromatic rice value chain for an agro-processing commodity is still in an immature state. Value chains are the way the food commodity industry has developed globally (Kaplinsky, 2000). APEX is strong at national networking for promoting the development of the aromatic rice industry. Besides, it has an ability to build partnerships: APEX successfully drew in the expertise of IRRI in rice processing and motivated millers to invest in mill modifications. It kept the Bangladesh Rice Exporters Association informed of all developments and enabled the upgrading of the skills of its lead members through monitoring tours. APEX initiated the Aromatic Rice Export Clearinghouse (AREC), which could be an essential body for promoting export. Identification of markets, timely information to millers with contracts for growers and assurance of export standards are all critical to this enterprise. The high level linking to policy makers and concerned business entities seems to be the core competency of APEX. However, they do not have a network of poor farmers in the aromatic rice growing area. Using the enterprise web, APEX realised it faces fundamental choices as to who should manage the farmer contracts for supplying aromatic rice to millers. The transaction cost of developing farmer contracts, and ensuring their execution, may be kept low by developing the capacity of local NGOs to perform this function. In this way, it would not detract from the core competency of APEX. 136

The Enterprise Web

INSIGHTS FROM THE USE OF THE ENTERPRISE WEB Learning about process The enterprise web provided a tool for describing each activity, discussing processes and each partner's contribution. In the first session in which the enterprise web was introduced, it was met with a sense of indifference. Investigators in each of the three case studies found it easy to talk about the technology and its adoption, but not about the organisational dimension of the extension process. This was a recurring hurdle in most of the sub-projects on extension methods research. Partners lost themselves continuously in the detail of talking about the technology. A breakthrough in understanding came when they engaged in developing promotional posters on their extension method. Each produced an enterprise web in Bangla to represent their method. The visual nature of the enterprise web helped project staff to clarify essential activities. Listing of activities with coloured cards and then arranging these with causal links opened up the discussion. Writing on each discrete activity clarified weaknesses and any necessary remedial actions. Successive interactions enriched the case studies. For example, in marketing the mobile pump, the economic returns were included for all four clients: the manufacturer, the dealers, the mechanics and farmers. The enterprise web was only introduced towards the end of the sub-projects, in response to the frustrations of individuals in describing organisational requirements to extend their technology. Its use was experimental and its development a little rushed. The increased clarity for analysis suggests that the enterprise web should be introduced at the beginning of a project rather than at the end. In pro-poor extension social capital is critical Each enterprise web included an activity of identifying poor farmers. NGOs that already have an established village level network have a comparative advantage. Built over many years, FIVDB and BDS can rely on extensive networks of resource-poor men and women. Each NGO has developed a strong social capital that includes working together and the development of trust (Coleman, 1988, 1990). Such social capital is crucial for economic growth (Fukuyama, 1995; Wilson, 1997) and offers great potential for companies who want to do businesses with the poor (WBCSD, 2004) or establish value chains in developing countries (Goletti et al., 2003). In contrast to FIVDB and BDS, neither IDE nor APEX has their own village network that can be mobilised for extending their respective technologies. Partnerships or outsourcing to an NGO with such networks offer great potential for minimising transaction costs and reducing the risk of not including resource-poor farmers. Introducing the concept of governance Based on their core competencies and organisational culture, each service agency 137


had to weigh up the risk of establishing partnerships (contracting outside the organisation) as opposed to doing the activity oneself (vertically integrating). The enterprise web enables complex delivery mechanisms with many distinct activities to be visualised by professionals from different technical backgrounds. This allows them to talk about it more easily and to identify potential weak links. It also unfreezes discussion on organisational structure to minimise risks in technology delivery and potentially to minimise costs, both principles of transaction cost economics. The justification of 'who does what?' on the basis of risk or transaction cost minimisation is not an issue that is on the current development agenda. This could change, as the enterprise web helps to build cross-disciplinary bridges by neutralising language (Box 11.2). Box 11.2 On the Struggle with CrossDisciplinary Language

The language in business literature and new institutional economics can appear quite daunting to an agricultural extensionist or scientist. Let's look at some of the words: transaction cost minimising, transaction cost economics, efficiency boundaries, governance structure, vertical integration, contract law for partnerships, strategic alliances, core competencies. The awareness of the importance of organisation in pro-poor development leads to wrestling with the above terms. Unfortunately, the language is not easy. The visual tool of the enterprise web is useful in breaking down language barriers to analyse the organisational aspects of extension and pro-poor business development.

CHALLENGES FOR THE ENTERPRISE WEB AS A TOOL Achieving rigor in visualising the enterprise web It is easy to miss activities or to clump activities together. The process of iteration itself can be conducted by different actors for validation of activities and for giving greater clarity on diversity of organisational structure that may be possible for extending a given technology. Committing resources The case studies presented here are a first learning experience. For most partners the enterprise webs still need to be tested in a discussion workshop with their senior management. APEX senior management were involved in the development of their case study and have already used their enterprise web diagram to talk with an international business group. The steps indicated in the enterprise web must be recognised by the management of the service provider and resources committed to address all activities, either alone or in partnership. 138

The Enterprise Web

Extension method costing (direct and transaction cost economising) A common weakness in all the extension method case studies is their lack of costing, showing that this is not part of the current culture of NGOs. To date, the donor community has not rewarded efficiency in terms of cost and risk minimisation. Donors that engage with government, NGOs and private sector could request greater analysis of costing for agricultural extension and reward accordingly. Without attention to cost minimisation and demonstrated risk minimisation the enterprise web may be met with indifference. Also, for more indepth analysis of risk and costing, special facilitating skills will be required as most scientists and extensionist do not have this expertise. Unfreezing the culture to go solo Larger NGOs have tended to integrate vertically and to limit their interaction with government research institutions. Contracting out, in the context of Bangladesh, has tended to present greater risk. And yet, for sustainable extension of agriculture technologies service providers need to look more closely at partnership for technical and social quality. IDE relied on local NGOs to identify the poor farmers and BDS developed independent duck hatcheries to meet duckling supply. The need for ongoing links to technical expertise by a non-governmental service provider can easily be ignored. With the enterprise web, links can be identified and costed, and a contractual arrangement defined. The enterprise web does provide a way of highlighting research-extension linkage as an important contract.

CONCLUSION The enterprise web helps to clarify essential activities in the dissemination of technologies; it is technology and actor specific. Each new technology can be analysed for the most effective and efficient pathway for extension in its own right. In addition, the cost of a specific activity in the extension process, such as identifying resource-poor farm households, can be accounted for. As marketing becomes increasingly critical, the enterprise web can be linked to supply chain management approaches. In other words, the divide between development and business is removed and this in turn allows the issues of poor farmer inclusion to be part of a wider debate.

REFERENCES Coleman, J. S. (1988) Social capital in the creation of human capital. American Journal of Sociology 94, No. Supplement: S95-S120.

139


Coleman, J. S. (1990) Foundations of Social Theory. The Belknap Press of Harvard University Press, Cambridge. Drucker, P. F. (1985) Innovation and Entrepreneurship: Practice and Principles. Heinemann, London. Fukuyuma, F. (1995) Trust: The Social Virtues and the Creation of Prosperity. Hamish Hamilton, London. Goletti, F., Purcell, T. and Smith, D. (2003) Concepts of commercialization and agriculture development. Agrifood Consulting International Discussion Paper No 8, Ha Noi, Vietnam, November 2003. Hennart, J.-F. (1988) A transaction cost theory of equity joint ventures. Strategic Management Journal 9, 361-374. Jarillo, J. C. (1988) On strategic networks. Strategic Management Journal 9, 31-41. Jarillo, J. C. (1995) Strategic Networks: Creating the Borderless Organisation. ButterworthHeinemann, Oxford. Kaplinsky, R. (2000) Spreading the gains from globalisation: What can be learned from value chain analysis? IDS Working Paper 110. Institute of Development Studies, Sussex. Kogut, B. (1988) Joint ventures: theoretical and empirical perspectives. Strategic Management Journal 9, 319-332. Magor, N. P. (1996) Empowering Marginal Farm Families in Bangladesh. PhD dissertation, Adelaide University, Australia (unpublished). Porter, M. E. (1985) Competitive Advantage: Creating and Sustaining Superior Performance. The Free Press, New York. Prahalad, C. K., and Hamel, G. (1990) The core competence of the corporation. Harvard Business Review May-June, 79-91. Reve, T. (1990) The firm as internal and external contracts. In: Aoki, M., Gustafsson, B. and Williamson, O. E. (eds) The Firm as a Nexus of Treaties. Sage, London, pp. 133-161. Thorelli, H. B. (1986) Networks between markets and hierarchies. Strategic Management Journal 7, 37-51. WBCSD (2004) Doing Business with the Poor. A Field Guide. Available at http://www.wbcsd.org Williamson, O. E. (1979) Transaction-cost economics: the governance of contractual relations. Journal of Law and Economics 22, 233-261. Williamson, O. E. (1981) The economics of organisation: the transaction cost approach. American Journal of Sociology 87 (3), 548-577. Williamson, O. E. (1985) The Economic Institutions of Capitalism. The Free Press, New York.

140

The Enterprise Web

Williamson, O. E. (1991) The logic of economic organisation. In: Williamson, O. E. and Winter, S. G. (eds) The Nature of the Firm: Origins, Evolution and Development. Oxford University Press, New York, pp. 90-116. Williamson, O. E. (1996) The Mechanisms of Governance. Oxford University Press, New York. Wilson, P. A. (1997) Building social capital: a learning agenda for the twenty-first century. Urban Studies 34 (5-6), 745-760. Wood, G. D. (1994) Bangladesh: Whose Ideas, Whose Interests? University Press Ltd, Dhaka, Bangladesh.

141

12 Integrated Rice-Duck

A new farming system for Bangladesh

Malik Anwar Khan Gazi Jashim Uddin Ahmed Noel P. Magor Ahmad Salahuddin

SUMMARY Rice-duck farming is a low-cost, organic farming method for small entrepreneurs, introduced in Bangladesh in 2001. Initially, convincing people that ducks were not harmful to rice was a major struggle. By raising ducks on rice paddy, no chemical fertilisers or pesticides are required, while 20% higher crop yields are obtained and net income on a cash cost basis increased by 80%. Labour requirements are continuous at a low level and supplementary feed requirements can be high for certain ecosystems. Female household members can conduct most of the duckrearing activities. By using an enterprise web approach, the case study helps organisations to explore how they can make best use of their competitive strengths to get involved in rice-duck farming. It also highlights the weakest links in the model. To make an integrated rice-duck system work for poor farm families, provision of ducklings and access to vaccines are key bottlenecks that have to be overcome. The first hurdle can be overcome by setting up village hatcheries, parallel to the government supply system, whereas the second depends entirely on contacts with the government vaccine suppliers. Over three years, rice-duck farms were established in more than forty villages, either as individual units or more recently on a community basis. The community rice-duck farming system reduced transaction and labour costs. 143


TECHNOLOGY Integrated rice-duck production is a low-cost, organic farming method for small entrepreneurs. Ducks are allowed to forage in the paddy 20 days after rice transplanting until the flowering stage about 2 months later. While the ducks forage, they remove weeds, eat unwanted pests, soften the soil with their bill and feet movements thereby releasing trapped nutrients, and their droppings provide natural fertiliser. By raising ducks in rice fields, no herbicides, insecticides and chemical fertilisers are required, findings based on three years of research (Ahmed et al, 2004). In the experiments, rice yields increased on average by 20%, which increased household rice availability by two to eight weeks. Duck eggs and meat also significantly increased household protein intake; surplus eggs and ducks were sold at the market for cash.

"Ducks are harmful to rice" is a misconception often voiced. In Bangladesh many farmers rear ducks and many grow paddy, but the opportunity to integrate has never been promoted.

Bangladesh has three rice seasons. There is the main monsoon or aman season (July November) and the now dominant irrigated boro or winter rice (November - May). The minor rainfed aus rice season (May - September) is important in Northeast Bangladesh. The rice-duck management is different for each season. Older ducklings are needed in the cold winter season, whereas a potential shortage of standing water for duck scavenging can result in a greater dependency on supplementary feeding and alter the economic viability of the technology. For aus and aman seasons standing water is not an issue and supplementary feeding is minimal. The partial economic analysis in Table 12.1 (Peter, 2004) shows a negative return on a full cost basis due to labour costs (particularly rearing), but on a cash cost basis there is an 80 percent increase in return, compared to cultivating rice alone. During an appraisal workshop, farmers raised the time for managing ducks and the need for supplementary feed as constraints. However, the increased cash income outweighed this. Women said that they used some of their 'leisure' time for rearing ducklings(Ahmed et al, 2004). An interesting point to note here was the first year's experience in Sylhet. BRRI scientists had a promotional video from Japan. The video showed clearly the increase in rice tillering, root growth and yield, the decrease in insects and weeds, and how ducks soften the soil with their bill and feet movements. Along with this was the added duck production for meat. The video was the only evidence. The experimental results of the first aman season coupled with the farmer feedback produced a response in Magor of, 'It is just like the video!'

144

Integrated Rice-Duck

Table 12.1 Cost-benefit analysis of rice-duck farming. Example of Khadimnagar union, Sylhet during aman season. Taka per 20 decimal plot (0.08 ha) ACTIVITY

RICE ALONE

RICE-DUCK

Costs Ducklings (30 @ Tk 12/duckling) Seed

360 30

Bamboo & labour for fencing

550

Fertiliser

150

Labour for weeding

360

Insecticides Labour for fertiliser & pesticide application

33 100

Feed (6 months)

1,530

Housing & labour for housing

145

Vaccine

90

Labour for rearing

2,300

Labour for moving in & out Total variable cost

22

900 673

5,897

1,600

1,735

Benefits From paddy From

ducks1

Gross return Net return on full cost basis Net return on cash cost basis 1

1,740 1,600

3,475

927

-2,422

1,387

2,493

Income from selling 6-7 month old ducks at 10% mortality rate

That first season was a confidence boosting season for each of us. Traditionally, women in Bangladesh conduct about 80% of duck-rearing activities, involving hatching, housing, nursing, feeding, and herding. Once rice farmers have learnt duck rearing techniques, it is sufficiently easy even for younger family members. Under their duck programme, the national NGO Friends In Village Development Bangladesh (FIVDB) already trained about 500 vaccinators in Sylhet, 30 in Moulvibazar and 100 in Sunamganj, two thirds of which are women. In future, more women may be trained to vaccinate ducks. They will be able to make cost savings by vaccinating their own ducks and earn extra income by vaccinating ducks of neighbours. 145


ACTORS AND NETWORKS An overview of all actors is given in Table 12.2. The Bangladesh Rice Research Institute (BRRI) provided overall leadership for testing and validating the rice-duck technology. FIVDB was established in 1981. Apart from formation of self-help groups, development of human and social skills, promotion of savings and microcredit, and adult education with village libraries and cultural activities, it supports sustainable Table 12.2 Actors involved in rice-duck project in Bangladesh, 2001-2003

146

ACTORS

CONTRIBUTIONS

LONG-TERM EXPECTATIONS

Lead organisations BRRI and FIVDB

Provide technical knowledge and support; establish links with different stakeholders

Both partners will explore the potential of rice-duck for organic production of paddy, duck meat and eggs

Partner organisation (BDS)

Provide necessary knowledge and support

BDS plans to extend rice-duck to more farmers

Hatchery owners (Small businesses)

Sell duckling to farmers

They will be able to increase duckling production; more hatcheries will be established

Farmers with less than 0.2 ha plots

Demonstrate rice-duck at village level; build up community awareness to involve more farmers

Community will shift from individual plots to co-managed communal rice-duck systems

Farmers' wives

Rear ducklings, sell eggs and meat; conduct rice post-harvest activities

Women get extra cash income and invest in better family nutrition

Other NGOs

At least 4 other NGOs have initiated new rice-duck plots in other villages

More NGOs will embark on rice-duck cultivation

Government agencies

DAE, Livestock Offices, govt. poultry farm and BADC provide seed, ducklings, vaccine and services

Government organisations (GOs) will provide more support to rice-duck

Union Parishad

Some UP members and chairmen are providing moral support and help to establish links

More involvement and moral support will be needed

Private duck vaccinator in the village

FIVDB, government organisations and NGOs coordinate and network and ensure they provide vaccines

More duck vaccinators will emerge where rice-duck practice starts


agriculture, livestock and poultry activities. Under its livelihood programme, it aims to improve homestead-based production, mainly of women-headed households, by promoting integrated production systems. FIVDB provides training on duckling production, and duck husbandry and management. Malik Anwar Khan, one of the founders of FIVDB and the coordinator of the Poultry Unit in the Livelihood Enhancement Programme has acquired a wealth of duck experience over 25 years during which he has published three books on duckling production, duck rearing and duck diseases. Malik Anwar Khan's passion for, and knowledge of how ducks can significantly improve rural livelihoods has made him a driving force in the rice-duck project. FIVDB's central facility has an effective duckling production unit with two automatic electric incubators and one Chinese incubator. The latter is a low-cost solution whereby eggs are kept at constant temperature in a bamboo cylinder; primary heat being provided by sun light or by a kerosene stove on cloudy days. FIVDB has established over 20 duckling production units with Chinese incubators in Northeast Bangladesh. Some of these units produce ducklings regularly, while others function only seasonally. In July 2001, BRRI teamed up with FIVDB to conduct research on integrated riceduck farming with poor households in Sunamganj, Moulvibazar and Sylhet districts of Northeast Bangladesh (Table 12.3). For each district, FIVDB appointed a project assistant with an agricultural diploma. Table 12.3 Rice-duck activities in three districts of Bangladesh, 2001-2003 SUNAMGANJ

MOULVIBAZAR

SYLHET

No. of unions

4

4

3

No. of villages

9

18

13

Farmers involved in research

40

48

64

Other farmers adopting rice-duck in extension phase

28

37

17

2,190

2,400

2,620

No. of ducks

FIVDB also collaborated with the local non-government organisation (NGO), Barisal Development Society (BDS), which played a significant role in networking with other NGOs, private sector and government institutions such as BADC, DAE, district livestock officer, government poultry farm, and the Union Parishad office of the local government. One agriculturist and two supervisors were involved from BDS. "We already had a friendly relationship and good understanding with BDS. To make a partnership, it is very much important that you know a person in that organisation and that there is mutual goodwill and understanding," said Malik Anwar Khan. 147


EVOLUTION OF THE METHOD The traditional practice of duck raising in rice fields involves herding the birds in paddy fields after the rice is harvested. Although integrated rice-duck farming has been practiced for generations in Japan, and is practiced in many other Asian countries (Cagauan et al., 2004), it was unknown in Bangladesh, until recently. When two BRRI scientists, Mr. Shaikh Tanveer Hossain and Dr. Gazi Jashim Uddin, visited Japan in 1998, they were impressed by the benefits integrated rice-duck farming brought to the farmers and by the environmental friendliness of the system. Upon their return, they discussed their experiences with the director-general of BRRI. Early in 2000, a first concept note was developed and submitted to PETRRA, but not approved. PETRRA stressed that the institutional support was not in place for BRRI to make the model work, unless strategic partnerships were developed. When, later on, a new proposal was submitted in partnership with FIVDB as the leading institute in Bangladesh with duck expertise, the proposal was approved. The first challenge of the project was to introduce the new concept to all the actors and consequently to evaluate which key activities and links had to be established to make the rice-duck system work under Bangladeshi conditions.

THE RICE-DUCK EXTENSION METHOD The first output of the rice-duck project was to validate the technology and the second output was to articulate the essentials to extend the technology. It is the latter for which the enterprise web tool was used. The different activities that require attention are depicted in Figure 12.1, and referred to in each of the paragraphs below. Select suitable ecosystems Those areas where duck rearing and paddy cultivation exist side-by-side are suitable for integrating both systems. Apart from saline water and upland or drought prone areas, most parts of Bangladesh are suitable. In places where line transplanting is already being practiced, such as in Comilla, the system can be quickly established. Under this sub-project, sites in the northeast and south tidal area were selected. Select resource-poor households In the northeast all rural poor households involved in the project had had a longterm relation with FIVDB and expressed their interest in embarking on rice-duck. They practiced homestead agriculture, had water facilities near their home, a possible duck-rearing space and owned some paddy land. BDS selected households in Barisal from within their group programme. Other examples of building on comparative strengths in selecting poor farmers can be

148


Requires training input

Provision of credit

Vaccine collection from government

Establishment of village hatchery

Supply of ducklings


Selection of resource-poor households

Supply of ducklings from government

Rearing of ducklings

Vaccine distribution

Vaccination

Provision of credit

Production of rice duck by resource-poor men and women Home consumption

Sales of rice, ducks and eggs

Figure 12.1 Enterprise web for rice-duck farming

found in the Farmseed (see Chapter 18) and the women-led group extension (Chapter 3) projects by the NGO Agricultural Advisory Society (AAS). After the initial success of the rice-duck project with FIVDB and BDS, another NGO AAS commenced rice-duck promotion in collaboration with MAC Bangladesh, an NGO in Srimongal upazila under Moulvibazar district. Their staff and farmers received training from FIVDB at PETRRA's request. But the way they introduced this new system differed. They set the pre-condition that the whole community had to agree to conduct rice-duck culture before AAS/MAC would provide support, anticipating that by doing so future social conflicts over natural 149


resource management would be minimised. In addition there may be some economy of scale in field supervision of ducklings and removing the need for boundary fencing of paddy fields (see Box 12.1).

Box 12.1 Community Approaches in Rice-Duck

The NGOs MAC Bangladesh and AAS discussed with communities in two villages (Sirajnagar and Uttara Varaura) of Srimongal in northeastern Bangladesh, about the benefits of integrated rice-duck farming. In both places, farmers had experience with rice and duck rearing, but farmers from Uttar Varaura did not combine them. Two systems emerged through group discussions and a trial in aman season 2003. In Sirajnagar village, 21 farmers with adjacent plots purchased 2,000 ducklings individually, but through MAC, who also provided support for vaccination and farmer training. Other community members with adjacent plots agreed on refraining from insecticide use; before they sprayed twice a season. Each farmer shelters, rears and grazes his or her ducklings individually. Because the land available for grazing is seven acres (2.8 ha), fencing is not required. The group, as per suggestion of MAC Bangladesh, started producing duckweed in their abandoned ponds an in situ source of duck feed. In Uttara Varaura village, seven farmers joined their plots totalling 3 acres (1.2 ha) on which they graze 350 ducklings owned by the group. But interestingly only three farmers shelter and rear the ducklings, while they take turn in grazing the ducklings. They meticulously keep track of all expenditures, including their time, and share costs and benefits with the group. 'Practicing rice-duck farming in a group works well for us," said Ratan Dev, one of the group members.

Supply ducklings A reliable supply of ducklings is essential for this model to work. Two parallel sources may be explored: government duckling suppliers or NGO/private smallscale duckling production units. In Bakergonj upazila in Barisal district, government farms provide ducklings, but not at the right time. This resulted in poor performance of the newly established rice-duck farms. To overcome similar problems, the project assisted in setting up a village hatchery or duckling production unit. For two months an experienced duck farmer from Sylhet district, trained by FIVDB, helped poor households in Barisal to establish a small-scale duckling production unit and to teach them the tricks of the trade. By the end of 2003, about 20 families were supplying eggs to two duckling units, each capable of producing about 600 ducklings per batch. Having seen the benefits for their members, BDS is planning to establish more hatcheries in their project area. To establish a new hatchery, at least 500 fertile eggs need to be obtained from any duck farm in the country. In order to ensure the eggs are fertile, the chosen farms must have at least one male for every eight female ducks. 150


Ensure access to vaccines, provide vaccination & disease treatment support Timely availability and provision of duck vaccines is the next essential element of the model. In Bangladesh, duck vaccine production is limited and in the hands of the government. Production takes place in the Livestock Research Institute in Dhaka and Comilla, and at the vaccine production centre of the Bangladesh Agricultural University in Mymensingh. Vaccines are then distributed to the districts. In some cities, vaccines may be available in the bigger agribusiness shops, but generally they can only be bought routinely at the district or upazila Livestock Office. Their staff and those of FIVDB, can diagnose duck diseases. For diagnosis and timely access to vaccines, good contacts with these government institutions are essential. Various companies such as Bangladesh Pharmaceuticals produce medicines. Duck medicines are similar to those for chickens, which because of wide-scale government support to the industry, are readily available in every pharmaceutical shop. The bottleneck clearly lies in the availability of duck vaccine, not the medicines. Once vaccines are available, the ducklings have to be vaccinated. Professional vaccinators were trained in areas where new duck farms were established. In some cases, village women were trained and became vaccinators for all rice-duck farms in their community. Rear ducklings This household activity is generally conducted by women. Rice-duck rearing families purchase ducklings from outside or produce them themselves. As housewives become experienced at hatching and nursing the ducklings the mortality rate decreases significantly. Ducklings are reared until they are 20-25 days old when they can graze in the paddy fields.

Hatching ducklings in your bedroom. Anowara Begum looks at her one-day old ducklings. She will nurse them for the next three weeks, until they are old enough to graze outside.

"It is very time consuming to look after the eggs and ducklings, but I now make better use of idle time and plan my day much more," says Anowara Begum from Donokandi village in Sylhet. She's been involved in riceduck culture since 2001 and hasn't regretted a single minute of it. Anowara Begum hatches the eggs and rears the ducklings in the bedroom of her mud house. Clearly, no large expensive infrastructure investment are needed. Implement rice-duck farming Resource-poor farmers can rear about 30 ducks on 20 decimals of land (a decimal is hundredth of an acre, or about 40 square meters, so fields are 800 square meters). Without the technical details, it is worth mentioning that

151


management costs can be significantly reduced if rice-duck is implemented at the community level with groups of farmers having adjacent plots, rather than by individual farmers (see Box 12.1). This improves the efficiency of duck herding, and protection from predators while reducing fencing costs. Feeding is required during two months of the dry and two months of the flood season when natural food is unavailable. If proper care is not taken, ducks lose weight and become prone to diseases. Better-off farmers with 300-500 ducks generally buy broiler food for when natural feed is unavailable, whereas poorer farmers with less than 100 ducks mix rice husks with weeds, and collect frogs and snails from the flood plains. In Srimongal, MAC Bangladesh and AAS promote the use of duckweed Lemna perpusilla. Although readily available, duckweed can also be cultivated to have a more steady supply (Skillicorn et al., 1993). Provide credit Credit may be required to set up a small-scale household duckling production unit and to embark on rice-duck farming. NGOs and non-commercialised banks are the main sources of credit. Typically setting up a small-scale household duckling production unit will cost Tk 15,000 (US$ 264). Resource-poor families who want to embark on rice-duck need about Tk 12 (US$ 0.21) per duckling and some additional capital to purchase feed and vaccines. These costs can be recovered in one season, so only a short-term loan is needed. However in Barisal, farmers who were not BDS members said that they were too afraid to take microcredit because they feared they would not be able to make weekly repayments (Orr et al., 2004). Provide training To optimise the rice-duck system, specific training is essential for paddy and duck farmers to integrate the two farming methods. It is essential that farmers are first made aware through village group discussions and video shows that ducks do not harm the rice. Women are trained at FIVDB's training centre, through practical field orientation at their homestead, group discussions, farmers' workshops, and communication materials such as videos, posters, leaflets and the monthly FIVDB newsletter Gram Bhanhob. A training module for rice-duck is available, along with a guide book by Khan and Ahmed (2004). In FIVDB's working area there are many male and female groups, village federations, farmer extension agents, community learning centres and community libraries. These community institutions have facilitated the implementation and management of field activities, and ensure long-term sustainability. It is anticipated that within a few years farmers will be capable of sustainably managing rice-duck systems themselves. 152


THE WEAKEST LINK Vaccine supply is the weakest link of the system, as this is entirely in the hands of the government. When selecting partners to implement rice-duck, establishing good links with the vaccine supplier in the region is of utmost importance. In future, it would be highly beneficial if private companies like Bangladesh Pharmaceuticals were to produce vaccines. With their efficient distribution system duck vaccines could become more readily available to farmers at crucial times. The supply of ducklings in the right quantity, at the right age and on time is also a weak link. Setting up a village duckling production unit can be pursued to overcome this weakness. Rice-duck is a new system and thereby its extension needs to be coupled with capacity building for both the service provider and farmers. As we needed to build confidence, each organisation and each village will need to build confidence in the technology itself.

INTERCHANGEABILITY OF ACTORS One organisation or a number of partners, depending on their comparative advantage, could potentially implement each activity of the enterprise web. Selecting households in a new area, for instance, is better done by local NGOs or community-based organisations, as they have a long-term engagement and commitment to their communities. All NGOs with an interest in agriculture can embark on rice-duck culture, if they get the necessary training. Large NGOs with credit programmes which are already working with poor farmers would seem obvious potential collaborators at first. "Most large NGOs are engaged with broilers rather than ducks, because it is a shorter-term investment," says Malik Anwar Khan, "so it is better to build partnerships with smaller NGOs." These days, government policy favours more highly intensive poultry farms.

Letting the ducks out.Young ducklings need particular protection from predators, such as birds of prey or mongooses. Any family member, but usually children and/or women, are given this responsibility.

DIFFICULTIES, RISKS AND ASSUMPTIONS

The first major obstacle to overcome was to change an old misconception that ducks are harmful to rice. This has been overcome by involving farmers and other community members in the research. 153


During the first season duck mortality occurred because of poor management during hatching and nursing and because of a scarcity of duck food and vaccines. Another concern, which was unexpected, was the loss of ducklings from the predatory mongoose. Also, sudden seasonal flash floods and droughts occurred in some areas. All these difficulties triggered project staff to emphasise specific technical issues in future training sessions and to jointly look for local solutions.

Organic rice. Ducks eat weeds, insects, and snails while fertilising the land with their droppings. They create perfect conditions for organic rice cultivation.

Snail collection to feed ducks is an important additional income source for women, but snails may become overexploited if no plan for sustainable management of the beel (seasonal water body) is developed (Sultana et al., 2001). Therefore, in areas where natural food scarcity occurs, experiments should be conducted to expand the integrated rice-duck system with rearing of snails or duckweed, as mentioned earlier. Or the nitrogen-fixing fern Azolla could be used to enhance nutrients and provide an in situ source of duck feed, as successfully tested in the Philippines (Cagauan et al., 2004). FIVDB also has experience in culturing earthworms (Khan, 1994). When water comes from communal irrigation sources, ducks may die following pesticide application in neighbouring fields, further emphasising the importance of a community approach. But above all, for Bangladeshi farmers timely access to crucial duck vaccines, as described earlier, remains the major risk.

SCALING UP

As rice-duck is a new concept for Bangladesh, changing people's perception is the first prerequisite to adoption. PETRRA introduced rice-duck to the nation through Rhidoye Mati o Manush (Land and People), a popular weekly agricultural programme on Channel i. Wherever ducks are available in the country, rice-duck could be introduced once there is a positive attitude towards the new system. Experienced farmers could become increasingly involved in convincing and training others. To scale up, FIVDB plans to experiment with local users-pay systems whereby poor farmer groups save funds and contribute a nominal fee in return for training and technical assistance. Rice-duck could become a suitable system for the organic production of aromatic rice, which is generally grown in larger blocks of land. The specific conditions that have to be fulfilled for this system are described in Chapter 14. The potential for rice-duck culture in Bangladesh is high: not only is rice yield increased by 20%, but the rice is also produced organically. Millions of babies eat cereals on a daily basis, so the market potential for organic rice is enormous. Certification for 'organic rice' can also provide a price premium for farmers. 154


CONCLUSION Three years of research have shown that rice-duck farming is economically profitable, environmentally friendly, and technically feasible for poor farmers in Bangladesh. However, before embarking on this new system, an organisation must carefully plan with whom to partner, and where credit, vaccines and ducklings can be sourced. Technical expertise is available in Bangladesh which was strengthened during this project, but this integrated system will only work if all the required conditions are fulfilled.

REFERENCES Ahmed, G. J. U., Hossain, S. T., Islam, M. R., Anwar, M. A. and Biswas, J. P. (2004) Integration of rice-cum-duck farming for resource-poor farm households. In: Workshop Proceedings: Emerging Technologies of the PETRRA Sub-projects. Organised by PETRRA-IRRI and BRRI, at BRRI Auditorium, Gazipur. Cagauan, A. G., Branckaert, R. D. S. and Van Hove, C. (2004) Rice-duck farming in Asia: Increasing its production potentials by integration with fish and the nitrogen-fixing aquatic fern Azolla. The First INFPD/FAO Electronic Conference on Family Poultry. Available from http://www.fao.org/ag/AGA/AGAP/LPA/Fampo1/freecom4.htm Khan, M. A. (1994) Hansh Paloner Adhunik Padhti. A book on the principles of duck farming system and its management. ITDG and FIVDB, Dhaka, Bangladesh. [in Bangla] Khan, M. A. and Ahmed, G. J. U. (2004) Technical Guide Book on Integrated Rice-Duck Farming Systems. IRRI, Dhaka, Bangladesh. Orr, A., Seema, F., Nabi, S. A. and Peter, J. I. (2004) From Technology to Livelihoods: Case studies of four PETRRA sub-projects. Consultancy report. IRRI, Dhaka, Bangladesh. Peter, J. I. (2004) Cost-Benefit Analysis of Three PETRRA Sub-Projects: Second Part. IRRI, Dhaka, Bangladesh. Sultana, P., Thompson, P. M. and Ahmed, M. A. (2001) Women-led fisheries management - A case study from Bangladesh. Global Symposium on Women in Fisheries, 29 November 2001, Koashiung, Taiwan. Skillicorn, P., Spira, W. and Journey, W. (1993) Duckweed Aquaculture. A New Aquatic Farming System for Developing Countries. The World Bank, Washington. Available from http://www.ntrnet.net/~skilli/duckweed.htm#Duckweed

155

13 Pump It Up

Developing a Pro-poor Mobile Pump Market

Badrul Alam Noel P. Magor Paul Van Mele

SUMMARY Poor farmers need affordable and appropriate irrigation technology to improve their livelihoods through increased rice productivity and diversified cropping. After being tested and refined, the International Development Enterprises (IDE) disseminated mobile pumps in Barisal district in the coastal area of Bangladesh, where tidal canal water is readily available. The mobile pump is a more durable metal modification of the bamboo treadle pump, and it is mobile. To create a supply chain, the international NGO IDE trained and established links between a producer, dealers and installers. Public and private sector helped in creating a market. Apart from the local entrepreneurs, the Department of Agricultural Extension (DAE), the local government and local NGOs also promoted the technology, with some of the latter facilitating access to credit for farmers. Full cost of a pump was Tk 1,650 (US$ 29). After two seasons, 157 poor farmers had purchased a mobile pump, of which 72 were made available at subsidised rate for demonstration purpose. We believe that for high volume marketing, both end users and manufacturers will need access to credit. This case study recommends strengthening and building upon existing local business chains. Developing a pro-poor enterprise for mobile pumps is as much about pumping up water as about pumping up confidence. 157


TECHNOLOGY The mobile pump is a foot-operated metal device for lifting ground water or surface water from ponds, canals, rivers and streams for irrigation. But above all, it is an affordable technology for poor farmers. It can easily be moved between plots and is easy for women and children to operate (about 30% of the pumps are operated by women). The users can make simple repairs themselves. Mobile pumps can be locally manufactured, distributed and maintained, as described in this chapter. The mobile pump competes with the diesel-operated low lift pump. But farmers say that with this heavy pump timely irrigation is not always available, as four or five farmers may compete for access at a time. With the mobile pump a farmer can apply water as needed and easily move the pump from one plot to another. Apart from its longevity and suitability for surface water, this is one of the major advantages over the treadle pump, which uses a fixed bamboo structure (see also Box 13.1).

Mobile pump

Treadle pump

Low lift pump

ACTORS IDE, an international NGO based in the USA, has worked in Bangladesh since 1984 (see www.ide-bangladesh.org ). IDE specialises in the transfer of appropriate, affordable and sustainable technologies to the private sector and facilitates rural mass marketing to create sufficient demand for a new product (IDE, 2003). This in turn acts as a stimulus to local small businesses to produce and market such products. Box 13.1 More on the Mobile Pump

Water source: Surface water or ground water Discharge capacity: 0.9 to 1.1 liter/sec Pump head: Tk 1,050 (US$ 18.5) or Tk 1,250 (US$ 22.0) depending on diameter Cost of accessories such as suction hose: Tk 600 (US$ 10.5) Lifting capacity: maximum pumping depth is about 7 metres Irrigation coverage: 0.2 ha for rice and 0.4 ha for vegetables Weight of the pump: 18 kg Life span: 4 to 5 years

158

Pump It Up

For example, IDE facilitated the production and marketing of the treadle pump (another manually operated irrigation device) through the establishment of 72 pump producers, 1205 pump dealers, 4000 well drillers. It has worked with over 246 NGOs. Since the early 1980s, when treadle pumps first became available in Bangladesh, about two million treadle pumps have been installed worldwide (Orr et al., 1991). It is this expertise that IDE brought to the production and marketing of the mobile pump. The local area network for producing, distributing and marketing the mobile pump was established by IDE with one part time manager (the senior author) and one experienced IDE field officer. The DAE and the local government helped in promotion, information dissemination and assistance in selecting farmers for demonstrations. Various NGOs (VOSD, BDS, Proshika, Caritas, BRAC and ASA) played a similar role, and added the potential of providing poor farmers access to microcredit. The other major actors were a producer, dealers, mechanics and installers, and poor farm families in Barisal district in the coastal area of Bangladesh.

ESTABLISHING A LOCAL NETWORK There are two preconditions for a sustainable market: (i) the existence of sustainable demand and (ii) the existence of an efficient supply chain. We developed demand through an awareness building programme and created a local production and distribution channel for the supply of quality pumps (see enterprise web in Figure 13.1).

Farmers like the mobile pump, because it does not need fossil fuel and they can easily take it from one field to another. Every evening they can take it back inside their house to prevent theft.

Select suitable ecosystem We used our existing experience for identifying an appropriate area and linked up with three villages in Uzipur upazila in Barisal district. This was the first systematic programme for establishing a mobile pump network in the coastal area of Bangladesh. Earlier on in an adjacent district, an IDE-linked producer of treadle pumps had already manufactured some mobile pumps, some of which had found their way to Uzipur. We talked with local dealers and one of the NGOs about the large potential of the mobile pumps in their locality. After all, Uzipur had ample canal water that was suitable for this simple irrigation technology. This suitability was further reflected in a review workshop with researchers, local senior extensionists and NGOs: "As surface water is available, the mobile pump was seen as

159


Building effective network


Promotion of mobile pump

Producer development


Dealer development

Building links NGO/CBO

Develop pump mechanics and installers

Micro finance



Building links local Govt./GO


Mobile pump used in crop cultivation by poor men and women Home consumption

Sales of rice, vegetables & other crops

Figure 13.1 Enterprise web for mobile pump uptake for poor men and women

highly suitable for greater Barisal for cultivation of boro rice (November - May), supplementary irrigation of aman (July - November), vegetable cultivation and for nurseries" (PETRRA, 2003).

160

Pump It Up

Establish an effective network We then contacted the DAE government extension offices, the local government, and potential partner organisations that were already active in the project area. The NGOs VOSD and BDS showed most interest, while Proshika, Caritas, BRAC and ASA received awareness training. The links between actors are essential for ensuring timely access to mobile pumps. The supply chain includes three network partners, namely producer, dealers and installers. By 2004, in less than two years, we established one producer, three dealers and 20 installers in the project area, with DAE block supervisors and six NGOs promoting the pumps. As none had any expertise in mobile pump technology, we trained about 30 people as a first step in building skills and confidence. Each network member in turn promoted quality pumps to farmers, with two NGOs making them available through microcredit. The other NGOs mainly engaged staff in the project, because of NGO solidarity and the good relationship they have with IDE. Badrul Alam (the senior author) helped out the other organisations on several occasions, and they were happy to return a service. One of Badrul's friends, from the time he was at University, holds a senior position in Proshika. Partnerships glued by people with a common history and interest often last longest.

Hardware shops are part of any market in rural Bangladesh. But only by becoming part of a local network, dealers can see their customer base expand.

Periodic pre-season workshops and motivational meetings for farmers with the producer, dealers, mechanics, NGO and DAE staff are crucial to consolidate the network. After the project, IDE will continue providing low level support to the network through its Business Development Services programme that is operating in the region. 161


Promote the mobile pump Advertising and market promotion activities are essential in creating a market. Prior to the establishment of a local producer and developing the skills of dealers and installers, only a very small number of mobile pumps were available in the locality. Essentially it was a new technology for farmers. Market creation in Uzipur aimed at raising awareness among farmers, producer, dealers, mechanics and local organisations. This 'new' irrigation technology would provide an alternative to: no irrigation; the indigenous swing basket method; or low lift pump irrigation. The activities also enabled farmers to identify and gain confidence in quality mobile pumps so that they were not to be cheated by unscrupulous dealers. Other promotional activities included: display of the mobile pump in the market; motivational meetings; one to one contact; farmer field days; promotion through 'miking' (using a microphone on a rickshaw to send out message, see photo Chapter 9); distribution of leaflets and promotional cotton bags; rickshaw and roadside signboards; and display in a local agricultural fair. By promoting the pump through all channels, and maintaining satisfactory profit margins, each network member contributed to the sales (Table 13.1). Table 13.1 Number of mobile pumps sold by network members to poor farmers NETWORK MEMBER

SUBSIDISED DEMO PUMPS (N=72)

PUMPS WITHIN PROJECT AREA (N=74)

PUMPS OUTSIDE PROJECT AREA (N=11)

Producer

-

-

11

Dealers

23

19

-

Mechanics

19

17

-

-

7

-

NGOs (BDS and VOSD)

17

25

-

DAE

11

5

-

Union Parishad member

2

1

-

Demonstration farmers

Select demo farmers Figure 13.1 shows that selection took place though three different means: private sector entrepreneurs; NGOs and community-based organisations; and the local government. Based on PETRRA's requirements and IDE's experience, with each of these partners we selected farmers who: have food security for three to eight months of the year 162

Pump It Up

own up to 0.4 ha of land have land free from tree shade have a plot near a roadside with easy access to many people are willing to cooperate with IDE agree to keep records as per project requirement are willing to motivate other farmers will use the pump at maximum capacity. But we also took into account more personal characteristics (see Box 13.2). Mr. Mozahar Hawlader, from Uzipur upazila, Barisal is 48 years old and poor. But he is enthusiastic, dynamic and jolly-minded and was therefore selected as a mobile pump demo farmer by Mr. Joynal Boati, one of the dealers. Although the total installation cost was Tk 1,400 (US$ 25), Mozahar could buy it at half the price in December 2002, under the condition of becoming a demo farmer.

Box 13.2 I Feel Good

"I own 60 decimals of cultivable land (2,400 square m). However, due to lack of irrigation facilities my land remained fallow during the boro season (November - May). After installation of the pump and receiving technical support from the DAE Block Supervisor, I cultivated half my land with rice and the remainder with vegetables," he says. Now, Mozahar not only has rice for the whole year, but also earned Tk 15,500 (US$ 273) from his tomato, eggplant, pumpkin, cabbage, lady fingers, puishak, lalshak and datashak. Mozahar rented out his mobile pump to his neighbours at the rate of Tk 10 per hour (US$ 0.18), earning him Tk 600 (US$ 11) in one season. Through his advocacy, five neighbouring farmers purchased pumps. The additional income from vegetables allowed him to connect to the electricity grid, buy new clothes for his son and wife, repair his house and install a hand tube well for drinking water. Mozahar concludes: "My family enjoys better hygiene now, because we have access to clean water for bathing and washing, and can afford soap. We are no longer dependent on the water lord and are more confident about the future of our family. We now play a role in the development of our community and have gained the confidence of other villagers. We feel good about ourselves."

Barisal being a district where strict religious and social rules prevail, we first asked any potential demo farmer whether they would allow their daughter or wife to operate a pump during field demonstration days. Although all of them agreed to it, not all kept their promise. But still, our focused gender approach resulted in at least 30% of the demo farmers being women, which in turn stimulated other farmers to bring their women along to the field days. After all, their wives or children could easily operate the pump. Identify and develop producers As recommended by Mr. Joynal Abedin, a seller of hand tube wells at Shikarpur 163


The mobile pump network and mobile phones go hand in hand.Hardware dealers prefer keeping small amounts of many different items. Being part of the local network, they can quickly forward demand to the manufacturer, who lives only 15 minutes away.

bazaar, we contacted Mr. Salahuddin at the Sumon engineering workshop as the potential local producer for mobile pumps. Mr. Salahuddin already produced manufacturing grills and treadle pumps, and repaired agricultural equipment. He had a good reputation in the area. After he showed enthusiasm in expanding his business, the IDE technical team decided to train him as producer of mobile pumps. Although Salahuddin had most tools and accessories required for the production available in his workshop, he still needed about Tk 30,000 (US$ 528) for the cost of jigs and fixtures, dies for piston plates and a rolling machine for the cylinder. He obtained all from various suppliers, partly on credit. In two seasons, Salahuddin produced 116 pumps, of which 11 were sold by himself directly outside the project area. The others he supplied to the selected dealers through the local network at a net profit margin of Tk 150 per pump (about 15% of the total cost price or US$ 2.64). He earned Tk 17,400 (US$ 306) additional income and clearly wishes to continue producing pumps after the PETRRA project ends. Mobile pump production is a seasonal business from November to February. Working capital can be a constraint which can lead to a backlog during the peak period. This is discussed further below. Establish local network of dealers Dealers form a critical link. We selected and trained three dealers who were already in the hand tube well business, well known in their own locality, motivated to sell mobile pumps, willing to store a small number of mobile pumps as per demand of poor farmers and prepared to work through the local network. Although they lived in a neighbouring upazila from the producer, it was only 15 minutes away, and after we built the network, they remained in regular contact through their mobile phones. The dealers played an important role in creating demand, in encouraging mechanics to promote the mobile pump, in training unskilled mechanics and in ensuring information flow among the network members by contacting the NGOs. But they also ensured availability of pumps and spare parts as per demand, and provided after sales service at a reasonable price. As they had a direct stake, dealers shared in the costs of the promotional activities by training farmers free-of-charge during field days. The dealers had a capital investment of Tk 246,000 to 354,000 (US$ 4330 - 6230) in their business, of which less than 10% comprised mobile pumps. Clearly, they sell

164

Pump It Up

mobile pumps along with other things such as hardware and sanitary materials, motor parts, hand pumps and tube well pipes. But small orders of five to ten pumps were often made to the producer with a week's notice. This demand-led production eliminated the need to involve a distributor or stockist in the local network. Table 13.2 shows that each dealer was actively marketing their latest technology, making a net profit of 160 Tk per pump (US$ 2.82). Table 13.2 Net profit of dealers after selling mobile pumps for two seasons NAME OF DEALER/SHOP

NO. OF PUMPS SOLD

NET PROFIT IN TK (US$)

Md. Joynal Boyati, M/S Bhai Bhai Sanitary

48

7,680(135)

Md. Shekander Shikder, M/S Shikder Machinery

36

5,760(101)

Md. Kazi Kamrul, M/S Zayed Enterprise

32

5,120(90)

Train pump installers To install mobile pumps accurately and for repair and maintenance we trained twenty pump installers. The majority install pumps as their main occupation and are linked to dealers (see Box 13.3). As a rule, we work with installers who are also farming themselves and have more than ten years of experience in installing pumps, either hand tube wells, shallow or deep tube wells.

Shajahan Boyati from Mundopasha village in Uzirpur upazila is a very good tube well mechanic, renowned in the region with his 10 years of experience. He is working for M/S Bhai Bhai Sanitary shop of Md. Joynal Boyati. But in the winter season he sometimes was without work, which kept him poor. With his twenty decimals of land (800 square meters) he could hardly grow enough rice to feed his family for three months of the year.

Box 13.3 The Mechanics of a Bright Future

When IDE approached him to receive training in the installation, repair and maintenance of mobile pumps, Shajahan enthusiastically accepted this technology. IDE also invited him to participate in different promotional activities as they had learnt that installers were crucial for promoting the technology. The shop sold twelve pumps thanks to Shajahan for which he got a commission of about Tk 25 or US$ 0.44 per pump. During one season, Shajahan installed fifteen pumps in farmers' fields and trained them in repair and maintenance. For each installation he received Tk 150 (US$ 2.64). He also made Tk 1,200 (US$ 21) for repairing and maintaining pumps. So his total extra earnings were Tk 3,750 (US$ 66). Since he started promoting and installing mobile pumps, he is eating fresh and healthy food, repaired his house and his children are better dressed. He will work hard to further increase sales of mobile pumps in and around Uzipur. After all, he is the mechanic of his own future.

165


Finance the end-users of mobile pump Mobile pumps are especially suitable for marginal, small, landless and resource-poor farmers, but many do not have the cash to buy one and rely on relatives or credit suppliers. Provision of credit clearly had, and will continue to have, a great impact on the market for mobile pumps. In Figure 13.1 this is an activity limited to NGOs. Table 13.3 shows that even when pumps were sold at 50% subsidised prices to marginal farmers, about a quarter required credit from outsiders. When sold at full cost, about half of the farmers took credit from NGOs (BDS and VOSD) or dealers. Table 13.3 Sources of capital for procuring mobile pumps SOURCE OF CAPITAL

SUBSIDISED DEMO PUMPS

PUMPS AT FULL COST

53

46

NGO

9

25

Dealer

10

14

Total

72

85

Own and relative

Training We trained all partner organisations in the supply chain, including 72 demonstration farmers. Every cropping season, for four to five hours, we taught farmers to install, repair and maintain pumps by themselves. Old pump users trained newcomers free of charge. During the last phase of the project six partner organisations contributed twenty five staff members as facilitators during field sessions, also without receiving any payment. This was based on the principle of reciprocal support, as described earlier on.

KEYS FOR SUCCESS The marketing of mobile pumps to poor farmers was successful because: The mobile pump technology was suitable for the tidal ecosystem, which resulted in irrigation water being readily available A systematic programme of raising public awareness ensured demand for the mobile pump and trust in quality of the products. (This approach also proved its value in promoting modern seed to the poor, as illustrated in Chapters 18 to 20.) Local leaders (DAE, local government, NGOs) were convinced of the suitability of the mobile pump for their locality, and helped in market creation (most case studies presented in this book show a similar approach) Training local entrepreneurs added more worth to existing businesses 166

Pump It Up

The producer was recommended by a local dealer and hence an existing local business chain was strengthened and incorporated in the expanded supply chain The mobile pump technology was promoted to local NGOs as having potential for their clients IDE was experienced in the establishment of local networks for other irrigation technologies and transferred that skill to mobile pump distribution. Clark et al. (2003) found that sales of treadle pumps by IDE in Bangladesh were 10fold those in West Bengal, India, although these areas are physically and socially very similar. IDE India had decided to manage quality control through centralising pump manufacture and establishing retail chains, whereas in Bangladesh manufacturing was decentralised, with quality monitored through the supply of pump machine tool templates. The result was that in Bangladesh manufacturers were able to innovate with pump types and sizes, and deliver customer-tailored pumps.

DIFFICULTIES, RISKS AND ASSUMPTIONS Availability of working capital for the producer and the price of inputs for the mobile pump are constraints at the production level. The enterprise web in Figure 13.1 does not show the link of producer to microcredit, because it did not occur in our project. Building a large stock of mobile pumps at the beginning of the dry season requires a huge capital. Often, small producers cannot do so due to the shortage of working capital. This causes a backlog of orders by the dealers and as a result an inadequate supply of mobile pumps may occur in the peak season. Pump production costs increased to a great extent. The major raw materials for mobile pumps, namely metal sheets and metal rods, are mostly imported and the prices doubled from 2003 to 2004. As these materials have diversified uses, they are sometimes in short supply and this destabilises the market. A third constraint is access to credit for marginal farmers. The expansion of the mobile pump sector needs to be linked to microfinance activities of NGOs. Rather than providing credit ourselves, which would result in conflicts of interest, IDE addressed this issue by careful engagement with local NGOs from the beginning. Access to credit for both end users and manufacturers will be critical for high volume marketing. In North Bengal, India, van Steenbergen (2003) found that the sales of treadle pumps were hampered by an undercapitalised private sector. But to overcome this, they only experimented with subsidies, not considering private or public credit suppliers. A third constraint is access to credit for marginal farmers. The expansion of the mobile pump sector needs to be linked to microfinance activities of NGOs. Rather than providing credit ourselves, which would result in conflicts of interest, IDE 167


addressed this issue by careful engagement with local NGOs from the beginning. Access to credit for both end users and manufacturers will be critical for high volume marketing. In North Bengal, India, van Steenbergen (2003) found that the sales of treadle pumps were hampered by an undercapitalised private sector. But to overcome this, they only experimented with subsidies, not considering private or public credit suppliers.

CONCLUSION IDE built the capacity of producers, dealers and installers who are now able to produce, market and maintain mobile pumps by themselves. Each indicated that the profit margins were acceptable. Poor women, as well as men, received training in pump operation, repair and maintenance; early signs indicate a positive impact on their livelihood. IDE strengthened the local network by linking skilled artisans to NGOs and local government. The strategic integration of the private and public sector during project implementation is key to the sustainability of the local enterprise web.

REFERENCES Clark, N., Hall, A., Sulaiman, R. and Naik, G. (2003) Research as capacity building: the case of an NGO facilitated post-harvest innovation system for the Himalayan hills. World Development 31(11), 1845-1863. IDE (2003) Poverty Reduction through Irrigation and Smallholder Markets (PRISM). Available at http://www.ideorg.org/downloads/PRISM%20Description.pdf Orr, A., Islam, A. S. M. N. and Barnes, G. (1991) The Treadle Pump: Manual Irrigation for Small Farmers in Bangladesh. RDRS, House 62, Road 7A, Dhanmondi, Dhaka 1209. PETRRA (2003) Proceedings of Regional Workshop, South Central Region. Held September 24th, 2003. IRRI, Dhaka, Bangladesh. van Steenbergen, F. (2003) Creating markets with the poor: selling treadle pumps in India. IIED Gatekeeper Series No. 107.

168

14 Adding Flavour

A value chain approach for aromatic rice

M. Hassanullah Md. Abdus Salam Salim Ahmed Noel P. Magor

SUMMARY APEX, a national NGO, formed a coalition with Mark Industries Ltd., the Bangladesh Rice Exporters Association and poor farmers to plan and implement a sub-project on "Technological Development of Production, Processing and Marketing System of Aromatic Rice in North-West Region of Bangladesh". APEX established a national inventory of indigenous and modern varieties of aromatic paddy. Through on-farm trials, farmers identified the three best performing highyielding aromatic varieties viz. BRRI dhan 34 and 37, and Basmati. APEX formed and trained 67 groups, totalling 330 resource-poor farmers, who may be called upon by any exporter or miller. A contractual marketing system ensured timely supply of desired aromatic rice varieties to millers, exporters, local and overseas consumers. Buy-back guarantee of paddy reduced transaction costs for farmers and middlemen. Farmers also received at least an 18% premium on top of the normal market price, increasing their income by on average Tk 4,100 per acre (US$ 170 per ha). The project helped to increase milling recovery from 52% to 63%, after which two mills also agreed to improve their packaging technology. To strengthen the value chain, an Aromatic Rice Export Clearinghouse (AREC) was established to help farmers, millers, and exporters with advice, logistics, arbitration, market intelligence and market promotion. The following case study uses the enterprise web as a management tool to pinpoint the weakest links in an emerging pro-poor aromatic rice value chain. 169


TECHNOLOGY Situational context Bangladesh has attained near self-sufficiency in rice production, whereas over the years the farm-gate price of rice has declined (Natural Resources Institute of UK, 1996; Project No: R-6689). NRI conducted a study on the prospect of expanding aromatic rice to meet local and export demand. The study focused on traditional varieties like Kalizira, Chinigura and Kataribhog. But these varieties have a problem of low yield and very low milling recovery. In 1998, BRRI developed high-yielding aromatic rice varieties BRRI dhan 34, 37 and 38. Bangladesh still imports Indian and Pakistani Basmati and Thai Jasmine rice; local demand is not met by local production. Therefore with professional production, processing and marketing support for aromatic rice a potential local niche market for local producers may be created. Modern and local aromatic varieties All aromatic rice varieties are cultivated in the rainy aman season (July - November), apart from Basmati, which is grown mostly during the dry boro season (November - May). BRRI dhan 34 is a relatively short grain aromatic variety, which gives 30-40% higher yield than local aromatic rice varieties like Kalizira, Chinigura and Kataribhog. The aroma, other characteristics and market price of milled BRRI dhan 34 is comparable to those of the traditional aromatic rice varieties. BRRI dhan 37 is a lodging-resistant aromatic variety that is photosensitive and suitable for the aman season. It has good-looking grains, a strong aroma and is comparable with Kataribhog, an aromatic land-race of Dinajpur. And it gives 25-37% higher yield. BRRI dhan 38 is very similar to Basmati, but has a slightly shorter grain (Table 14.1 and see Nasiruddin and Ahmed, 2004 ). Production system Common belief holds that mostly rich farmers cultivate aromatic rice, which raises the question as to how the poor could benefit from it. But our socio-economic Table 14.1 Classification of modern and local rice varieties based on grain size RICE CLASS Short

VARIETY

7

Basmati

Adding Flavour

survey in 2001 revealed that resource-poor farmers allocate a greater proportion of their land to aromatic rice than large farmers, 14.6% of their total rice land compared to 9%. They could benefit even more from growing aromatic rice if they were organised as production groups and linked with the markets through a system of buy-back guarantee with millers and traders. Milling system By surveying 14 mills, we learnt that low head rice recovery and poor quality milled rice resulted from poor quality paddy, over drying, one time high pressure milling, over milling, and the lack of a length grader and mist polisher. By upgrading the milling technology, quality of milled rice improved significantly and recovery rate increased from 52.5% to 66.3% (Rickman, 2003; Salam et al., 2004; Salam and Azad, 2004). We demonstrated this modernised system to millers and developed a training programme for them. Through the project they also gained access to improved milling technologies from visits to Thailand. Marketing system A nationwide market survey in 2002 revealed that a family of 5.5 members ate on average 28.1 kg of aromatic rice per year, with consumption increasing in recent years (Hassanullah et al., 2004). Consumption varied among different economic classes: upper class people with an annual income of more than Tk 1,500,000 (US$ 26,400) consumed 58.4 kg of aromatic rice per year, whereas a labourer consumed only 5.5 kg. In 2002, Bangladesh imported 50,000 tons of Basmati. The country exported only about 1,100 tons of aromatic rice from a total production of 486,000 tons. With the projected national consumption being 703,000 tons in 2010, the production of aromatic rice will need to increase to meet demand.

Customers want the best. Increased milling recovery is one of the incentives for millers to invest in modern technologies. By doing so, also the quality of aromatic rice improves, complying with international expectations.

The aromatic rice trade channel is a long and complex one, involving many stakeholders. Our surveys indicated that the major problems for the value chain were: the short-bold grain types of local varieties low yield (1.5 to 2.0 tons per ha) of local varieties poor recovery and quality in traditional milling system uncertain paddy price for farmers lack of contact with overseas markets. 171


ACTORS AND NETWORK The value chain approach required a coalition of farmers, traders, millers and exporters (Table 14.2). This was facilitated by APEX, a voluntary organisation for community development. Other specific members were Mark Industries (Pvt.) Ltd., and the Bangladesh Rice Exporters Association (BREA). The coalition is very single minded about exploring all aspects of the value chain for aromatic rice. Table 14.2 Profile of various actors in aromatic rice value chain ACTORS

TYPE OF ORGANISATION

SKILLS AND ASSETS

MOTIVATION TO CONTINUE COALITION

APEX

NGO

Experience of organising poor for mutual benefits

An important commodity to support rural poor

BREA

Trade organisation

Mobilising exporters for export of aromatic rice

To assists its members for supply of raw materials

MARK Industries

Manufacturing industry

Professional engineering expertise and equipment

To expand business

AWLIA

Engineering workshop

Technical skills

To expand business

IRRI

International Rice Research Institute

High level professional skills

To contribute to future rice development in Bangladesh

Farmer groups

Informal organisation

Land and expertise to produce aromatic rice

To gain economic gain by cultivating aromatic rice

Millers

Industry

Milling equipment

To procure quality aromatic paddy for better milled rice

Exporters

Trading agencies

Trading link and skills

To expand business

AREC

Service agency

Information and business contact

To continue providing services

APEX is an action-oriented organisation that endeavours to improve the socioeconomic condition of disadvantaged rural people and empower women through generation of employment opportunities. APEX chose aromatic rice as a potential commodity that can accrue benefits to the poor and helped to organise poor farmers in groups. APEX also has the ability to link in at high levels in government and to the business community. Mark Industries (Pvt.) Ltd. manufactures a wide range of products from pharmaceuticals to chemicals to agricultural machinery. In recent years, the company showed a high capability to fabricate equipment for urea briquetting, pulping, packaging, rice harvesting, husking and milling. Experienced engineers and technicians provide customer services and undertake R&D;. but not enough for innovating the rice milling system. For this project, they therefore joined with the 172

Adding Flavour

Awlia Engineering Workshop, which specialises in production and installation of various components of automatic rice mills and their maintenance. BREA (Bangladesh Rice Exporters Association) aims to bring production, processing and marketing of aromatic rice under one umbrella, as in Thailand. The organisation works closely with the Ministry of Food and Export Promotion Bureau. It can mobilise its fifteen members to undertake any R&D for the benefit of aromatic rice export. Mr. Joe Rickman, head of the agricultural engineering unit at the International Rice Research Institute (IRRI), Los Baños, Philippines, helped to improve the existing milling efficiency in Bangladesh. The Aromatic Rice Export Clearinghouse (AREC) was established with supplementary support from PETRRA to provide aromatic rice stakeholders with information and advise them on protocol and logistic services, and arbitrate if needed. But also to help out with merchandising, market intelligence, and market promotion. A committee representing APEX, exporters, millers, and research and extension staff has been managing AREC in the premise of APEX, but is moving to new premises shortly.

THE ENTERPRISE WEB FOR AROMATIC RICE Figure 14.1 shows three clusters of activities, covering production with farmers, processing with millers, and marketing with merchants and export association. Each is interdependent but also has specific requirements. And all are in their early stages of development. Production technology

Selection of suitable ecosystem Aromatic rice cultivation requires special agro-ecological niche to ensure better quality and aroma. We identified Naogaon, Rajshahi, Dinajpur, Sherpur and Bogra districts as most suitable for aromatic rice.

Selection of resource-poor farmers As per PETRRA's criteria, we identified farmers who have less than 0.6 ha of land and 3-8 months rice provisioning ability per year. Initially, 36 farmers, men and women, were selected and twelve groups formed to conduct six trials in three experimental sites. Intensive dialogue took place to convince the selected farmers about the benefits of forming groups to make contractual arrangement with millers and exporters. 173



Foundation. seed supply By NGOs

Group level seed production by farmers


Adaptive research on production technology

Provision of credit to RPFs

Selection of RPFs

Crop production skill development

Post-harvest programme

Production of aromatic rice by resource-poor farmers (RPF),men and women, in groups

Mediation grower contracts for best prices

Selling paddy to millers

Provision of credit to millers

Miller skill development

Rice to local retailers Support standards and identify markets

Millers Provision of credit to exporters

National Exporters

Figure 13.1 Enterprise web for mobile pump uptake for poor men and women

174

Rice to international retailers

Adding Flavour

Adaptive research on technology During the participatory variety selection, the twelve groups visited each trial at the growth, panicle formation and harvesting stage. In consultation with the breeder, the group members observed three modern varieties alongside a predominant local variety. After observing their desirable characters, farmers selected BRRI dhan 34 and 37 for aman season and Basmati for boro. Although the aroma of Basmati was higher in aman season, farmers decided to grow it mainly during boro based on its yield performance. Basmati yielded on average from 3.0 to 5.0 tons per ha in aman and boro season, respectively. A dedicated farmer Munshi Shamsuddin said, "APEX motivated me to spare two acres of land for BRRI dhan 34. Due to the bumper yield and price contract with them, I received almost 7,000 Taka more profit compared to Chinigura [variety] last year."

Seed system Once the best varieties were known, we collected foundation seed of BRRI dhan 34, 37 and 38 from the Rajshahi regional BRRI station. Farmers were taught about the production technology of foundation seed of Basmati. To obtain foundation seed, certain conditions are to be met (see Chapter 17). Dr. Mohammad Nasiruddin, former director research of BRRI, taught the beneficiaries the seed production technologies. In the future, farmer seed producer groups could be established to ensure quality seed supply. Various innovative systems are given in Chapters 18 to 20.

Skill development for production and post-harvest handling In 2002, we trained 41 farmers for variety purification and seed production and preservation and developed improved cultivation technologies. BRRI dhan 34 and BRRI dhan 37 yielded 3.0 tons per ha compared to 1.2 tons per ha for local aromatic rice. In 2003, we formed another 55 groups, bringing the total to 320 resource-poor farmers (236 male and 94 female), and trained them in commercial production of these varieties during the following sessions.

Learning from local experts. Involving experienced aromatic rice growers in new producer groups enhances learning among all members.

Investment support for the poor Microcredit from local NGOs or aromatic rice production loans needs to be available to the poor for production and storage. The first loan, granted during the cropping season, should help to pay their operational expenses. But as prices of aromatic rice are at their 175


lowest during the harvesting period, a second injection of financial support would allow farmers to store their rice and sell it a couple of months later at up to an 80% highe price. The length of the loan could even be tailored to link to the higher prices around the time of religious festivals such as Eid Ul Fitr and Eid Ul Azha. Processing

Contract mediation between farmers and millers Contract arrangement for production and supply of desired quality of paddy required an intensive dialogue between the millers and the farmer groups. APEX worked as contract witness and mediator between parties, after having discussed the terms and conditions of the contract (see Figure 14.1). Other contractual arrangements will follow the same approach in the future.

Miller skill development to meet export standards We chalked out a programme and gave hands-on training to 40 millers and technicians in two batches. Mr. Joe Rickman of IRRI, with expertise in rice processing, acted as a resource person.

Investment support for millers This is an essential activity in the total system. Owners of M/S Raj Auto and Rokeya Auto Rice Mills perceived that the improved system not only gave higher milling yield but also higher head rice recovery. They estimated that the extra return was worth Tk 6 (US$ 0.11) per kg of output. Daily output of the two mills was 90 tons and 135 tons, respectively. At 100 tons per day of output, this means they would have US$ 11,000 extra profit per day. So, interested millers should have access to easy term loans for modernising and expanding their old mills (Table 14.3). Marketing Contract production and marketing was an appropriate system for poor aromatic rice producers, who received an 18% premium (Tk 472 per maund compared to Tk 400 Table 14.3 Investment analysis for upgrading rice mills (at 100 tons per day) INVESTMENT ANALYSIS

176

Break-even point sales volume (annual)

1,105 tons (18% of capacity utilisation)

Break-even point sales value

Tk 26.5 million (US$ 466,000)

Pay-back period

Four years from the date of operation

Net present value

Tk 41.8 million (US$ 736,000)

Internal rate of return

30%

Adding Flavour

per maund; one maund equals 37.3 kg). It enabled many marginal growers to produce a quality product on demand. Information and requests to comply with specific quality standards could be easily conveyed from the export market back to the millers and through to growers. This feedback mechanism allowed for a swift response, only because farmers were organised in groups. This is one of the reasons why working through NGOs who can coordinate many farmer groups offers extra advantages compared to working with large, but individual farmers.

Market exploration We identified potential markets in Europe, the Middle-East, Africa and Southeast Asia. Supported by donor agencies and the Export Promotion Bureau, a series of trade missions was planned to establish contacts with export markets and make detailed feasibility studies.

Aromatic Rice Export Clearinghouse AREC played a critical facilitation role in drawing growers, millers and marketers together so as to achieve a desired volume of exports and thereby address a niche in the market in a coordinated way (see Figure 14.1). AREC may eventually be transformed into an organisation of aromatic rice growers, processors and market agencies.

Investment support for exporters

Attracting attention. Making attractive packages is part of the marketing strategy, as Bangladeshi aromatic rice will be on the shelf competing with this of other countries.

In neighbouring rice exporting countries such as India, Thailand and Pakistan, the government keeps a buffer stock of exportable rice to facilitate exporters. Exporters are also adequately financed to buy large stocks. In Bangladesh, as export of aromatic rice is an emerging business, exporters do not have many overseas contracts yet and hence are uncertain about the size of orders they will get. But once they receive an order, they need to respond swiftly, not let importers wait. Timely supply of quality product is of critical importance, especially in the early stages of building international relations. Hence, adequate financial support should be available to the exporters to buy in bulk in accordance of the value of Letter of Credit, send by the international clients. Awareness raising and policy advocacy One of the major tasks was to create awareness about the new technologies and products. Based on the initial trial results and with support from the agronomist working in the project, we published a leaflet in Bangla on aromatic rice 177


cultivation and distributed this among farmer group members. Extension staff of the Department of Agricultural Extension and NGOs viz. RDRS, Proshika and RSDA working on aromatic rice also received copies. We also produced a leaflet on exportable Bangladeshi aromatic rice highlighting the qualities and grades, and handed them out to traders and exporters. APEX in collaboration with BREA participated in the international rice congress held in Beijing, 2002. It sponsored a Bangladeshi stall and displayed samples of Bangladeshi aromatic rice. Publications were distributed among congress goers to create international awareness among potential buyers. A video programme was made and a book published to further increase awareness among stakeholders and the general public at large. We conducted eight regional workshops on production and milling, in which about 500 representatives of all stakeholder groups participated. A national dialogue was held for policy makers to adopt appropriate strategies for promoting export of aromatic rice. Training Through study tours and trials, we exposed groups in which all stakeholders were represented to the aromatic rice production, milling and packaging technologies. To establish good mutual understanding, millers and exporters also took part in 41 demonstrations for farmers. Similarly, when we showed the improved milling system to potential millers, we also invited farmer group leaders and traders. These activities helped in building system transparency and mutual understanding of the complexities of the value chain. The sub-project dealt with commercial entrepreneurs who were involved in production, milling and trading. Although some common problems emerged, each actor faced quite unique technological problems. High-level professionals and technicians provided services for personal guidance and counselling to farmers, millers, traders and exporters.

KEYS FOR SUCCESS Farmer contracts Buy-back price guarantee given to the beneficiaries motivated them to actively participate in the crop and seed production training and ensured maintenance of field standards. Technical links for upgrading processing and marketing From the inception of the sub-project, poor head rice recovery and quality of milled rice were perceived as major problems for promoting aromatic rice. As 178

Adding Flavour

technical expertise was not available in Bangladesh, input from Mr. Joe Rickman of IRRI was significant to the success of the project. AREC aimed to establish a continuing relation with IRRI and other international institutes dealing with livelihood improvement. Partnership between key actors For aromatic rice, the key actors are the producers, millers and exporters; and direct participation of all is a prerequisite for success. In this sub-project, the NGO APEX organised farmer groups and brought the different actors together, but in future any NGO with an interest in agriculture could take the lead and help negotiating contracts between millers and farmers. At the production end there is an interchangeability of agents for securing contracts. Mark Industries and Awlia Engineering are private businesses, who expanded their markets by selling innovative machinery for improved milling and by providing maintenance support. The Bangladesh Rice Exporters Association assisted the exporters for marketing and exporting of aromatic rice.

POTENTIAL PITFALLS The institutional base of resource-poor producers In the second year, the number of producer groups increased to 67 for trial commercial production. A contract arrangement was made with a miller and an exporter to provide buy-back guarantee, but we fear that two cropping seasons is too short to develop a strong institutional base of producer groups. In future, it would be better to train existing groups under local NGOs viz. RDRS, Proshika and RSDA or community-based organisations and link them to millers who in turn are linked through AREC to exporters. Examples of existing farmer seed producers are given in Chapters 18 to 20. In the development of the market chain, inclusion of resource poor farmers must be purposeful. The transaction cost of millers to work with poor farmers at first sight appears to be higher compared to dealing with large farmers. However, when working through NGOs who can coordinate groups, the reverse may be true. But a social capital development cost is needed to build the local level organising capacity, which is required to draw up and comply with production and marketing contracts. It is important for this to be self-financing. A sustainable cost-sharing mechanism developed by the RDRS federations could serve as an example (see Chapter 20). Millers received an extra Tk 6 per kg of output (see above) and growers receive 18 percent premium. Maybe a better contractual arrangement can be made for farmers so that they receive part of the Tk 6 per kg extra that the millers obtained. It is here that a complete financial and transaction cost analysis is needed of each activity in the enterprise web. 179


Institutional development of BREA BREA is a key actor in marketing aromatic rice. We used the platform, but did not foresee the need to build capacity of BREA for long-term sustainability. Although it brought exporters together in a common platform, in 2004 it still had to establish its legal status through registration with the government authorities. Immediate attention is needed to reinforce this institution for exporting rice. To strengthen BREA's operations and service delivery to all its members, not only aromatic rice producers, AREC could operate as a cell in BREA's premises to reduce management and operational costs. Access to international market Modernisation and expansion of the aromatic rice industry depends on access to international markets. Unless contracts are established, it will be difficult to expand production and improve the milling system. This sounds like the chicken or the egg riddle, but reality tells us that the bulk of increased production will not be absorbed in the local market and increased cost of milling will not be recovered by selling to the local market. Hence, links should be further developed with the international market. Bangladesh competes in the export market with India and Myanmar, both of whom provide export subsidies (Hossain and Deb, 2003). Although Bangladesh has recently introduced export subsidies for agro products this does not include aromatic rice (FE Circular, 2004). Government support for this industry in its infancy is necessary. Apparently, the value chain approach for aromatic rice could benefit resource-poor farmers. But how to develop the capacity of the mills to meet standards for local and export markets needs to be explored. Export substitution should be possible but will be dependent on achieving consistent standards.

CONCLUSION The advancement of resource-poor farmers through production of aromatic rice is dependent on developments in production, processing and marketing technologies and institutions. There are essential upward linkages to businesses and marketing houses. The formation of AREC is an important key in building volume of trade and ensuring standards. The enterprise web is in its early stages of development and does require consolidation.

REFERENCES FE Circular No 05, July 25, 2004. Bangladesh Bank. Hassanullah, M., Salam, M. A. and Hossain, S. (2003) Aromatic Rice of Bangladesh: Market Survey Report. APEX, Apt. 6/A, Mirpur Lodge, Hse 14, Avenue 1, Block A, Mirpur 180

Adding Flavour

10, Dhaka 1216 and IRRI, Dhaka Bangladesh. Hossain, M. and Deb, U. K. (2003) Liberalisation of the crop sector: can Bangladesh withstand regional competition. Centre for Policy Dialogue-IRRI Policy Brief No 4. Centre for Policy Dialogue, GPO Box 2129, Dhaka 1000. Nasiruddin, N. and Ahmed, S. (2004) Farmers' Recommended Varieties of Aromatic Rice in Northwest Region of Bangladesh. APEX, Apt. 6/A, Mirpur Lodge, Hse 14, Avenue 1, Block A, Mirpur 10, Dhaka 1216 and PETRRA-IRRI, Dhaka, Bangladesh. Rickman, J. F. (2003) Rice Milling in Bangladesh. Power point presentation held February 15, 2003 at Retired Army Officers Welfare Club, New DOHS, Mohakhali. Salam, M. A. and Azad, S. Z. (2004) Improved technology for production, processing and marketing of aromatic rice in Northwest Bangladesh. In: Workshop Proceedings: Emerging Technologies of the PETRRA Sub-projects. Organised by PETRRA-IRRI and BRRI, at BRRI Auditorium, Gazipur. Salam, M. A., Azad, S. Z. and Ahmed, S. (2004) Evaluation Report on Technological Development of Production and Marketing System of Aromatic Rice in North-west Region of Bangladesh (sub-project number: 29 02). IRRI, Dhaka, Bangladesh.

281

Part V: Pro-poor seed systems

15 Innovations in Seed Systems

Solveig Danielsen Md. Khairul Bashar Mark Holderness

INTRODUCTION Seed is an important vehicle for technological innovations. And good seed is one of the first and foremost prerequisites for a healthy and vigorous crop. However, seed is more than an agricultural input and a source of new technologies. A range of developmental, socio-economic, environmental and political issues are also covered by seed: food security, natural resource management, agrobiodiversity, intellectual property rights, social dynamics, gender, cultural and religious dimensions, policies and regulations. A strong and resilient rice seed sector that addresses the specific needs of poor farmers for timely access to seed of appropriate genetic, physiological and phytosanitary quality plays a crucial role in ensuring food security and rural economic development for the ever growing population of Bangladesh. Seed security is particularly important in Bangladesh, where natural disasters occur frequently, resulting in total loss of farmer seed stocks in some areas. The complex of problems that farmers face in their livelihoods calls for a holistic approach to delivery of rice seed that builds on the comparative strengths and advantages of stakeholders involved in generation, delivery and uptake of rice technologies. Two factors have been pivotal for the emergence of innovative pro-poor rice seed systems in this country: 1) the new seed policies that seek to promote diversification and decentralisation of seed production and 2) the poverty-orientation of the PETRRA project and its strong focus on creating innovative mechanisms such as strategic networks for effective delivery and uptake of rice technologies. 185


The challenge In farmer consultations under PETRRA, limited access to good seed came up again and again as a key constraint. However, formal rice seed supply to resource-poor farmers is particularly challenging. Rice is a low-value self-pollinated staple crop, and rice seed can be produced relatively easily and stored onfarm over a number of cycles without genetic degeneration. In such a system it is difficult to create a stable demand for external seed supply; private companies in Bangladesh have so far been uninterested in this area because the profit margin is low and the demand unstable. Although 65% of Bangladesh's rice area is planted in improved high-yielding varieties, resource-poor farmers have limited access to external sources to renew their seed. Renewing seed is important because rice yields often decline after 4-5 years of recycling. The seed gets 'tired'.

Rice yields often decline after 4-5 years of recycling, in part due to accumulation of seed borne diseases, weeds and genetic impurities. The seed gets 'tired' and needs renewal. The sporadic demand from farmers for new seed, the lack of private rice seed suppliers and the limited ability of government institutions to ensure timely delivery of affordable seed to the poorest farmers mitigate against sustainable external supply systems. Consequently, farmers do not have timely access to quality seed and do not benefit from technological innovations that may improve their livelihoods. Evidence from around the world shows that different ways can be designed to support local seed systems (Louwaars, 1994; Tripp and Pal, 2001). Sustainable, small-scale seed enterprises that address the needs of small-scale farmers are hard to establish (Kugbei et al., 2000) and once external support to such systems stops, they tend to collapse (Tripp, 2001; Almekinders and Thiele, 2004). One of the arguments is that it is almost impossible to combine development goals with profitability of the supply systems. A number of sub-projects under PETRRA have taken up the challenge to develop innovative, poverty-oriented rice seed supply systems. Five of these will feature as case studies. Historical context Because of the importance of rice to the nation and to the poor, there has always been a strong government role in rearing and distributing rice seed. This has been centralised through the Bangladesh Agricultural Development Corporation (BADC) under the Ministry of Agriculture, mandated to ensure the effective delivery of good seed to farmers. Until recently, BADC only worked with the public research

186

Innovations in Seed Systems

sector such as the Bangladesh Rice Research Institute (BRRI), and the Seed Certification Agency (SCA), to produce rice seed according to guidelines and policies of the National Seed Board. BADC only supplies around 5-6% of the total rice seed demand and the poorest are seldom reached. Most farmers save their own seed or acquire it through informal trade and exchange. In recent years, questions also arose as to the sustainability and efficiency of government-supported rice seed delivery and control systems. Complaints about the inadequate supply and quality of seed through the subsidised and centralised system were common. Various NGOs and private companies sought licence to produce and sell seed themselves, often as part of credit and input supply systems (Huda and Smolders, 2001). Discussions around these themes in the early 1990s led to the formulation and approval of the national seed policy in 1993, and signing of the national seed act in 1997. The policy was established to: "make the best quality seed of improved varieties of crops conveniently and efficiently available to farmers, with a view to increasing crop production, farmers' productivity, per capita farm income and export earnings" Significant changes in the seed policy included: the reorientation of BADC to operate on a commercial basis to the extent possible; the ability of other organisations to register as seed dealers; and the introduction of voluntary certification and the category of truthfully labelled seed. The certification of breeder seed and foundation seed remained compulsory and the SCA was recognised as the only mandated agency to do so. Rice varieties remain notified, such that new varieties have to undergo extensive testing to obtain approval for release. Certified seed needs approval of the SCA. Without certification, seed may be labelled and sold as truthfully labelled seed at one's own responsibility. If a customer is not satisfied with the seed, the actor who labelled the seed can be legally prosecuted. If sold by farmer seed producers, as with the Farmseed method described in Chapter 18, it is termed quality seed. So these three types relate more to legal accountability than actual quality levels. These changes in policies and regulations have not always been straightforward and reservations have been expressed about the quality assurance systems and public accountability of a private-sector supply system. Converting the SCA from a controlling to a service-oriented institution is a major change as well. PETRRA case studies Given the new legal framework for seed supply in Bangladesh, PETRRA took up the challenge to stimulate innovations in seed systems, as described in the case studies in this part of the book. The first two cases (Chapters 16 and 17) describe a new approach to technology

187


identification and the development of a rice seed network to ensure effective communication and coordination between the diverse actors in the system. The next three cases (Chapters 18 - 20) examine new models for improving farmers' access to quality seed. These are hereafter referred to as Farmseed, developed by AAS (Agricultural Advisory Society); Grameen Seed, developed by GKF (Grameen Krishi Foundation); and the federation model by RDRS (Rangpur Dinajpur Rural Service). Common to all three systems is that poor farmers were selected to receive training as seed producers and seed provision was uncoupled from credit (Table 15.1). Table 15.1 Seed innovation systems under PETRRA SYSTEM VARIABLES

FARMSEED

GRAMEEN SEED

RDRS FEDERATION MODEL

Variety selection

18 varieties demonstrated; 3 were selected by farmers

Varieties selected by GKF based on knowledge of local preferences

Local market survey; federation seed committee gathers information from its members

Amount and price of foundation seed supplied to producer

3 kg per farmer Tk 18 (US$ 0.32) per kg

At least 2 kg per farmerTk 18 per kg

10 kg per farmerTk 20 (US$ 0.35) per kg

Type of seed produced

Quality seed

Truthfully labelled seed


Value addition

On-farm drying and storage

Drying, cleaning, storage and packaging at the GKF processing unit

Drying, storage and packaging at the federation processing unit

Quality control

Germination test on-farm; no label

Seed testing (germination, moisture, purity) done at the GKF processing unit; labelling

Federation seed committee conducts field inspection and seed testing (purity, germination, moisture); labelling

Marketing strategy

Seed producers sell and exchange their own seed

Seed producers are given first priority to buy back truthfully labelled seed from GKF for resale in the community; retailers sell remaining seed at local markets

The federation seed committee buys truthfully labelled seed from seed producers for sale from its premises and local markets; promotion with banners at bazaars, banks etc.

Client groups

Farmers in the community; local markets at up to 20 km

Farmers in the community; local markets

Federation members; local markets

Selling price*

Tk 10-15 (US$ 0.180.26) per kg

Tk 16 (US$ 0.28) per kg

Double grain price, ~ Tk 16 per kg

* Grain price is Tk 8-9 ($0.14-0.16) per kg. Price of truthfully labelled seed from the governmental BADC varied from a minimum of Tk 14.5-16 and Tk 18 (data 2003-2004).

188


In this chapter we will bring out some generic lessons from the case studies, on how they aim to improve production and deliver quality rice seed to the poor. Keys for success and the challenges of institutionalising and sustaining new approaches will be emphasised and the Bangladeshi experiences compared to those from other countries.

VARIETIES AND SEED-BASED TECHNOLOGIES Around 65% of Bangladesh's rice area is planted in improved high-yielding varieties (HYV) (Hossain, 2002). Most of these are developed for irrigated systems. Over the last decade, around 15 new varieties have been released. BR 11 for aman (July November) and BRRI dhan 28 and BRRI dhan 29 for boro season (November May) have become popular with farmers. The lengthy variety release procedures, the inadequate participation of farmers in variety testing and selection, and the lack of improved varieties with biotic and abiotic stress tolerances are some of the major constraints to expanding the rice seed markets (Hossain, 2002). Rice biodiversity The widespread adoption of high-yielding varieties has led to the replacement and erosion of indigenous rice genetic resources in Bangladesh (Bashar and Sarkar, 1997). Local varieties have considerable importance in self-sufficiency and as a source of desirable agronomic, nutritional and culinary traits; some of them even give comparable yields to modern varieties. Local varieties are adapted to specific local conditions, e.g. saline or deep-water areas. Sometimes no improved varieties are available for such conditions and farmers rely entirely on their own varieties. Some RDRS federation members, for instance, grow local deep-water varieties beto and darash as there are no improved varieties for these conditions. Some indigenous varieties are being exploited as high-value aromatic rice (see Chapter 14). Bangladesh, being close to the centre of origin of rice, has a national rice gene bank that contains around 5,500 partially characterised accessions. Forty improved varieties have been released by BRRI since 1970 and of these only eight are widely adopted (Hossain et al., 2002), suggesting that there may be a large under-exploited pool of rice genetic diversity in Bangladesh. There is an increasing awareness among BRRI researchers and others about the potential value of this national heritage. The involvement of BRRI in collecting and characterising local rice biodiversity is described in Chapter 17 and by Bashar et al. (2004). Local varieties are not addressed in the case studies. In another PETRRA subproject on rice diversity in southwest Bangladesh, around 200 local landraces were collected from four districts, ranked and further selected according to people's criteria and preferences. These could potentially feed into the rice seed network, and give farmers more options to choose from.

189


Identifying appropriate technologies Farmers require varieties and seed technologies that fulfil their needs and are appropriate to their conditions. Research and development of varieties and seed technologies in the past was top-down. Government extension systems often offered pre-selected varieties and technologies to farmers via demonstration plots, and training and visit (T&V) extension. While developing new seed systems under PETRRA, scientists have increasingly recognised that an integrated effort by scientists, extension workers and farmers in technology development and evaluation is an important precondition for successful adoption and feedback. Following up on participatory variety selection work by breeders at BRRI (Salam et al., 2002), its Adaptive Research Department established action research groups with poor farmers, NGOs and government extension workers to select and validate a broader range of technologies, not just varieties. Adoption rates increased, feedback mechanisms were established and farmers gained self-confidence as decision-makers. This model has not yet influenced the new seed models addressed here, but in the future it probably will. RDRS has been a keen participant in the action research groups and the positive experiences are now being taken to the federations to enhance uptake of new technologies. Collaboration between NGOs, community-based organisations, private sector, and government research and extension has been institutionalised in the focal area forums, endorsed by the state minister for agriculture (see Box 21.1). This is just one example to illustrate the value of cross-fertilisation of ideas between different PETRRA sub-projects. These participatory principles of stream-lining variety selection, breeding and approval processes by early engagement with communities have been used successfully in participatory wheat variety selection by CIMMYT and their partners in India, Bangladesh and Nepal, and on various other crops elsewhere (Elings et al., 2001; Ortiz-Ferrara et al., 2001; Bellon and Reeves, 2002). Farmseed worked with farmer groups to ensure the best match to local needs and likes, and from 18 varieties, three were chosen for multiplication. The selection of varieties by GKF and RDRS was based on thorough knowledge of local preferences and local market surveys, respectively. Local demand and preferences are more easily captured by local organisations than by large seed companies and BADC.

DEMAND AND FARMER AWARENESS The driving force behind sustainable seed supply systems is stable demand for seed and access to outlet markets. When farmers produce entirely for their own household, as is the case of most Bangladeshi rice farmers, access to grain markets is unimportant. The PETRRA stakeholder consultations confirmed this: farmers gave

190


lowest priority to harvest prices. However, in the long-term, given the increasing urbanisation of Bangladesh, there is likely to be a shift towards market-oriented agriculture. Profitability, rather than subsistence will increasingly become more important for stimulating production (Hossain, 2002). Once food security is ensured, access to stable outlet markets will be an important incentive for farmers to invest in quality rice seed. The government deliberately keeps prices of paddy low since the landless and urban poor depend on cheap rice. This could potentially discourage rice seed and grain producers. Even Bangladeshi rice farmers who sell little rice have many incentives to take up new, appropriate rice technologies. An increase in productivity leads to: less dependence on purchased rice, greater disposable income, more efficient land use, release of labour for other purposes, and improved diet, health and children's education. Recognising the value of good seed The basis of demand for quality seed is recognition of its value, whether purchased or produced on-farm. Farmers' perception of improved seed and varieties is highly variable, ranging from specialised knowledge based on the experiences of generations to lack of attention to even the basic features of seed quality. Generally, farmers have a notion that seed needs to be renewed periodically and that improved varieties give a number of benefits, but they often undervalue maintaining good seed to enable expression of its genetic potential. "First you have to raise value in peoples mind," says A. K. M. Zakaria from the Rural Development Academy in Bogra. For the last five years he has been training farmers, men and women, in improving the quality of their farm-saved rice seed under the PETRRA Seed Health Improvement sub-project (SHIP). Farmers who realise that good seed gives vigorous plants and good yield are in a position to judge whether his or her scarce capital should be invested in improved seed. Today most farmers do not have a real choice, since many have no place to buy quality rice seed. Assuring timely supply of affordable seed All PETRRA seed case studies mention two consistent keys for success: timely delivery of good seed and affordable price. The main problem with seed supply through BADC is that seed of the right variety is often not available at the right time or that the poor have no access to them at all. The new seed production models have a built-in distribution system through local seed producers or community institutions. This lowers transaction costs for poor farmers and helps ensure timely delivery. In the Farmseed model, production and distribution are totally integrated. The seed producer undertakes production, processing, quality control and sales in the local community (Table 15.1). The client and the supplier are neighbours, making it easy to

191


gauge demand, sell at competitive prices and enjoy trust in the quality. In the Grameen Seed model, farmers also act as producers and distributors, buying back processed and packaged truthfully labelled seed from GKF in any desired amount for resale in the community. Retailers sell the rest on local markets. Its price equals market prices, Tk 16 (US$ 0.28) per kg. GKF strives to stimulate exchange of information and experiences at union level in order to establish trust and transparency. In the RDRS model, federation members have first priority to buy seed from the federation seed store at market price, about Tk 16 per kg. Federation members say that the new supply model has changed seed acquisition strategies for some. Recognising that RDRS seeds yield better than their own, federation farmers are willing to buy seed more frequently because the price is acceptable and availability is high. A common strategy is to purchase one 10 kg bag of rice seed, the maximum affordable amount for many, and supplement with their own if needed. Demand and trust is partially established by multiplying seed and training seed growers; other farmers from the community are invited to come and see variety performance, impact of seed quality etc. Seed is brought to the doorstep of the farmers on time, fulfilling one of the ambitions of the rice seed network. Resource-poor farmers are seed producers, retailers and the principal client group. Farm-saved seed: scope for improvements The advantages of attention to seed quality in farmer-saved rice seed are often neglected and there is scope for increasing farmer awareness of the value of good seed. SHIP has paid particular attention to this, using participatory learning and action research to improve farmers' awareness of good seed and the ways of producing it with low levels of inputs. The SHIP experience shows that, when aware of its potential, farmers value good seed and are willing to pay a premium price for it from a reliable source. Besides having a 5-15% yield increase, seed producers sell their own-processed seed to other farmers in the village at Tk 10-12 (US$ 0.18-0.21) per kg, up to 50% higher than the grain price. Participatory learning improves farmers' awareness of good seed. Trained farmers demarcated their seed plots to increase community awareness, engaged in farmer-tofarmer extension and Going Public (Chapter 9), and in the production of videos for further scaling-up (Chapter 5).

192

Stimulating formal seed supply for farmers while enabling them to produce better seed themselves gives farmers the option to select whatever source of seed is best for them at a given time. Under unfavourable conditions, poor farmers may have to


eat rice seed saved for the next season, forcing them to acquire seed from whatever source is available. For them, enhanced capacity to produce and store quality seed is of particular importance (Almekinders and Louwaars, 1999). Matching seed and outlet markets - the long-term perspective PETRRA seed sub-projects do not specifically address the potential of seed and grain markets. Currently, in Bangladesh the demand for quality seed is so high that once it is available at the right time at an affordable price "the seed almost sells itself" as one BRRI researcher said. Statistics from the rice seed network indeed confirm that there is a large demand for quality seed and that decentralisation is a promising way forward. The estimated increase from 5 to 15% in formal sector seed supply in five years is impressive. As many farmers, even in fully developed markets such as Europe, continue to save seed for some years, formal sector supply of rice seed in Bangladesh may have a maximum potential of around 30% by 2020, based on a projected decrease from 5 to 3 in average number of years over which seed is recycled (Talukder et al., 2004). Once the formal rice seed market becomes saturated, market analyses and planning will become more important to ensure an even, regional and national distribution of seed. Market assessment through the rice seed network (Chapter 17) and focal area forums (see Box 21.1) will enable organisations to coordinate production and distribution.

QUALITY ASSURANCE Bangladesh is a good example of how new directions in seed policies and regulations can stimulate the formal seed system by allowing new players to produce and distribute seed. In most countries, the regulatory system is a 'policeman' imposing quality standards and control measures that the government system cannot comply with itself. The seed systems introduced by PETRRA help ensure the supply of truthfully labelled seed of improved varieties to poor farmers. The quality assurance lies in the continuous supply of breeder and foundation seed from trustworthy sources (BRRI and BADC), in the processing and in the technical training and backup that seed producers and others receive through the projects. Quality assurance and trust is the responsibility of all players in the seed chain. Local organisations are important for assuring quality. NGOs and community-based organisations (CBOs) often have their own well-defined clientele group, which they are mandated to support and to whom they are accountable. A big part of the trust in the system lies in the confidence between the local organisations and their members. Community-based quality control is powerful where producers and supplier can be readily identified. In South America, this has also been called

193


'neighbourhood certification' (Iriarte et al., 2000). Increasing distance from the seed source makes other forms of quality assurance necessary. In the Farmseed and Grameen Seed models where production and marketing is integrated, the social pressure on the seed producers ensures quality control: once their colleague farmers lose trust in the quality of the seed, they will be out of business. Under the RDRS federation model, the distance between seed producer and client is slightly increased, as seed is sold by a federation seed committee. If there is only one local seed supplier, there is no competition and the market cannot ensure quality. The case studies illustrate how three leading NGOs diversified farmers' sources of seed. Farmseed opted for groups of 20 small-scale seed producers per village; GKF has groups of 10 seed producers/sellers per ward (1-3 villages), while RDRS sells through their federations seed committees. While seed can be tested locally for germination, moisture and purity by NGOs or community-based organisation or on-farm with simple equipment, seed health testing requires more advanced facilities. The lack of routine testing of breeder and foundation seed by the national seed certification authorities is a weakness in the system. Seed health testing is only done at the Bangladesh Agricultural University and only upon request. The lack of systematic surveys for seed-borne diseases may hamper seed quality and lead to epidemics. In Tanzania, official seed certification authorities perform spot-checks of the decentralised seed production (Box 15.1). Box 15.1 Quality Declared Seed in Tanzania

The government of Tanzania promotes on-farm production of quality declared seed (QDS) to improve smallholder farmers' access to quality seed through decentralised production (FAO, 1993). QDS farmers receive government training in seed production and management: roughing, variety maintenance, disease management, harvest, seed processing and storage etc. Quality control is done on-farm by the farmers, while the official seed certification authority performs spot-checking of about 10% of the seed and occasional field supervision. QDS farmers receive fresh foundation seed every second year. Since the system started in 1998 many poor farmers in remote villages have been enabled to produce good seed and supply neighbouring farmers, using an approach similar to the Farmseed model in Bangladesh. Success depends on the crop, location, the farmers' ability to produce good seed, and the certification authority's ability to maintain the service. QDS has upgraded the quality of farmsaved seed and the trust in QDS seed is higher than in other purchased seed. QDS farmers are accountable to their neighbours/ clients for the quality of their seed. Among the problems that still remain to be solved is how to make the system sustainable in a large country like Tanzania with many remote villages.

COMMUNICATION AND PARTNERSHIPS The PETRRA case studies show the importance of fruitful partnerships between public, private and community-based organisations. Technical support from BRRI 194


contributed much to the success, and the NGOs acted to complement, rather than substitute for state institutions. Farmers valued the seed grown by trusted neighbours. In Brazil, successful cases represent a mix of local government initiative, civil society and a supportive central government (Tendler, 1997). Farmseed emphasised networking of skills and resources among NGOs and community-based organisations. The network partners consider training their staff on rice and seed production as incentives for participation. Farmer trust and confidence in the system is strong in Farmseed. AAS built a vast network with local NGOs and community-based organisations, with hardly any communication gap between the partner organisations and the farmers. AAS has shown an extraordinary ability to mobilise and enthuse these organisations. AAS staff have been inspired by their own work and this spreads to their partners. All the positive achievements of Farmseed come back to their ability to engage pro-actively with other organisations and institutions, capitalise on their local network and experiences and stimulate collaboration. This mentality is also reflected in the flexible and lean structure of AAS (Chapter 18). GKF established a network of local leaders, field staff or block supervisors of the Department of Agricultural Extension (DAE) and NGOs to bring in new customers and provide information about seed and varieties. These actors, along with other farmers from the village, are invited to field days. This creates demand and announces which persons are the sources of trustworthy seed. Visits to the GKF seed processing centre also help raise awareness about the added-value of seed processing. The long-term objective of this new GKF approach is to reach all farmers in any union where the model is introduced (Chapter 19). A RDRS in Rangpur worked with federations to revolve inputs from participating farmers and to link with BRRI and DAE block supervisors for technical support and training. RDRS became a partner with BRRI to conduct action research in farmers' fields to validate new varieties and to experiment with other rice-related topics involving students from the Bangladesh Agricultural University. The federations now serve as a forum to gather feedback, narrowing the gap between research and adoption of new technologies. The RDRS federation model allows for efficient exchange of information and seed between federations, and creates the opportunity to establish a business (Chapter 20).

"The kitchen is our school now. We learn from other women (extension agents) when they train us in our homes."

Relationships between government research, government extension and NGOs are not

195


always clear and there can be competition in roles. Establishing true partnerships, building on complementary strengths, takes time and effort. Transparency is crucial. The studies presented illustrate the importance of communication and transparency for networks and partnerships. Success is rooted in the clarity of purpose in each organisation, and in enthusiastic staff. The importance of individual fiery souls as major driving forces cannot be over-emphasised. The rice seed network provides a loose umbrella for diverse institutions, and allows several seed supply models and partnerships to be applied. This is the beauty of the system: how communication, coordination and flexibility can be ensured, is addressed in Chapter 17.

MAINTAINING THE INCENTIVES For the system to work and expand, all links in the seed chain must be operational. The strengths and weaknesses of the seed chain as experienced by the PETRRA subprojects are summarised in Table 15.2. These mainly deal with the technical aspects of institutionalising the seed system. Fundamental to these innovations have been the changes in mentality and ensuring that all actors maintain their benefits, shared responsibility and accountability. PETRRA has been instrumental in shifting mindsets, with the initial stakeholder consultations as a first step and eye-opener. Farmers As the beneficiaries of the new seed systems, poor farmers have an incentive for commitment once they recognise the benefits. The needs of the clients must be considered, as local demand for quality seed will be the first incentive for seed producers. The emotional value of farmers towards their own rice seed and the growing feeling of ownership also help guarantee that the innovations will last. The case studies describe this in various ways: 'Resource-poor seed producers are freelancers, not contractual labourers. They are directly reaping the economic benefits from their work and the social benefits from becoming respected community members' (Farmseed). 'The training provided to seed producers does not only benefit seed production, but also their crop production overall. Higher yields and better seeds help raise incomes. Also, women feel recognised for their important role in rice production' (GKF). 'Building on community institutions, farmer seed producers and farmer trainers has increased efficiency of human, social and physical capital. Ownership is ensured by giving the civil society a voice in project decision-making. Creating a conducive learning environment for women has given them more voice in rice seed cultivation and agriculture' (RDRS).

196


NGOs The advantage of NGOs and community-based organisations in a seed system is that they are not driven entirely by profit, the way seed companies are. Most of them Table 15.2 Strengths and weaknesses of rice seed supply systems in Bangladesh, 2004

1

SYSTEM VARIABLES

STRENGTHS

WEAKNESSES

Conservation and use of germplasm

BRRI1 holds the national rice gene bank of ~5500 accessions of Bangladesh origin; BRRI ensures ready access to germplasm

BRRI's collection and evaluation of local germplasm is outside the mandate of the rice seed network

Breeding/variety selection

BRRI increasingly involves NGOs to identify technology identification with farmer groups

Most partners do not select varieties, but use 2-3 varieties already widely available in the area; farmer participatory variety selection still limited; delay in variety release

Production of breeder seed

BRRI researchers are highly skilled and motivated; BRRI can multiply any variety or accession in small quantities on request; current level of production can be sustained by BRRI after PETRRA

The rice seed network depends on breeder seed supply from BRRI, which relies on government support and commitment; BRRI not allowed to recover costs from breeder seed production, may hamper motivation; limited buffering from natural disasters

Multiplication of foundation seed

More producers since policy reform, more coverage and flexibility; production by NGOs is incentivebased; BADC supports development of new pathways

Supply system needs improvement and strengthening; vulnerable to natural disasters, potentially reduces buffer capacity of BADC

Multiplication of truthfully labelled seed

Many organisations involved at various levels; strong local anchoring, feedback and interaction; low transaction costs

Risk of fake seed producers accessing the market; potentially unequal competition from BADC due to government subsidies

Processing/storage

High capacity of BADC; NGOs pay service fee to BADC and other organisations for seed processing; constructive interaction among service suppliers improves national coverage

Capital costs constrain alternative service providers; processing and storage capacity may hinder increase in seed production; risk of losing trust and transparency if seed is processed in a centralised unit

Quality assurance

Rice seed network partners receive training in seed production and management; social capital of local NGOs, CBOs or small private seed producers ensures trust; some NGOs have a long record of seed production and quality control, so their seed quality is much better than farm-saved, esp. in the dry season

No seed health testing in the official seed certification scheme; staff turnover at SCA due to structural problems; capacity of SCA to assess all requirements for breeder and foundation seed; quality assurance system for truthfully labelled seed depend on market forces and self-regulation

Distribution/marketing

Positive effect of selling less foundation seed to seed farmers; production and distribution networks close to the client, seed farmers participate in marketing; rice seed network is framework for assessing seed markets

Limited business and organisational skills of many organisations; limited market analyses and coordination within and among organisations

Outlet markets

Potential for linking seed networks to grain markets

Not addressed in rice seed network. Currently beyond its scope

The Bangladesh Rice Research Institute (BRRI), Bangladesh Agricultural Development Corporation (BADC) and Seed Certification Agency (SCA) are governmental institutions.

197


such as through microfinance. PETRRA gave partner organisations free training, which helped motivate NGOs to consider agriculture as a tool of social have a broader mandate and are committed to their clients, allowing them to concentrate on the needs of the poor. In spite of the importance of agriculture to Bangladesh, few NGOs in Bangladesh have much agricultural expertise; they developed from a basis of social improvement, development. Continued commitment to seed supply will rely on the ability of the NGOs to recover their costs. Under Farmseed, AAS did not recover costs, but they currently explore the production and sales of foundation seed as a way of maintaining income once donor support ends. The business plan of GKF includes full recovery of costs for training by slightly increasing seed prices. Mr. Jabbar foresees that within two years seed producers will no longer need special training. They could still get technical advice through the union office. Federation members, seed committees and RDRS share the profit from seed sales on a 20:40:40 basis. All agree on this split, which will continue after the project ends. The RDRS share goes to a revolving fund and has already allowed RDRS to expand its model to other federations. Private sector One of the reasons for the dramatic increase in private sector participation in the rice seed network (Chapter 17, Figure 17.3) could be that they received free training in seed production from BRRI, which provided these small-scale seed entrepreneurs a business opportunity. Once the advantages of the training become more widely recognised, it may be feasible to charge for it, which could help sustain BRRI's capacity building role in the seed system. Government institutions Government institutions have a number of specific roles to play in seed supply (see Box 15.2). BRRI is a good example of how an institution can change from the bottom-up by getting involved in community-oriented research and capacity building. Researchers engaged in PETRRA say that they changed their views on many things. BRRI now receives a lot of credit for the PETRRA successes and such recognition is bound to bring in more new ideas. Science has its own value system, which is not always in tune with communities and the poor. Research institutions need to listen to real demands from the people, with mutual respect for the knowledge of other partners. Dr. Musherraf Husain, head of the Adaptive Research Department at BRRI, confesses that PETRRA has opened his eyes and revolutionised his views of his own role as a scientist:

198


'I have worked in research for 22 years, but now I realise that I have been working for the rich farmers all these years. I have seen with my own eyes that resource-poor farmers can be highly innovative and working with them has been very enriching for me. There is no way we can improve the livelihoods of poor farmers without involving them as active partners in the search for appropriate solutions.' Ensure capacity to meet increasing demand for breeder seed and for emergencies Maintain the national gene bank Strengthen the Seed Certification Agency and include seed health testing in certification of breeder and foundation seed

Box 15.2 Government Responsibilities in Seed Supply

Ensure capacity and involvement of DAE and BADC Support technical training of NGO staff Guarantee fair competition between public and private institutions Quick and smooth variety release Establish intellectual property rights and plant breeder rights Stimulate grain markets

This is where it all starts. The ground for development of new institutions has been prepared. Maintaining the incentives for BRRI is crucial to meet the increasing demand for breeder seed. Without this, the rice seed network will collapse. Since BRRI, as a public sector institution, is not allowed to recover the costs from selling breeder seed, it needs different incentives, e.g. sufficient funding and stimulating working conditions. Government institutions are difficult to move. There is a lot of suspicion in the system towards the new roles of institutions in seed supply. For instance, some wonder if the private sector and NGOs can guarantee seed quality. This reflects the difficulties many government employees have in keeping motivated, as they are caught in a bureaucratic, inflexible system with limited resources or rewards to accept new responsibilities. Involving public institutions like DAE, BADC and local governments is important for scaling up, given their long experience, established agricultural programmes, communication links and nation-wide coverage.

CONCLUSION The seed ordinance in 1997 provided a legal framework to develop new approaches. Through PETRRA, new kinds of actors successfully rose to the challenge of helping smallholders grow and sell quality rice seed to other poor farmers by making best use of existing institutional systems. Improved technical, business and 199


organisational skills, with effective public-private partnerships allowed enthusiastic, open-minded people to build sustainable, demand-driven seed supply systems.

REFERENCES AND RECOMMENDED READING Almekinders, C. and Louwaars, N. (1999) Farmers' Seed Production. New Approaches and Practices. Intermediate Technology Publications, London. Almekinders, C. and Thiele, G. (2004) What to do with the seed for small-scale farmers after all? Cultivos Tropicales (submitted). Ashby, J. A., Quiroz, C. A. and Rivers, Y. M. (1989) Farmer participation in technology development: work with crop varieties. In: Chambers, R., Pacey, A. and Thrupp, L. A. (eds) Farmer First. Farmer Innovation and Agricultural Research. Intermediate Technology Publications, London, pp. 115-122. Badstue, L., Bellon, M. R., Juárez, W., Manuel, I. and Solano, A. M. (2003) Social relations and seed transactions among small-scale maize farmers in the central valleys of Oaxaca, Mexico. Economics Working Paper 02-02. CIMMYT, Mexico. Bashar, M. K. and Sarkar, H. C. (1997) Rice genetic resources conservation and utilization in Bangladesh. In: Hossain, M.G., R. K. Arora and Mathur, P. N. (Eds) Plant Genetic Resources - Bangladesh Perspectives. Proceedings of a National Workshop on Plant Genetic Resources, held at Bangladesh Agricultural Research Council, Dhaka, 26-29 August 1997. NCPGR/BARC/IPGRI, pp. 66-70. Bashar, M. K., Khalequzzaman, M., Ahmed, M. S., Harun-ur-Rashid, E. S. M., Sarker, M. R. A., Kamruzzaman, M., Sen, S. and Ashraf-ul-Alam, T. (2004) Completion Report on Rice Diversity and Production in the Southwest of Bangladesh: Using Diversity and Local Knowledge to Create Sustainable Livelihoods in the Coastal Area of Bangladesh. IRRI, Dhaka, Bangladesh. Bellon, M. R. and Reeves, J. (eds) (2002) Quantitative Analysis of Data from Participatory Methods in Plant Breeding. CIMMYT, Mexico. Cromwell, E. and Wiggins, S. (1993) Sowing Beyond the State: NGOs and Seed Supply in Developing Countries. ODI, London. David, S. (2002) Farmer seed enterprises: a sustainable approach to seed delivery? Review Paper. CIAT. Kampala, Uganda. De Vries, J. and Toerrigs, G. (2001) Securing the Harvest. Biotechnology, Breeding and Seed Systems for African Crops. CABI Publishing, Wallingford, UK. Elings, A., Almekinders, C. and Stam, P. (eds) (2001) Why focus the thinking on participatory plant breeding? Euphytica 122 (3), 423 - 600. FAO (1993) Quality Declared Seed. Technical guidelines for standards and procedures. FAO

200


Plant Production and Protection Paper 117. FAO, Rome, Italy. Friis-Hansen, E. (2004) Concepts and experiences with demand-driven advisory services. DIIS Working Paper no. 2004/7. Danish Institute for International Studies, Denmark. Huda, N. and Smolders, H. W. J. (2002) Seed regulatory frameworks in a small farmer environment: the case of Bangladesh. In: Louwaars, N. P. (ed.) Seed Policy, Legislation and Law - Widening a Narrow Focus. The Haworth Press Inc., London, pp. 177-193. Hossain, M. (2002) Rice research and poverty alleviation in Bangladesh. CPD - IRRI Policy Brief 2. IRRI, Dhaka, Bangladesh. Hossain, M., Janaiah, A., Husain, A. M. M., Naher, F. (2002) Rice seed delivery system and seed policy in Bangladesh. CPD - IRRI Policy Brief no. 1. IRRI, Dhaka, Bangladesh. Iriarte, V., Terrazas, F., Aguirre, G. and Thiele, G. (2000) Local seed systems and PROINPA's genebank: Working to improve seed quality of traditional Andean potatoes in Bolivia and Peru. In: Esbern Friis-Hansen and Bhuwon Sthapit (eds) Participatory Approaches to the Conservation and Use of Plant Genetic Resources. IPGRI, Rome, pp. 154-161. Kugbei, S., Turner, M. and Witthaut, P. (2000) Finance and Management of Small-Scale Seed Enterprises. ICARDA, Aleppo, Syria. Louwaars, N. P. (1994) Integrated Seed Supply: institutional linkages in relation to system efficiency, biodiversity and gender. In: David, S. (ed.) Alternative Approaches to Bean Seed Production and Distribution in Eastern and Southern Africa. Proceedings of a Working Group Meeting, Kampala, Uganda, 10-13 October 1994. Network on Bean Research in Africa. Workshop Series No. 32, CIAT, Kampala, Uganda. Louwaars, N. P. (ed.) (2002) Seed Policy, Legislation and Law - Widening a Narrow Focus. The Haworth Press Inc., London. Nathaniels, N. Q. R. and Mwijage, A. (2000) Seed fairs and the case of Marambo village, Nachingwea district, Tanzania: implications of local informal seed supply and variety development for research and extension. AgREN Network Paper 101. Ortiz-Ferrara, G., Bhatta, M. R., Pokharel, T., Mudwari, A., Thapa, D. B., Joshi, A. K., Chand, R., Muhammad, D., Duveiller, E. and Rajaram, S. (2001) Farmer participatory variety selection in South Asia. Research Highlights of the CIMMYT Wheat Program Research 1999-2000. CIMMYT, Mexico, D.F., pp. 33-37. Salam, M. A., Islam, M. R., Bhuiyan, M., Khan, A. A. and Faruquei, M. (2002) Participatory variety selection: an initiation in Bangladesh. In: Witcombe, J. R., Parr, L. B. and Atlin, G. N. (eds) Breeding Rainfed Rice for Drought-Prone Environments: Integrating Conventional and Participatory Plant Breeding in South and Southeast Asia. Proceedings of a DFID Plant Sciences Research Programme/IRRI Conference, 12-15 March 2002, IRRI, Los Baños, Philippines, pp. 64-65.

201


Sperling, L. and Ntabomvura, B. (1994) Integrating farmer experts into on-station research. In: Feldstein, H. S. and Jiggins, J. (eds) Tools for the Field: Methodologies Handbook for Gender Analysis in Agriculture. Kumarian Press, Connecticut, pp. 128-135. Sperling, L., Scheideggeer, U. and Buruchara, R. (1996) Designing seed systems with small farmers: principles derived from bean research in the Great Lakes region of Africa. AgREN Network Paper 60. Talukder, M. Z., Ahmed, M. S. and Bashar, M. K. (2004) Present status and future plan for rice seed production and processing in Bangladesh. Paper presented at the Research and Extension Workshop, BRRI, Gazipur, Bangladesh. Tendler, J. (1997) Good Government in the Tropics. Johns Hopkins University Press, Baltimore. Tripp, R. (2001) Seed Provision and Agricultural Development: The Institutions of Rural Change. ODI, London. Tripp, R. and Louwaars, N. (1997) Seed regulations: choices on the road to reform. Food Policy 22(5), 433-446. Tripp, R. and Pal, S. (2001) The private delivery of public crop varieties: rice in Andhra Pradesh. World Development 29(1), 103-117. Van Mele, P. (2002) Seed Health Improvement sub-project: Uptake Pathways. CABI Bioscience, Egham, UK. von Lossau, A., Kasten, W. and Almekinders, C. (2000) Support for the Informal Seed Sector in Development Cooperation - Conceptual Issues. GTZ, Germany and CGN, the Netherlands, Eschborn. Wright, M. and Turner, M. (1999) Seed management systems and effects on diversity. In: Wood, D. and Lenne, J. M. (eds) Agrobiodiversity: Characterization, Utilization and Management. CABI Publishing, Wallingford, pp. 331-354.

202

16 More Is More

Validating rice varieties with NGOs and poor farmers

Musherraf Husain

SUMMARY New rice varieties could raise yields and improve poor farmers' quality of life, yet most have little or no access to them. Traditionally, the Department of Agricultural Extension (DAE) presents just two or three new rice varieties, and only to larger farmers, so adoption rates are low. The Bangladesh Rice Research Institute (BRRI) assumes that those varieties are good for a specific region, based on its own understanding of the varieties. But in this project, BRRI improved the validation and adoption of new varieties by presenting more of them, and by collaborating with NGOs to work with poor farmers with a farm size of less than 0.6 ha or less than 9 months of rice-provisioning ability. During a group discussion, each farmer selected a variety to test in his or her own field and received training. The fields were near each other and marked with signboards, so visiting farmers could easily make comparisons. Local staff from the DAE facilitated the training and field days with the NGO. This partnership allowed female farmers to also judge the new varieties. Adoption of the varieties was higher than before, with farmers adopting 3-4 out of the 8-10 new varieties. Once farmers identified an interesting new variety, they disseminated it themselves. Poor people can be early adopters of new rice varieties. Collaboration between DAE and NGOs opens new doors for validating and disseminating both modern and promising local varieties. 203


ACTORS AND NETWORKS BRRI's Adaptive Research Division is charged with finding new rice technologies and validating them in farmers' fields. In this project, staff helped design the research trials, decided the rice varieties to be validated and provided technical back up. They also provided training and new rice varieties to the partner agencies' staff and farmers. DAE, the government extension agency, has a national network of extension agents in the field. Overall, staff have a broad knowledge of agriculture and of local customs and institutions. They have less experience in action research, as they usually focus on disseminating technologies, especially to larger farmers. While testing the new method, DAE staff worked closely with the farmers doing field trials, and helped them solve day-to-day problems. The NGO Rangpur Dinajpur Rural Service (RDRS) works to improve the livelihood of the landless, marginal and poor farmers. It has a strong agricultural, women's empowerment and credit programme, with 30 local offices and 256 federations. A federation works at the union level, and comprises 30-40 groups of poor households. About 268,000 households are affiliated to these federations. Most of the farmers who tested rice varieties with BRRI under this new method belonged to a federation. How the RDRS federations became involved in rice seed is described in Chapter 20. RDRS selected the sites, formed farmer groups and helped with the research activities.

EVOLUTION OF THE METHOD Previously, BRRI's Adaptive Research Division demonstrated rice varieties to farmers through the DAE, without involving NGOs. Farmers had few options for choosing varieties. Just a few rice varieties were shown through large demonstrations; farmers were passive recipients. The government agencies thought of large farmers as innovators, and provided them with seed and other inputs, the whole system being highly subsidised. Poor farmers were considered laggard and no consideration was given to gender. Field days were organised to generate large-scale awareness and interest among farmers. Partnerships with NGOs were not really considered. This style continued until 2001, until, stimulated by the PETRRA project, BRRI's Adaptive Research Division gradually came to understand its own limitations and made major philosophical adjustments. At first, the project provided inputs, but soon input costs were recovered from participating farmers through RDRS's credit scheme and revolving funds. The project triggered fundamental changes in one of the major national agricultural research institutes, BRRI. Throughout the country, BRRI's Adaptive Research Division can now conduct adaptive research for a range of rice technologies cost effectively by training NGOs and DAE field workers. 204

More Is More

THE TECHNOLOGY VALIDATION METHOD Offer many varieties BRRI learnt to offer poor farmers more varieties at once. BRRI used to show farmers two or three modern varieties that were already being grown to some extent in the region. In this project, we also offered some unreleased 'varieties', along with some released varieties that have been adopted in other areas and an exotic one (IR64). BRRI has spent 30 years developing rice varieties for specific agroecosystems and has an institutional memory of when and where specific varieties perform best. The Adaptive Research Division decided to show farmers only varieties released within the last ten years. Through another project on biodiversity by the Genetic Resources Division of BRRI, some NGOs started collecting local varieties for conserving in the national gene bank (see Chapter 17). Select sites We tested our method from 2001 to 2004 in several natural and social environments in Northwest Bangladesh, one of the poorest areas of the country, in ten upazilas of Rangpur, Dinajpur, Kurigram, Lalmonirhat, Nilfamari and Thakurgaon districts. The upazilas were chosen so that the local RDRS office could supervise the field activities. Identify local NGOs We chose partner NGOs with an interest in agriculture, that worked with farmers, and were receptive to collaboration with government agencies. RDRS was selected as a partner NGO, because it has a pro-poor philosophy, 30 local offices in the greater Rangpur region and can draw on extensive rural networks through its federations. RDRS in turn identified the NGOs Janani and BRAC, and the private sector East-West Seed Company as local partners. BRRI brought all parties together to develop a mutual understanding of the responsibilities of NGO staff and the DAE extension officers.

Offer many varieties to a group of people and after trying they will tell you which one they like most.

Farmers evaluate varieties The NGOs identified farmers to participate in the research, receive training and become demo farmers. Ten people were selected per group in each season and in each upazila. BRRI used to present just 2-3 varieties to farmers. But in this project we showed poor

205


farmers about 8-10 rice varieties for validation. Farmers planted each new variety beside one of their own, and staff helped farmers compare them at different intervals. In the project area, farmers mainly preferred short duration, bold grained rice varieties. The research brought DAE, NGO staff, and farmers closer together. Develop capacity within partner organisations and farmers In one-day sessions, BRRI trained staff from the partner organisations in such fundamentals as rice varieties, agronomy, insect pests and diseases, fertiliser, seed and post-harvest. For each season, the NGO and DAE selected ten farmers, men and women, who were not involved in training previously, and five neighbours of each of the participants. This allowed an exchange of experiences with the community. Staff from the local partners were invited to attend the training and field days, so they could help spread the new rice varieties. Mrs. Bulbuli Rani is a poor, but innovative woman. She established a small rural sewing enterprise after having received a small loan from RDRS, one of the NGOs working in Northwest Bangladesh. Her strong leadership and enthusiasm for agriculture has made her a role model in the region for involving women in action research.

Involve women in research About 40% of the farmers who helped with the research were women, and they contributed more than BRRI had expected. The poor women were often the first to adopt a new variety. Mrs. Bulbuli Rani, the vice chairperson of the Mohendra Nagar federation in Lalmonirhat district has been an advocate and innovator; she asked to be involved in the research. In 2004, she was working with BRRI scientists and an MSc student from the Bangladesh Agricultural University to develop hybrid rice seed (see Chapter 20). Overall, the men did not oppose women participating. Between 2002 and 2004, about 330 farmers were trained, of whom 30% were women. We held about 10 field days and visits. The field days were large events and a bit formal, but visits to the research plot to exchange views were usually done in small groups of researchers and farmers. Women participated in every step. Feedback mechanism Farmers' response to the varieties was communicated to the researchers and extension staff. The partner organisations regularly monitored the field and resolved problems quickly. BRRI staff held group discussion with farmers to gather feedback on the varieties. Once farmers found a variety superior to their existing one, they adopted it quickly. The demo farmers all retained seed of new varieties to plant the following season, while

206

More Is More

neighbours and farmers who were involved in the field days demanded seed from either DAE or RDRS. This helped the Genetic Resource Division of BRRI to set targets for breeder seed production based on actual demand. NGOs involved in the rice seed network can sell farmers either foundation seed or truthfully labelled seed of the varieties they want, in a timely manner (see Chapter 17). Figure 16.1 depicts the method, developed under the value-based PETRRA project, which pays attention to poverty, gender, participation and partnerships.

Make prior selection of potentially suitable technologies for specific ecosystem Select & characterize sites

Identify partners GO/NGOs

Identify 10 resource-poor demo farmers and 5 neighbours ofeach of them for training

Conduct group meeting to introduce and allocate 8-10 varieties to demo farmers

Action research for locationspecific technology validation

Training & field visits Field days

Feedback from all resource-poor farmers through interviews and measurement of crop yield

Technology for resource-poor farmers identified & recommended

Fig. 16.1 Validating varieties with government-NGOs-farmers 207


KEYS FOR SUCCESS Partner NGOs must be enthusiastic about working with male and female poor farmers. Women play a key role in growing rice, and involving them in training and knowledge sharing helps improve productivity, as highlighted in Part II on gender. By presenting more varieties for validation, poor farmers can become early adopters and play an important role as extension agents to reach the vast farm community.

RISKS, DIFFICULTIES AND ASSUMPTIONS At first, DAE had little experience or interest in working with poor farmers, especially women. In future, it will be important to keep DAE field staff committed. A new experience can be unsettling, but NGO partners help reduce this discomfort. Partnerships with NGOs give DAE hands-on experience working with a new target group. Through many contacts with researchers and extension agents, poor farmers of the RDRS federations gained more confidence to express their needs. In the future, we expect them to express their demands for varieties to RDRS and DAE, assuming that the farmers continue to enjoy good relationships with government and NGO extension agents. Strengthening links with the rice seed network will be crucial to seeing that these demands are satisfied. In Chapter 20, Samsuzzaman and Van Mele describe the crucial role federations can play in this.

SCALING UP The method may be applied and fine-tuned in other environments by other government agencies and NGOs to validate and disseminate pro-poor technologies other than seed. Once a technology has been validated, NGOs may disseminate it through their own network. RDRS, for instance, is already disseminating varieties through their federations in a sustainable manner. The focal area forums offer decentralised multiple-actor platforms for validating technologies and channelling the voice of the poor about issues broader than seed (see Box 21.1).

CONCLUSION More is more: more NGOs, more farmers, and more varieties. To ensure a pro-poor and gender-sensitive approach, government agencies must work with NGOs and poor farmers, men and women. Giving farmers many varieties to test at once allows poor men and women to be innovators and become advocates in farmer-to-farmer extension. Farmers will grow and disseminate the varieties they like. Incorporating the method in multiple actor platforms further stimulates institutional change.

208

17 Building a Rice Seed Network

Md. Khairul Bashar Ahmed Salahuddin Paul Van Mele

Village women spread the word

SUMMARY Since the signing of the national seed act (Amendment) in 1997 and national seed rules in 1998, NGOs and private agencies have been able to purchase breeder seed from the Bangladesh Rice Research Institute (BRRI); they plant breeder seed to raise quality seed for farmers. Until recently, only 5% of the rice seed was supplied to farmers by a few government and private agencies. A more efficient national seed system would have to supply farmers with timely, adequate quality, modern variety seeds at affordable prices. With this in mind, BRRI developed a public-private network to ensure a continuous supply of breeder seed under the PETRRA project. We trained and technically supported a wide range of partners to produce foundation and quality seed. Over four years, the supply of quality seed to farmers has increased from 5% to an estimated 15%, while the number of organisations involved increased from three in 1998 to 54 in 2003. Apart from a better-coordinated demand assessment, decentralised production and dissemination of seed, the network enables a quick response to sudden seed shortages due to natural disasters. As its coordinator is also head of the national rice gene bank, the network also helps partner NGOs conserve rice biodiversity by collecting and evaluating local varieties. 209


ACTORS AND NETWORK BRRI is responsible for producing breeder seed of the varieties they develop and recommend. BRRI produces and supplies breeder seed to the government's Bangladesh Agricultural Development Corporation (BADC) which multiplies the seed and distributes it nationwide. In the old model, only BADC produced foundation seed on its farms (Figure 17.1). Certified or truthfully labelled seed (TLS) was produced from foundation seed through their contract growers in 15 zones of the country. Under the new model, BADC is no longer the sole producer of foundation seed. They now share their physical plant, such as processing centres and stores, with other players in the seed business, and so gradually become a service provider. The governmental Seed Certification Agency is not directly involved in the network, but plays a vital role by providing quality control services from breeder seed to foundation seed to certified seed. By 2003, quite a few private seed producers had become involved in the seed network, mostly operating at about the same scale as the local NGOs, but selling their seed on the open market. Throughout the country, an increasing number of NGOs have embarked on agriculture and now distribute quality seed to the poor. Although it is difficult to classify the wide range of seed producers, an overview is given in Table 17.1.

EVOLUTION OF RICE SEED NETWORK Before PETRRA. Until the late 1990s, the private seed sector was not interested in

a

b BRRI

BADC

Breeder Seed

BADC

Contract growers

BRRI

Private Sector

NGOs

Contract growers Farmer seed producers

Foundation Seed

Figure 17.1 Old (a) and new (b) seed systems in Bangladesh

210

Building a Rice Seed Network

Table 17.1 Comparison between seed producers under the rice seed net

Mandate(s)

LOCAL NGO

NATIONAL NGO

PRIVATE SECTOR (PS) BADC

Improve livelihood by forming groups and providing microcredit

Enterprise development They have no mandate for sustainability to benefit poor farmers Provide service through business

To provide service to the nation by producing and distributing quality seed

BRRI To develop modern rice varieties and sustainable rice production technologies adapted to different seasons and ecosystems To provide breeder seed to multiple actors upon request To maintain rice gene bank

Focus on resource-poor

Yes

Yes

No

No

Yes under PETRRA

Motivation to participate in network

Highly self-motivated and feel privileged to be part of valuebased seed network

Motivated to expand their work

Highly motivated and feel privileged to be partner of network

Motivated

Highly motivated

To sell their products and service to the network partners

Leading the network is part of BRRI's mandate

Pre-1998 experience of seed production

No

No

No

Yes

Yes

Strong points

A lot of scope to work Have fixed, nationwide with poor farmers clientele groups locally Push seed sales to their Work with groups clients through credit support Can assess demand

Similar to local NGO except Syngenta, but sell seed through dealers in open market

Highly skilled manpower All kinds of technical support (demand-led Enough physical variety development, facilities and logistics throughout the country training, monitoring etc.)

and distribute seed efficiently

Have skilled manpower and physical facilities.

Competitive advantage

Enjoys the trust of the community

Have well-established nationwide customer base

Small seed producers/sellers already know their

Due to government subsidies, they can sell seed at low price customer base Syngenta through a wellestablished marketing has good marketing channel network and attractive packaging

Not applicable

Weaknesses

Little experience with seed, less trained manpower and facilities

Work with people irrespective of poverty levels

Price higher than BADC

Limited physical facilities

Low market coverage

To become established as seed entrepreneur

Higher price of seed than BADC Inadequate demand assessment resulting in unsold seed being sold the following year

Have own processing centres and marketing channel

For small entrepreneurs, human resources and physical facilities are insufficient

Sale of produce is not ensured

Maintain seed network throughout the country

Insufficient skilled No control over variety manpower in breeder and quantity selection seed production as it is decided by Seed activities Promotion Committee of Ministry of Agriculture

211


growing or selling rice seed. NGOs lacked the equipment and skills, resulting in questionable seed quality. Of the large volume of seed needed by farmers, BADC was able to provide only about 5% of it, suggesting that there was a nationwide scarcity of good, formal seed. After the national seed rules took effect in 1998, NGOs and private sector agencies started approaching BRRI for breeder seed. This was on a first-come, first-served basis, and without any screening of partners as to their capacity to produce foundation and quality seed. Phase 1. Increasing breeder seed production. In 1999, the Genetic Resources and Seed Division of BRRI launched the first phase of this PETRRA sub-project to increase the supply of breeder seed and to help all categories of seed producers to improve their knowledge.

Farmers sell their rice at the weekly haat or market. Increasingly farmer seed producers develop a business mentality and compete with commercial seed dealers. Until 1997 quality seed was only sold through the government system, but policy changes have opened up the market to NGOs, private sector and farmers.

Phase 2. A network emerges. In the second phase of the project, it became clear that BRRI had to screen for the most suitable partners. A memorandum of understanding was signed between BRRI and several NGOs and private seed companies with enough technical capacity and land to grow foundation seed. This gave them a guaranteed supply of breeder seed, and a legal framework for producing foundation or quality seed under BRRI's supervision. In 2004, several other organisations were about to sign a memorandum of understanding. BADC is no longer the sole producer of foundation seed. For an overview of the major differences between the old and new seed system see Figure 17.1 and Table 17.2. PETRRA provided an environment for its different sub-projects to regularly exchange experiences at the uptake forum and laid the groundwork for a rice seed network. All organisations that were involved in the nine seed uptake sub-projects increasingly requested breeder seed from BRRI and automatically became members of the seed network. Phase 3. Local leadership leads to regional networks. This decentralisation into focal area forums (see Box 21.1) should lead to fewer direct requests for breeder seed from BRRI headquarters, as the networks coordinate local demand and as BRRI regional stations produce more seed.

HOW DOES THE RICE SEED NETWORK OPERATE?

Identify partners and establish network BRRI established a rice seed network at the national and regional levels to get quality seed easier and faster to poor farmers. It allows for local demand to determine which varieties will be produced and how much. 212


At the national level, we initially chose partners based on their existing capacity and nationwide coverage, either directly or through strategic networks with local organisations. So far, BRRI has signed a memorandum of understanding with three Table 17.2 Comparison of old and new seed distribution systems OLD SYSTEM

NEW SYSTEM

Planning

Narrower

Wider

Quantity

Based on sales records of the previous season.

Seed producers within the network take stock of needs and place their demands for amounts of breeder seed directly to BRRI.

Variety

No knowledge about farmers' preferences. Appropriate varieties demanded by farmers were not available in all locations.

Based on preference surveys, quality seed of locally requested varieties is produced and distributed easier and quicker.

Quality

Quality of seed was questionable.

Capacity of NGO and private sector is built and monitored by BRRI scientists to ensure quality of foundation seed production.

Availability

Not ensured. Only 5% of the national seed demand could be supplied.

Seed supply has increased to meet 15% of the national demand.

Dissemination

Slow. Network of dealers was weakly developed within districts.

Quicker. Farmers have timely access to quality seed.

Demand

NGOs (BRAC, GKF and Podakhep), and one private company (Syngenta). The NGOs have their own foundation seed farms, and to some extent their own processing and storage facilities. Other NGOs such as AAS, Proshika, RDRS and Shushilan along with some private companies like Supreme Seed and Alpha Agro Seeds are still in the process of signing a memorandum of understanding, but they are already allowed to buy breeder seed (Figure 17.2). The number of partners in the network grew from three in 1998 to 54 in 2003, of which about 35 are small-scale agribusiness dealers. They easily build on their existing customer base and respond well to demand. Unlike the private companies, local NGOs have not started mass-producing rice seed (Figure 17.3), because they perceive it as risky and lack the expertise. Nevertheless, those who have started on a small scale are now expanding; because they now know they can get a fair price and easily sell all their seed. 213


Small seed entrepreneurs and local NGOs use breeder seed directly to produce truthfully labelled seed. Most of them use BADC processing and storage facilities, so that both genetic and physical purity is guaranteed. It is a waste to use breeder seed to produce truthfully labelled seed, but it is how the network is experimenting, learning and developing. Over time, these seed producers will need to rely on foundation seed suppliers in their area rather than on breeder seed from BRRI. To respond better to location-specific demands, focal area forums have been created in Northwest and Northeast Bangladesh, where regional BRRI stations already

BADC

Syngenta

BRRI RS

BRRI HQ & Regional Stations

Contract Farmers AAS Local NGOs

RDRS Proshika GKF

BS

DAE

Private Sector Federation

Padokhep

Own Group

AAS

Farmseed

FS

QS

R ESOUR CE-POOR AND OTHER FARMERS

BRAC

BADC

Paddy

Figure 17.2 Flows of breeder (BS), foundation (FS) and quality seed (QS) through the rice seed network, 2003 214


produced and distributed seed (see Box 21.1). Although first set up to talk about regional issues on rice and to explore ways to deliver information consistently to farmers, seed came up as issue no. 1 in both areas. Partners were selected who have a license from the Ministry of Agriculture and who are specifically working with resource-poor farmers in that focal area. In the Northwest, RDRS has taken the lead, while in the Northeast, albeit a bit slower, AAS is emerging as leader. The national and regional networks have been strengthened through formal agreements, training, regular monitoring of foundation seed production plots by BRRI and interactions of all partners in planning workshops.

4

GO NNGO

ton breeder seed purchased

3.5

LNGO

3

PS

2.5 2 1.5 1 0.5 0 1997

1998

1999

2000

2001

2002

2003

year Figure 17.3 Evolution of breeder seed purchase by government, national and local NGOs and private sector

Assess quantity of breeder seed required to meet demand of specific varieties The demand for breeder seed gauged at the national and focal area level. At the national level, the demand analysis, planning of production and distribution, and policies are discussed at the seed promotion committee under the Ministry of Agriculture. To better assess local needs, the Adaptive Research Division of BRRI

215


has adopted a pro-poor partnership and bottom-up approach, as described in the previous chapter. Planning workshops for the focal area forums are organised at the BRRI regional stations, or at any of the partner organisations. Supply breeder seed to partners BRRI has the sole responsibility to produce and distribute breeder seed, which is supplied to network partners in 10-kg bags. Personnel of the Seed Certification Agency monitor the distribution of breeder seed and seal the bags with a tag. Apart from BRRI headquarters, now two of the nine regional BRRI stations produce breeder seed of certain varieties as well. In 2003, demand for and production of breeder seed was double compared to the previous years, despite BRRI having raised the price per kg by 400%. Even after this price hike, some small-scale private seed producers still buy breeder seed to produce truthfully labelled seed. This waste of resources is hard to avoid at this stage of development, as mentioned earlier and illustrated in Box 17.1. Ideally, a national strategy should be developed to decide who should produce foundation seed and where. BADC, BRRI regional stations, experienced NGOs and large companies could cover the whole country.

Box 17.1 A Telephone Request for Breeder Seed

After introducing himself, Mr. Basith Salim, at the other end of the line asks: "Would it be possible to get five kg of BR19 breeder seed? I need to produce about 50 tons of truthfully labelled seed to satisfy my customers." He is the owner of the private agro business AB Krishi Prakalpa and started producing seed about three years ago. "Have you tried the local BADC office to ask for foundation seed?" replies Dr. Bashar over his mobile, the rice seed net coordinator and head of the Genetic Resources and Seed Division, at his office in BRRI, Gazipur. Salim continues, "Sure I have, but you know very well that this is not a popular variety in Bangladesh apart from my district (Sunamgonj), so it is not available from BADC. What can I do?" Bashar knows that the seed network is still in its infancy and that more efforts are needed to ensure appropriate foundation seed is available everywhere in Bangladesh. "I will get my staff to prepare a bag. This time you are lucky, you don't have to come yourself because my staff is travelling to the region next week. They will bring it," assures Bashar. "Thank you so much, Bashar, you are really very helpful. Oh and by the way, I got to know in my shop that some farmers from Moulvibazar and Sylhet districts like Akhnisail, a local variety. Shall I go and collect some for the rice gene bank?"

216


Train the trainers For years, the senior author (Bashar) was regularly invited as speaker in training programmes organised by BADC. Since PETRRA, the techniques used for maintaining purity of breeder seed are taught equally to the network partners through training, monitoring and visiting their foundation seed production farms. On-the-job training courses on seed production and preservation are organised by the Genetic Resources and Seed Division at BRRI. The training also helped develop an esprit de corps. Trained personnel then teach quality control to groups ofsmallholder seed producers. Various mechanisms to work with poor seed producers are described in the following three chapters. Monitor seed quality regularly Personnel from the Seed Certification Agency monitor breeder seed production at BRRI headquarters in Gazipur and the regional stations (Table 17.3). At the same time, BRRI scientists help to solve technical problems in the field. The first monitoring takes place in the field, the second during drying and processing and the last one during storage, when samples are taken for laboratory testing. During supervision, the Seed Certification Agency recommends several techniques to improve breeder seed production, such as leaving a one line gap after each six lines during transplanting, drying one variety at a time on the drying floor and threshing only one variety at a time. Table 17.3 Actors involved in quality control under rice seed network, 2004 SEED CLASS

PRODUCER

ACTORS FOR QUALITY CONTROL

Breeder seed

BRRI

Breeders and staff of Seed Certification Agency

Foundation seed

BADC, private sector and NGOs

Producer, BRRI scientists and staff of Seed Certification Agency

Certified and TLS seed

BADC, private sector and NGOs

Producer (optional Seed Certification Agency and BRRI)

Per request, BRRI scientists monitor the foundation seed production plots of the partner organisations so that they can maintain the required field and seed standards set by the Seed Certification Agency. Although certifiers are limited, they also visit the foundation and certified seed plots of partner organisations to give them tags after satisfactory results. But what will happen with quality control of foundation seed after the project ends? Considering that the incentives for BRRI staff to visit 217


every foundation seed farm upon request of a partner will diminish, policy support is needed to strengthen the Seed Certification Agency. Conserve rice biodiversity PETRRA has stimulated synergies between her sub-projects. The head of the national rice gene bank (Bashar) and coordinator of the national seed network is also the principal investigator of another PETRRA subproject on rice biodiversity. As such, the rice seed network started to become a hub for local varieties or landraces to enter the formal system. BRRI dhan 34, for instance, was the first local aromatic variety that became registered, and now the network has opened up opportunities for a broader range of local varieties to be collected, screened, registered and become legally protected by intellectual property rights.

Honest Neighbour. When driving to Moulvibazar district on April 27, 2004 to interview a group of farmers trained as seed producers by AAS, Paul Van Mele wonders how over the past years they have ensured quality control. Obviously the national Seed Certification Agency will not have visited their 2,500 seed producers. Before Paul can raise the question, Nikesh Gop, a 26year old seed producer in Uttar Baruara village illustrates what is at stake: "In Boro 2003, my BR28 seed showed poor germination when I tested it, so I decided not to sell any seed that season. It is better to sell nothing than to lose my reputation in the village."

Some NGOs are currently playing an innovative role in securing rice biodiversity. In Khulna in Southwest Bangladesh, about 16 NGOs have been engaged in germplasm collection of local varieties. Through the rice seed network, we identified a few suitable NGOs who help to evaluate these local varieties with farmers in a wide range of different agroecosystems and to multiply the most important ones.

KEYS FOR SUCCESS Changed seed policy helps seed producers to get access to breeder seed. A well-equipped breeder seed unit has been established at BRRI. Ensured access to breeder seed motivates partners. Increasing demand of quality seed ensures that all seed is sold. Partners share information and resources. Partners get technical knowledge through training and monitoring. Quality control is emphasised. The rice seed network developed in a flexible, learning by doing environment provided by PETRRA.

DIFFICULTIES, RISKS AND ASSUMPTIONS A favourable natural environment is essential for growing good seed. Natural disasters can disrupt the seed flow. Proper field selection is essential. Although it seems risky that the quality of seeds produced by smaller partners scattered throughout the country cannot be monitored by an official

218


agency, the ultimate quality control happens by pressure of their customers. Of course, this bottom-up quality control only works when farmers have sufficient choice in seed suppliers. To meet the growing demand for breeder seed in the future it will be necessary to invest in physical and human resources. Increasing the price of breeder seed from Tk 25 (US$ 0.45) per kg to Tk 100 (US$ 1.75) reduced the misuse of seed and increased the revenue for BRRI. However, the breeder seed unit at BRRI cannot use this money for its self-sustainability and to accommodate the growing demand. Policy changes are required to make this happen. The nine regional stations of BRRI have enough land to grow more breeder seed. Also foundation seed could be produced here, but so far BRRI is not officially recognised to do this. Revising the law with regard to BRRI's and BADC's mandate would be recommendable. Small-scale seed producers generally lack access to bulk processing and storage facilities, which prevents quality seed from reaching poor farmers at a reasonable price. BADC, selling seed at a subsidised rate, distorts the market. To avoid this, seed should become registered as an industry. Taking stock of existing facilities and developing a framework of who produces what seed nationwide, is something the rice seed network identified as essential, but little time was actually given to bringing this into practice. Delegating responsibilities from BRRI headquarters to other players could reduce the network's management load, freeing energy to spend on further expanding the network and on producing breeder seed.

SCALING UP By establishing the network, the supply of quality rice seed to farmers has increased from 5% to about 15%. The network allows BRRI to asses the demand for its breeder seed more accurately, and so to produce the right amounts, but it also enabled partners to quickly help one another at times of acute seed shortages due to natural disasters. As there are very few seed outlets in Southwest Bangladesh, since 2003 seed production has been promoted in six villages, of which three are under the supervision of the NGO HEED Bangladesh and three are under the NGO Shushilan. In each village, nine farmers were trained to grow foundation and quality seed. They received breeder seed of BR23 and BRRI dhan 40, the most popular varieties in the region (see Box 17.2). The smallholder seed producers are now keenly aware of the demand for seed by clients in the area. In 2004, another seed producer group started producing BRRI dhan 28 and 29.

219


Box 17.2 Overcoming Local Seed Shortages

In Southwest Bangladesh, hardly any agricultural support services exist. Lack of access to quality seed is a major issue. BRRI triggered an innovation by providing breeder seed and training farmers and local NGOs in that region to grow foundation and quality seed. PETRRA, through the rice seed network, has created an environment for people to solve their own problems of quality seed supply.

Some network partners such as ABCD and BRAC have since developed a memorandum of understanding with the Bangladesh Agricultural Research Institute (BARI) to get breeder seed of other crops. Smallholder seed producers have already applied their skills to other seed crops, including onions, wheat, mustard and potatoes (see Chapters 18 and 20).

CONCLUSIONS The network now has many partners and is capable of producing quality seed, with profitable businesses emerging. The rice seed network has improved access of quality seed at the grassroots level, resulting in a rapidly growing demand for quality seed. Policy reform is needed to optimise breeder seed production and performance of the rice seed network.

220

18 Farmseed

Putting farmers at the heart of the seed system

Paul Van Mele

SUMMARY Poor farmers need better and more affordable access to quality seed in order to improve their livelihoods. Over three years, the national NGO Agricultural Advisory Society (AAS) developed an innovative decentralised seed production and distribution system, called Farmseed. Because they operate with a limited budget and staff, AAS formed a strategic network with 64 NGOs and community-based organisations (CBOs), of which half remained viable partners. Teaching their staff to produce rice and quality seed was the main incentive for collaboration. From a large number of rice varieties, communities selected the three most suitable ones during on-farm demonstrations. AAS started selling foundation seed of these varieties in three-kilogram bags to resource-poor farmer groups on a seasonal basis, and trained them in the field. Women's groups received training on post-harvest, while other villagers learnt about quality seed through regular field days and by interacting with the community seed producers. Farmer-produced quality seed easily fetches a 25% higher price than common farm-saved seed. At the same time, community members' exchanges of seed or seedlings for good consumption rice, function more as social glue than as a profit-making business. Once local capacity is built, the system shows promising signs of sustainability by recovering costs through the production and sales of foundation seed. By 2004, modern rice varieties have reached about 20,000 farmers through 2,000 seed producers. Apart from the new Farmseed model, which AAS is now testing for onion and potato seed, this case study offers us some honest insights in the strengths and pitfalls of working with a vast number of local partner organisations. 221


ACTORS AND NETWORKS "Rice seed is too bulky for the private sector to be truly bothered about, and the high transportation cost makes a decentralised seed production system a necessity," says Harun-Ar-Rashid, executive director of AAS, which has focused its energy on agriculture and smallholder farmers since 1990. Although Bangladesh is blanketed by the farms of smallholders, organisations that concentrate on them are scarcer than one would expect. That is why it is important to learn how a relatively small NGO like AAS tried to make a difference. Through many projects AAS has been the driving force in developing a good working relationship with over 150 NGOs and community-based organisations across the country. Capitalising on this strategic network was one of their main reasons for success. But that's only part of the picture. Otherwise, how could one explain that the Farmseed project with a small budget and only about 10-15 AAS staff trained some 7,000 poor farmers and motivated about 13,000 more over just three years? AAS is unique in having a balanced number of men and women on its staff; all are motivated young graduates from local agricultural universities. They are committed to making this new system of farmer seed production and distribution work, and not just during the project's life cycle. Their eagerness to experiment with pro-poor mechanisms turned AAS into one of the most pro-active partners, implementing nine different sub-projects under the PETRRA project. Harun-Ar-Rashid interacts with farmers as much as possible, because they are the ones testing the technologies.

Three years ago, AAS engaged 64 partner organisations of which about half still remain in the Farmseed project. As Table 18.1 shows, the partners who stayed with the project were community-based organisations and local NGOs, with ties to the farm communities, but little previous experience in agriculture. Apart from the elaborate strategic network with mainly local organisations and based on reciprocal respect and mutual benefits, Harun signed official agreements with his former governmental employers. The Bangladesh Agricultural Development Corporation (BADC) was to supply foundation seed, and the Bangladesh Rice Research Institute (BRRI) to provide technical training. Their input strengthened the system. The stakes are high according to Harun: "The quality control throughout production needs to be ensured otherwise it will be a disaster." If poor quality seed enters at any stage, relationships based on trust will be undermined and farmers will lose confidence.

222

Farmseed

Table 18.1 Characteristics of partner organisations1 in Farmseed model STATUS & PARTICIPATION

NATIONAL NGO

REGIONAL NGO

LOCAL NGO

CBO

Initial number of partners

4

15

31

16

Partners remaining at end of project

1

3

14

13

Drop out rate (%)

75

80

55

19

Number of non-agricultural programmes

High

Moderate

Few/ Moderate (70/30)

Very few

Agricultural programmes

Few

Few

Few

Moderate and small scale

Interaction with resourcepoor farmers

Moderate

Moderate

High

Very high

Agricultural graduate or trained field staff

Few

Very few

Very few

None/ Minimal (90/10)

Moderate

Moderate

Positive

Very positive

Minimal

Minimal

Moderate

Maximum

Partnership attitude Level of interaction with project 1

Non-governmental organisation (NGO) and community-based organisation (CBO)

Table 18.2 presents the different categories of seed quality. Until the late 1990s, the governmental BADC was the only source of foundation seed, although in theory any reliable source could provide it. By the on-set of this sub-project in 2001, also large NGOs such as BRAC and Proshika produced foundation seed, and passed it on to their contract farmers to rear certified or truthfully labelled seed. A few NGOs have their own seed farms. Staff shortages at the Seed Certification Agency (SCA) makes that most seed producers cannot get their seed labelled as foundation or certified seed. Instead, they label it themselves as truthfully labelled seed within the provision of the seed law (see Chapter 15). Producers then bear full responsibility with regard to quality. With Farmseed, farmer seed producers produce quality seed, but it is not graded and stored by a third party to be sold as truthfully labelled seed, or certified by SCA to be sold as certified seed. Foundation seed injected into the community becomes a source supporting quality seed production. The seed remains in the community and quality is guaranteed through pressure of the seed producers' peers. To replicate Farmseed, it is important for an organisation to understand in which niche it can perform best. In the seed sector, smaller NGOs like AAS could best 223


Table 18.2 Main producers and distributors of rice seed in Bangladesh SEED CATEGORY

CURRENT PRODUCER

POTENTIAL ADDITIONAL PRODUCER

Breeder seed*

Bangladesh Rice Research Institute (BRRI)

No changes expected

Foundation seed*

Government (BADC) Large NGOs Private sector

Other NGOs with a strong agricultural agenda like AAS

Certified seed*/ Truthfully labelled seed

Government (BADC) Large NGOs Private sector

Regional and local NGOs and more farmer seed producer groups; lack of official certification personnel will result in seed being truthfully labelled, rather than certified

Quality seed

Farmer seed producers

No changes expected

Farm-saved seed

Farmers

No changes expected

*Only these receive certification from the Seed Certifying Agency.

explore a role as foundation seed producer. A small role at first glance, but one of big potential impact for poor farmers.

CURRENT SITUATION OF THE SEED SECTOR As mentioned earlier, the private sector is mainly involved in producing and selling vegetable seed and only to a very limited extent sells hybrid rice seed from China. Since the government approved the national seed act in 1997 and signed the new seed rules in 1998, the seed sector in Bangladesh started decentralising production and distribution. BADC still puts its seed on the market via private dealers, but now also large NGOs have captured part of the market and sell seed through their own centres (Figure 18.1). Until recently, only large NGOs have developed their own seed system, striving for vertical integration of the production process. They sell their seed as part of a credit package at slightly inflated prices compared to the subsidised seed of BADC. In this new system, farmers pay a relatively higher price for the seed, but in return enjoy greater certainty of seed quality and timely availability. Most poor farmers who are not members of a credit group, however, are not aware of or informed about new varieties and quality seed. To resolve this, AAS developed the Farmseed model whereby foundation seed is provided to poor seed producers through various local NGOs and community-based organisations. The whole approach boosted quality seed exchange between farmers, as will be discussed later. 224

Farmseed

At the end of the project, AAS started experimenting to see if they could produce foundation seed efficiently themselves. This would further reduce the number of stakeholders in the seed supply chain from four to three. Farmseed created a shortcut: seed produced by seed farmers became directly available in the community without being channelled back into government agencies or NGOs for processing and marketing. Farmer seed producers are free agents, not hired labourers. They themselves reap the profits from their work and the respect of their neighbours. Peer pressure ensures quality control: they must keep their neighbour-customers happy with the quality of the seed, or go out of business.

Large NGO (BRAC, GKF)

NGO Selling Centre

Farmers in Centre Credit Groups

Private Dealers

Farmer

Contract Farmer

Breeder Seed Supplier (BRRI)

Government Agency (BADC)

FS Farm Contract Farmer

Small National NGO (AAS)

New Link Explored

Resource-Poor Seed Producer

Breeder Seed

Foundation Seed

Resource-Poor Farmer

Certified, TLS or Quality Seed

Figure 18.1 Experimenting with inclusion of smaller NGOs in the rice seed chain by AAS

225


EVOLUTION OF THE MODEL Historians help us to learn from the past to better understand the present, and avoid making the same mistakes in the future. Archaeologists decipher the life ways of a vanished society by interpreting the remains of its material culture. By describing the past of Farmseed, we can trace significant changes in the philosophy of AAS. In May 2000, AAS started its first PETRRA project demonstrating modern rice varieties. The PETRRA uptake forum insisted that the emphasis had to be on testing new extension methods, and not just demonstrating technology. At the time, PETRRA had approved about ten projects and its main challenge was to get this new working philosophy across. During field days at regular intervals, farmers evaluated different varieties. Just before harvest, they expressed their priorities. Of the 18 varieties under evaluation, only three had made the final selection depending on season, soil fertility and field elevation. The basis for Farmseed was laid; AAS now had a solid understanding of farmers' preferences of varieties. They started training farmers and distributed 10-kg bags of foundation seed, which they purchased from BADC. However, the amount of seed provided proved too high: using one seedling per hill, 10 kg could plant half a hectare. As most fields of poor farmers are smaller than this, they ate at least two thirds of the seed. Learning from its experience, AAS started to repackage the seed into three-kilogram bags and Farmseed became increasingly recognised as a workable method by project staff and farmers alike. From then onwards, AAS started looking into improving its packaging. Innovations don't just happen; they are made to happen through a concentrated effort.

HOW TO SET UP A FARMSEED MODEL? Build partnerships and train trainers Initially, AAS selected numerous partner organisations that were active in five districts in Northeast Bangladesh, all of them local NGOs and community-based organisations. As their staff had little agricultural expertise, training helped build their skills and confidence. In total about 100 staff (including some from AAS) were trained in producing rice and quality seed at the BRRI Training Division. Four batches received classroom and field training for a week, with a tailor-made curriculum. Build resource-poor farmer seed producer groups Although many NGOs have set up farmer credit groups in Bangladesh, an important difference of Farmseed is that it uncoupled the link between credit and 226

Farmseed

inputs. New groups were formed and the community selected its members. Only those who were truly poor were allowed. Farmers were then asked about their interest in becoming trained as a community seed producer. Their wives would be trained as well. Male and female groups were formed with an average of 25 and 28 motivated members, respectively. Each group selected a coordinator who was not necessarily the same person as the one of the credit group. Group coordinators got enhanced social status, being elected by their community members as smart, trustworthy and to whom they could turn to for advice. Assess suitable varieties and validate technologies Foundation seed provides the fuel for the Farmseed engine. Yet, one needs to know whether the engine runs on diesel, petrol or gas. Through on-farm demonstrations, farmers selected the three most promising of 18 varieties by season-long monitoring. Improved seed storage practices, tested by BRRI, were validated and fine-tuned by roughly 50 women in 25 different villages. Adding one tablet of naphthalene per kg of seed in storage kept the seed free of pests for six months. Ensure access to foundation seed Nearly all farmer seed producers requested new foundation seed every season, so ensuring timely access is one of the most important requirements to make Farmseed work. Knowing which fuel is needed doesn't help if there is no petrol station around when you are running empty. Currently, foundation seed can mainly be procured from BADC. Large NGOs already produce their own, but this is used within their own seed production system. If smaller NGOs with a strong agricultural background could purchase breeder seed from BRRI, they could produce their own foundation seed. AAS, along with some other NGOs, is experimenting with this to see if it fits their organisation. This is described in Chapter 17 on the rice seed network.

Resource-poor farmer seed producers cover different age groups. But all are dynamic and motivated to better their lives.

Conduct group training activities Every cropping season, both men and women learnt about rice and seed production. One training at the community level typically lasted 2-3 hours. Additionally, women-led group extension focused on post-harvest technologies (Chapter 3). In total, about 7,500 227


members received training (Table 18.3). Later on, and drawing on the experience of the NGO Shushilan, AAS learnt about the importance of training women in all aspects of agriculture, as described in Chapter 4. Table 18.3 Number of farmers trained under Farmseed SEX

GROUPS

TRAINING SESSIONS

MEMBERS TRAINED

SEED* PRODUCER

MEMBERS

Men

160

800

4,000

2,000

13,000

Women

125

125

3,500

Total

385

925

7,500

n.a. 2,000

13,000

*Based on performance, the community selects trustworthy high quality seed producers.

"I no longer have to buy, but can actually sell seed," says Shamima Akhter during a village fair in Kishoreganj. "My husband, mother-inlaw and neighbours respect me much more now," she concludes with a smile. Shamima was trained by AAS in the women-led group extension project (see Chapter 3).

Create incentives for all actors At the start of the project, seed and fertiliser were provided to get poor farmers motivated. The second season, only seed was provided. Once trained, farmers could buy seed from AAS at a price slightly higher than this fixed by the government (BADC). In 2004, seed producers bought their foundation seed at Tk 18 (US$ 0.32) per kg and sold their quality seed at Tk 10-15 (US$ 0.18-0.26) per kg, depending on the season, availability, and purchasing and bargaining power of their clients. Although most seed was sold or exchanged within the community, some took the initiative to sell their seed at markets as far as 20 km from their home. More than 30 partner organisations contributed about 50 staff members as trainers or facilitators during field sessions, without being paid for it. Apart from the training, no other incentives were given. Once seed producers were established, some of them took over as communication hub between AAS and the other farmers. In 2003 and 2004, AAS produced 5-10 tons of foundation seed to see if it fits their organisation. If they could produce 20 tons, AAS would be able to recover all its costs and supply seed to its seed producer groups without increasing the price. Control quality Before harvest, the seed producer groups, AAS staff and other members of the community assess the seed quality during a field day. About half of the 4,000 farmers

228

Farmseed

trained as seed producers now still produce quality seed (Table 18.3). The quality of their seed easily equals that of certified seed. Knowing the seed producer eliminates the need for a label. Nikesh Gop, a 26-year old seed producer in Uttar Varuara village in Moulvibazar district illustrates clearly what is at stake: "In boro 2003, my BRRI dhan 28 seed showed poor germination, so I decided not to sell any seed that season. It is better to sell nothing than to loose my reputation in the village." Having learnt how to accurately test seed germination, strengthened seed producers' confidence. As farmers increasingly became aware that proper drying and storage was needed to maintain seed quality, the demand for plastic drums rose. How AAS dealt with this is described in Chapter 3 on women-led group extension. Monitor regularly Receiving feedback from poor farmers, men and women, was made easier by working with partner organisations, especially with the community-based organisations since these are an integral part of the same community. Working with the latter proved more successful than with NGOs. Let it roll, but keep eyes on the ball Depending on the variety and local conditions, farmers assigned between 30-70% of their field to seed production. Each seed-producing household used their own specific strategy to produce, store, consume, sell and exchange seed or seedlings. Flexibility is important for farmers in times of distress. Farmers kept about 10-20% of the seed they produced, and sold or exchanged the rest within the community. From its start to 2004, exchange of seed increased from 5% to more than 30% in the project area.

Bartering in Bangladesh.Women exchange rice grain or seed for various consumables within their community. Not only with their neighbours, but also with village-to-village vendors of hora malsa cooking pots and other consumables.

Once the system was established, it was important to keep regular contact with the partner organisations and group coordinators. Foundation seed must be regularly supplied, along with the vast number of ingredients that keep a relationship going. AAS, for instance, involved many organisations in other projects, further strengthening their agricultural capacity.

KEYS FOR SUCCESS One of AAS's strengths is its small management unit and decentralised offices throughout the country, making it lean, 229


flexible and able to deliver at low cost. But their staff also has strong technical expertise in agriculture; Harun has good links to government research and development agencies; and the whole organisation can rely on a well-established network of NGOs and community-based organisations, consciously developed over the past 15 years. Bangladesh is the country of the NGOs, one often hears. The fact is that there are plenty and nearly all lack technical expertise in agriculture, an ideal starting point for collaboration. On the other hand, recognising your limitations is probably equally important. AAS, having a limited number of staff, continuously tries to liase with trustworthy partners who work with local communities. But despite its long experience in the field, it is still often a matter of trial and error when moving into a new area; only half of the partners remained in the Farmseed project after four years. "Those partners who do not have a sincere interest in agriculture quickly leave, but this is not always easy to tell from the beginning," says Harun. The ones that stayed had strong ties to the community, which AAS needed. Although most lacked agricultural knowledge, they were eager to expand their programme and receive training from AAS. To develop Farmseed, partnering with community-based organisations worked better than with NGOs, since they live in the community and are farmers themselves, reducing the communication gap with other poor farmers. Besides, and as mentioned earlier on, most NGOs do not have any agricultural expertise. AAS also reflected on the past to capitalise on its experiences. In the nine PETRRA sub-projects it implemented, the success and lessons learnt from one project were integrated in the others (see Chapters 3, 4, 8 and 9). Harun is a friendly and respected man who looks after his staff well. He happily takes in every learning experience that arises and that helps build the capacity of his staff members. When two scientists from the UK-based intergovernmental organisation CABI Bioscience recently knocked on his door to try to establish community plant health services, he immediately seized the opportunity and had four of his staff members join in the training course. AAS is an innovator. It works proactively. Although the Farmseed project still had another year to go, AAS understood that incentives had to be ensured at all levels of the system. For the group coordinators social recognition was enough in the initial phase, but some money would be required to keep them motivated over a longer time. This partly explains why Mr. Harun started exploring the possibilities of producing foundation seed in-house. If successful, the revenues could be reinvested to sustain the system.

DIFFICULTIES, RISKS AND ASSUMPTIONS Especially in the early stages, capacity building and motivating farmers requires 230

Farmseed

investment. Under the current mandate, AAS cannot sell anything for profit, so it remains questionable whether a small and cost-effective technical NGO can keep on providing the same level of service to poor farmers without support from the public sector. Surely, AAS would require starter funds to initiate Farmseed in new areas. Networking is key to the success of Farmseed, yet stronger institutions with more experience in agriculture often stepped out of the partnership. The model may have competed with their vested interests or mutual benefits were not high enough to maintain a synergistic relationship. Quality control has to be ensured at all levels. In two places, farmers faced problems with bakanae, a seedborne disease that also survives in the soil. It was important to have good links with research institutes and communication with local partner organisations so that seed could be grown in other, disease-free areas. This is one of the reasons why AAS set up Farmseed in Rajshahi in Northwest Bangladesh, where bakanae was not a problem. Thanks to the swift response of AAS, infested plants are continuously rogued to avoid the disease further developing and spreading to other areas. To strengthen its human capacity in pest and disease management, AAS joined the Plant Health Services Initiative, established by CABI Bioscience (www.globalplantclinic.org).

Farmers can easily be taught to recognise bakanae infested seedlings (they are elongated) and to remove them in the seedbed and field, but not all diseases have such clear symptoms. Basic training in diagnosing plant health problems in the field is often lacking in integrated pest management (IPM) programmes.

SCALING UP Convinced of the potential of Farmseed, AAS recently took this model to Rajshahi using their own limited funds. By testing it in Northwest Bangladesh, AAS anticipates gathering more proof of its nation-wide applicability. In 2004, it started testing the potential of Farmseed for growing onion seed with about 20 farmers, and hopes to apply the model to other crops including potato, garlic and banana suckers.

CONCLUSION Over four years, the small yet dynamic national NGO AAS built bridges to both technology providers and poor farmers via 64 nongovernmental and community-based organisations. Technologies were validated and adapted; and an effective, low-cost training programme developed. A major incentive for the partner organisations was the development of new knowledge and skills for their personnel. While AAS had a strong agricultural expertise, few of the partner organisations had, explaining the symbiotic relationships in this strategic network. 231


Farmseed, a decentralised and community-based seed production and distribution model, developed under PETRRA, impacted on the lives of around 20,000 poor farmers and shows potential for sustainability once the model is established.

232

19 Grameen Seed

Grameen experiments with a pro-poor seed business

Paul Van Mele Ahmad Salahuddin Md. Abdul Jabbar

SUMMARY The Grameen Krishi Foundation (GKF) evolved from the Grameen Bank in 1989 as an independent NGO in the field of agricultural development. Under their current seed programme, GKF signs up mainly small to medium farmers to produce seed. Lack of mechanisms to assess production targets and create local demand is one of the reasons of poor efficiency of the existing GKF seed system. Recently, GKF experimented with an innovative method by focusing on the poor not only as seed producers, but also as customers, and to help sell quality seed. Per union, ninety farmers that own less than 0.5 ha were selected in a way that ensured maximum geographical coverage. GKF trained them as seed producers, seasonally sold them foundation seed and collected the seed they harvested for drying, grading, storing and packaging as truthfully labelled seed. At the beginning of the next season, the poor seed producers were given priority over seed retailers to buy back their processed truthfully labelled seed. In this new Grameen seed system, incentives are ensured for farmers, retailers and GKF, making it a financially sustainable business. By inviting seed producers, sellers and retailers to the seed processing centre, system transparency is created and confidence built in the quality of the seed. As seed is grown and sold in the same area, all farmers in the system get better access to quality seed. This new seed innovation system works independently of the GKF credit system, but could be complemented by it. Involving other NGOs and local officials helps to further increase the customer base. 233


ACTORS IN TRADITIONAL GKF SEED SYSTEM The seed sector in Bangladesh is steadily decentralising. GKF, among several other NGOs and private companies, has established a memorandum of understanding with the Bangladesh Rice Research Institute (BRRI) to buy breeder seed. Currently, only these NGOs are legally entitled to produce foundation seed on their own farms. This seed is then mainly passed on to their contract farmers to produce truthfully labelled seed (TLS) or certified seed. GKF has a credit programme based on the Grameen Bank principles, which is targeting the poor (Table 19.1). Their agricultural development programme, however, is less poverty targeted, with seed production being dominated by small to medium farms of up to 1.2 ha of land. The GKF unit office manages these seed producers, while addressing other duties such as credit, livestock, poultry and hybrid Table 19.1 Overview of Grameen Krishi Foundation programmes VARIABLE

CREDIT PROGRAMME

AGRIC. DEVELOPMENT PROGRAMME

Target audience

Poor people owning no land or less than 0.2 ha

Small and medium farmers, owning 0.2 to 1.2 ha of land

Main focus

Credit to support mainly non-agricultural activities

Credit to support agriculture

Training provided

Yes

No

corn production. The total working area of a unit office covers 8-15 unions, with each union comprising nine wards (see Box 19.1). The unit office has no defined role in marketing, or in deciding how much seed to produce. The GKF headquarters in Rangpur, Northwest Bangladesh, handles this. Box 19.1 Administrative Units in Bangladesh

6 divisions 1 division = 10-12 districts 1 district = 5- 12 upazilas 1 upazila = 5-12 unions 1 union = 9 wards 1 ward = 1-3 villages

For reasons of convenience, seed production has so far been confined to a few farmers. All truthfully labelled seed produced by contract farmers is collected at the processing centre in Mirbagh, about 15 km from Rangpur. It is responsible for supervising the foundation seed farms, processing and distribution, assisted by headquarters and the Dhaka liaison office (Figure 19.1).

The existing GKF seed production system has certain weaknesses. Due to the topdown, centralised structure, information about the type of varieties and amount of seed required at the village level is inaccurate. The contract seed producers are given no voice to express local farmers' preferences for certain varieties and mostly receive foundation seed from other areas than their own. No attention is paid to the local

234

Grameen Seed

Breeder seed supplier (BRRI)

Head quarters (Rangpur)

Liaison office (Dhaka)

Seed rocessing entre (1)

Unit ffice with foundation seed farm (5)

Unit office without foundation seed farm (82)

Contract seed producers

Private sector dealers

Farmers

Breeder seed

Foundation seed


Figure 19.1 Traditional GKF rice seed production and marketing system Seed leaving the seed processing centre is processed and packaged

context and dynamics in the farming system; seed producers are passive implementers. To a limited extent, unit offices sell TLS seed to farmers upon demand, but are not involved in promoting seeds. As with all major non-governmental seed producers in Bangladesh, seed production and distribution lacks strong interaction with other stakeholders. The marketing system is even more poorly developed than the production system with no mechanisms for assessing or creating demand, often resulting in seeds remaining unsold at the seed processing centre. The system lacks a network and long-term business plan. Neither unit offices nor liaison office are used as pro-active marketing units, or have planned relation with private retailers. 235


EVOLUTION OF GRAMEEN SEED The project described here started under PETRRA in 2000, and its evolution offers some lessons in terms of triggering change in a value-based innovation system. Changes in the system are presented in Table 19.2. At first, PETRRA struggled to get the concept of extension research across to their partners. All of the nine sub-projects approved in the first phase simply did extension instead of trying to improve the seed systems as a whole. The first principal investigator of the GKF sub-project was a person working in one of the unit offices. He kept on demonstrating varieties without trying to bring innovations to the system. Over the years, PETRRA improved the concept of extension method research, as part of its value-based research system. New projects and new phases of existing Table 19.2 Traditional GKF versus Grameen seed innovation system

236

VARIABLE

TRADITIONAL GKF SEED SYSTEM

GRAMEEN SEED INNOVATION SYSTEM

Decision-making and variety selection

Centralised, top down

Decentralised with feedback from seed producers and poor farmers

Integration of system components

Processing and marketing are not linked to production

Integration of different sub-components

System focus

Production focused

Holistic with stronger emphasis on market development

Seed sales

Often not all truthfully labelled seed is sold

All truthfully labelled seed is sold

Linkage with other actors

In-ward looking, no concerted effort to link seed system with other actors

Multiple actors (government extension, NGOs, retailers and others) are key to market development

Role of seed producers

Regarded as hired labourers

Considered as producers, consumers, seed sellers and marketing agents to expand customer base

Organisation of seed producers

Contract seed farmers act as individuals

Group approach

Poverty focus

Inclusive, but small and medium farmers dominate

Targeted, only the poor are trained as seed producer

Gender

Not considered

Women are trained, with special emphasis on post-harvest

Location focus

Not important, seed can be produced anywhere

Seed is produced in area of customer base

Local context

Not considered

Sensitive to local needs and aspirations

Grameen Seed

ones were approved only if they emphasised innovating extension and uptake methods, rather than doing extension. PETRRA's management worked closely with GKF to rethink their existing model, injecting new ideas and facilitating group learning (see also Chapter 1). Two years later, the principal investigator of GKF left the project and their managing director appointed a new person, Mr. Md. Abdul Jabbar, who already had 11 years of experience working in the GKF seed sector. Being in charge of the seed processing centre, he understood all too well the weaknesses of the GKF seed system. Innovating systems, or experimenting with new organisational models, is harder than experimenting with new technologies and it takes courage. It is our experience that value-driven people are the best vehicles for change. When asked what he finds the most rewarding aspect of the project, Jabbar reflects a minute and says: "I am very happy that I could facilitate the knowledge dissemination to poor farmers and improve their access to quality seed. Seeing poor people's livelihoods improve is very rewarding." Although the experience presented here is based on work by the seed processing centre, which inherently limits the scope for extrapolation to the national level, the model would most probably not have been developed by a person who had not been exposed to the different components of the system. The Grameen seed model presented here should ideally be tested for another 2 years, redefining the role and incentives for the unit offices. Figure 2 shows a much broader anticipated role of the unit office as a hub from which all unions could be supplied with quality seed. In the new model, the seed processing centre also takes responsibility for coordinating seed production and marketing, and so could be turned into a seed centre with a broader mandate.

THE GRAMEEN SEED INNOVATION SYSTEM Build resource-poor farmer groups Choosing poor farmers is key to the success of the model, as a first step towards integrating production and marketing. With the prospect of building a large customer base, attention is paid to maximum geographical coverage both within and between the groups. In the project, all 9 wards of one union were covered. In each village, 10 farmers were selected during a village meeting, based on three criteria: farmers had to (1) be rice producers; (2) own less than 0.5 ha of land; and (3) have a rice provisioning ability of 6-10 months. If more than 10 people complied with these criteria, further selection was made to avoid geographical clusters within the village or having several brothers or other family members in the same group. All farmers in the group were registered and lists kept by GKF.

237


During the same meeting, the 10 farmers elected their own group leader. Farmers mainly chose people: (i) who have courage to speak out and communicate easily with other farmers, NGOs, government extension staff, retailers; (ii) who want to

Breeder seed supplier (BRRI)

Head quarters (Rangpur)

Seed centre coordinating production & marketing (1)

Unit office with foundation seed farm (5)

Unit office without foundation seed farm (82)

Union staff (8-15 per unit office)

Resource-poor seed producers (90 per union)

Private sector dealers at village level

Farmers

Breeder seed

Foundation seed

Figure 19.2 Grameen seed innovation system Note that downward seed flows have been processed and packaged

238


Grameen Seed

learn new things; and (iii) who have a certain education. Although Mr. Jabbar suggested rotating the leadership within the groups, it is too early to see evidence of this happening. Train farmers (husbands and wives) on rice and seed production For practical reasons, three groups of seed producers came together in a gathering place, such as the village school or club, for a one-day training session at the beginning of each season. During the first training, the wives of the group leaders were also invited to attend, bringing the total participants to 33. In following seasons, three other women were invited on a rotational basis. Rice and seed production were addressed pragmatically , with special emphasis for women on post-harvest. Women were asked to share their learning with neighbours. Women more easily share new knowledge through informal networks than men do (see Chapters 3 and 4). Better knowledge of the whole cropping cycle resulted in more balanced use of fertilisers and less use of pesticides. Out of the 90 farmers trained, about 80 had significantly higher yields and all produced better seed. Develop incentive-based production and marketing mechanisms The training provided to seed producers did not only benefit seed production, but also their crop production and well-being overall. Higher yields and better seeds help raise incomes and women feel recognised for their important role in rice production. The emotional value of farmers towards their own seed is often underestimated in national seed systems. Processing seed separately for each farmer would not be feasible, but with the Grameen seed innovation system, farmers are at least guaranteed to get TLS seed produced by their peers in their own locality.

Women share what they have learnt from training sessions with other family members first. Later on, they also discuss this with friends and neighbours while doing joint activities such as cleaning taro, a local tuber crop.

Seed producers need to buy 2 kg of foundation seed for their seed production plot, and as much TLS seed as they want for personal paddy production and for selling to their neighbours. By selling processed and packaged TLS seed to their neighbours, they make an additional profit of Tk 2 per kg (Table 19.3). Price of paddy is at its lowest during harvest and poor farmers have few facilities to store seed properly for extended periods. GKF offers seed producers Tk 1 per kg on top of the prevailing market price for paddy. GKF buys unprocessed TLS from seed producers at 239


Table 19.3 Price (Tk per kg) that clients pay to different seed suppliers in the Grameen seed innovation system, boro season 2003 SUPPLIER

GKF

SEED PRODUCER

DEALER

CLIENT GKF Seed producer Dealer Farmer

9 18/14* 14.4 16

16

*Seed producers buy at least 2 kg of foundation seed at Tk 18 per kg for truthfully labelled seed production and as much truthfully labelled seed at Tk 14 per kg as they want for personal paddy production and for selling to their neighbours, a slightly preferential price compared to what dealers pay. (In 2004 the Taka was 57 to the US dollar).

Tk 9 per kg (US$ 0.16) and sells it back by the start of the next season at a minimum of Tk 14 (US$ 0.25). As the total cost for processing and packaging, including the salary of the workers, is Tk 3 per kg (US$ 0.05), GKF makes a profit of Tk 2 per kg (US$ 0.035) of processed seed, besides paying the salaries of seed plant workers. Another innovation is that Grameen seed producers are given priority to buy back their TLS seed after it has been processed, stored and packaged, and this at a slightly lower price than retailers would have to pay. Seed producers purchase their TLS seed at Tk 2 below the market price, while retailers are only given a profit margin of 10%. In this way, a business mentality is developed among the farmers, while retailers are still pleased with their profit margin. Cost-benefit analysis indicate that by selling the seed at Tk 15.2 per kg (US$ 0.26) would allow GKF to fully recover training costs and make the system financially sustainable without any project intervention. Most importantly, clear incentives to all stakeholders are maintained. TLS seed producers also sell the seed, ensuring an exponential increase in access to quality seed for all farmers in the village. Production and marketing is integrated at the system's heart by the poor seed producers (Figures 19.2 and 19.3). Seed producers also value dearly the guarantee of timely access to foundation seed. Intensively monitor seed producers for three to four seasons A classroom session was used to introduce new topics, but the real learning took place in the field. Throughout the season, seed producers were monitored and helped to set up seed production plots in one of their fields. Plot size varied depending on the farmer, but was generally around 600 square meters. If farmers had more than one field, they were encouraged to set up a plot in the field that is most accessible to other community members. This way, seed production plots also served as demonstration plots. 240

Grameen Seed

Neighbouring farmers Sells 190 kg TLS (5) GKF Seed Processing Centre

Buys 2 kg FS (1)

Resource poor seed producers

Buys 200 kg TLS (4)

2 kg FS for seed plot (2)

Sells 240 kg TLS (3)

10 kg TLS for paddy production (5)

600 m2 seed plot

Foundation seed (FS)

4,400 m2 paddy field

Truthfully labelled seed (TLS)

Figure 19.3 Example of multiple roles of a seed producer in Grameen seed innovation system (numbers indicate order of flow)

After the project had gone through three seasons the seed producers were qualified enough to work more independently. In terms of investment, GKF contributed two staff members to train 90 seed producers over nine wards and one union, Mr. Jabbar being appointed for 33% and a field staff for 50% of their time. Expose multiple actors to quality seed producers Currently, poor seed producers sell truthfully labelled seed, as do local retailers. Local leaders, government extension and NGOs also bring in new customers. All these actors, along with other farmers from the village, are invited to a field day. This not only creates demand, but also announces which persons are the sources of trustworthy quality seed. The long-term objective of this new GKF approach is to reach all farmers in any union where the model is introduced. Link quality control to collection system Trust is crucial in selling seed, especially for truthfully labelled seed which involves 241


no official certification. Stakes are high for all seed suppliers, because distrust in a system always spreads faster than trust. Every season, GKF ensures that registered seed producers buy their foundation seed. At harvesting time, union staff are contacted and the seed quality is verified. Varieties should not be mixed and seed should be pest free. An assessment is also made of variety uniformity, seed maturity, colour and moisture content. Preferably, seed should have less than 14 % moisture. If it is higher, a correction factor is used to adjust for the weight. Farmers bring their seed to the unit office, where bags are labelled with information such as the name of the farmer and variety, and they get paid on the spot. From the unit office, a truck collects all seed of one and the same variety within a given area. Occasionally, two different varieties are collected, keeping the bags separate and discarding seed that drops out of the bags.

Mr. Jabbar shows a polythene lined seed bag in the GKF seed processing centre during a visit by a farmer seed producer group. The guided tour helps to give them confidence in what happens to their seed.

Build system confidence for seed producers, retailers and clients Transparency creates credibility and confidence, which is crucial for attracting customers, especially in rural areas where word-of-mouth advertisement prevails. Apart from the field days, GKF organises visits to their seed processing centre so seed producers and retailers can appreciate how it adds value. Truthfully labelled seed is packaged in specially lined 10-kg bags that prevent moisture absorption. A leaflet with information about the production technologies is placed in the bag, which is sealed with a tag describing the seed. On request by poor farmers, 5-kg bags are also made.

KEYS FOR SUCCESS Main keys for success have been discussed at length in the above section. They include: Ensure good performance of new varieties through appropriate variety selection and training of smallholder seed producers Create demand by exposing farmers and retailers to new varieties Build transparency and trust in the seed innovation system Ensure incentives for all stakeholders

242

Grameen Seed

Integrate production and marketing Ensure access to foundation seed Add value through seed processing Build local capacity for quality control

DIFFICULTIES, RISKS AND ASSUMPTIONS The seed processing centre took the lead for this project. Mr. Jabbar feels it is important to let the method grow organically. Better let it grow strong and slow, than to bloom too soon and then collapse. As GKF has only one seed processing centre, building confidence in and transparency of this part of the system is a challenge for scaling up. Perhaps a video could be produced to overcome this barrier. Copies could be held in each of the 87 GKF unit offices and shown to new seed producer groups as part of their training. Institutionalising and mainstreaming this new Grameen seed model within GKF requires a strong commitment from the managing director. In 2003, this position has seen three changes. PETRRA tried to approach the top Grameen management to advocate this pro-poor seed innovation system, but without success so far. As IRRI, being an international research institute, has a permanent presence and strong social capital in Bangladesh, discussions may still take place in future.

SCALING UP While exploring how GKF sees their position evolve given their well-established competitors, Jabbar explains their long-term business plan: "We have already established 87 unit offices covering hundreds of unions in Rangpur and Tangail region. In most places, the demand for quality seed is very high, so there will be no problem in selling seed. In case other established actors such as retailers or NGOs cannot respond to the demand of their clients, they are encouraged to buy from our GKF farmers directly. That is why we invite them all during our field days, so that they can get to know each other."

Taking seed processing seriously. Seed processing by BRAC, the country's largest NGO, is taken very seriously, but their plants do not operate at full capacity. If BRAC were to sell services to other NGOs and private sector seed producers, all actors including the poor farmers- could benefit.

GKF is also a member of the Northwest focal area forum, in which research and

243


development activities are better coordinated between government, NGO and private sector (see Box 21.1). The Grameen seed model will not require too much adjustment of the existing GKF structure, as the emphasis of their agricultural development programme already changed from irrigation to input supplies and credit. The old irrigation focus had geographical limits and addressed a limited number of small to medium farmers, while input supplies and credit involve many more customers, in a much wider area. If Grameen management, and especially the founder Dr. Yunus, can be convinced, the system has great potential to spread to other parts of the country and improve the livelihoods of poor farmers, men and women. To overcome the limitations in seed processing capacity, agreements may need to be established with other service providers such as the governmental BADC and the NGO BRAC, who have seed processing units around the country. This would allow GKF to further replicate their model.

CONCLUSION The Grameen seed innovation system breaks down the barriers of quality seed access at the grassroots level. As farmers across Bangladesh lack access to quality seed, the market potential for GKF is enormous. By involving poor seed producers in seed marketing, no additional costs are required to develop a sales network, which may lead to much more rapid adoption of new varieties. If mainstreamed, the new seed innovation system would trigger a shift in GKF's agricultural development programme, namely to focus on the poor and provide agriculture-based training for both men and women at the community rather than at the upazila level.

244

20 Innovating with Federations

Community institutions take the lead in seed marketing

Syed Samsuzzaman Paul Van Mele

SUMMARY The NGO Rangpur Dinajpur Rural Service (RDRS) has a strong expertise in organising poor households into federations, comprising 30-40 groups and operating at the union level. So far, RDRS has established 256 federations covering 268,000 households in Northwest Bangladesh. In 2000, PETRRA stimulated RDRS to experiment for the first time with uptake methods for rice seed. The project described here is the first of a growing number of innovative RDRS projects on rice. The federations give poor farmers a support framework to become successful rice seed producers. Trained seed producing households get Tk 1.5 ($0.025) per kg over and above the grain price. The federation dries, stores and repackages seed from about 20 households into 10-kg bags of truthfully labelled seed. A seed committee, comprising five federation members, is responsible for the quality control and marketing of the seed. They sell the seed at twice the grain price. To ensure sustainability of this innovation system, incentives are ensured for all actors, with profits being shared between the seed committee, RDRS and the federation on a 40:40:20 basis. By August 2004, RDRS had trained more than 500 rice seed producers in 28 federations, who together provided quality seed to roughly 14,000 resource-poor farmers. 245


ACTORS AND NETWORKS RDRS is a regional NGO that mobilises the rural poor for social and economic development into community-based organisations. RDRS provides vocational training and microcredit to groups of about 20 poor households. Each group elects a chairman and a vice chairman; the latter must always be a woman. After about five years, when a group is considered socially graduated, it can join the federation, which comprises 30 to 40 groups and operates at the union level. Through a revolving funds mechanism, the approach has been expanded to more than 250 federations, some of which have started establishing networks with other institutions for increased access to services and resources. Under the PETRRA seed uptake project, various federations were selected based on their interest and availability of resources, like suitable land. RDRS, with its long experience in establishing groups and running microcredit programmes, is responsible for managing and recovering input costs from participating farmers. Three senior staff members of the Bangladesh Rice Research Institute (BRRI) provided training on seed production and processing to field staff of the Department of Agriculture Extension (DAE) and RDRS. These are in turn responsible to train farmer trainers and monitor seed producer groups.

EVOLUTION OF THE METHOD Quality rice seed is available in the region through the governmental Bangladesh Agricultural Development Corporation (BADC), but the main problem is that seed of the right variety is often not available at the right time. A delayed access of 2-3 weeks seriously disrupts farmers' planting and harvesting. Before working with PETRRA, RDRS had a vegetable seed promotional programme, but seed production was limited and only for use by the target households. In 2000, PETTRA stimulated RDRS to embark on research on uptake of rice seed. Although farmers were trained to produce quality rice seed at the federation level, both RDRS and the federations still lacked experience in selling seed. So RDRS approached the Bangladesh-German Seed Development Project which strengthened the production, business and marketing capacity of RDRS's staff through several training-of-trainer sessions. The emerging RDRS seed innovation system is continuously evolving (see photo).

THE RDRS SEED INNOVATION SYSTEM Select federation Federations are selected based on their interest and availability of resources like

246

Innovating with Federations

Mr. Imul Mosadek from the Bangladesh Agricultural University collects data for his MSc on hybrid rice seed with the help of Ms. Bulbuli Rani and her husband. Bulbuli is vice-chairman of one of the RDRS federations that started growing seed.

suitable land, grain storage facilities and interest of their members to become seed producers. Establish seed producer group and seed committee The executive committee of the federation forms a seed committee comprising five members, of which at least one is a woman, and who are in charge of seed quality control and marketing. The federation along with RDRS then selects about 20 households who are interested in producing quality seed and willing to follow the guidelines. Overall, men are trained to become seed growers while the women are trained in seed processing. However, women are increasingly involved in field activities themselves, as described in Part II on gender. Train master trainers Prior to PETRRA, RDRS had no experience of working with rice. It was a big leap for them, but they did it when they recognised how important rice is for alleviating poverty. Three BRRI senior scientists trained 15 RDRS staff, who in turn taught about 250 farmer seed producers. At first, RDRS covered the costs via projects, but later on federations started to organise and pay for the training from its profit margin from selling seed. In each of the seed-producing federations, RDRS staff also trained 3-6 farmer trainers on rice cultivation, who in turn taught about 2,500 farmers. Improved rice production stimulated demand for quality seed. Currently, farmer trainers facilitate sessions on a hire contract basis about twice a month, receiving Tk 50 (US$ 0.90) per session. This money comes from the overall federation budget, as such reinvesting profit from the revolving funds and other activities. 247


Train federation-based seed producer groups About 15 to 25 federation members are organised into a seed producers' group and receive training three times a season: during planting on improved production techniques; during maximum crop growth stage on crop protection; and during harvest on threshing, seed drying, and so on. Training takes place at the federation's facilities and in the field. Moving away from residential training, where participants stay for three to five days in the RDRS training centre, has made it easier to offer training to women. The family approach in agricultural training is used, with far reaching impacts on household decisionmaking and women. This approach was tested by several other NGOs under PETRRA, one experience being described in Chapter 4. The vice chairman of the Mohendra Nagar federation in Lalmonirhat district and member of the seed committee, Ms. Bulbuli Rani said: "Women need to receive training on field activities, not only post-harvest. This is especially important for activities such as weeding and rogueing, as our husbands may be out of the house to do another job at a crucial time in the season." "Having training at the federation facilities allows us to attend more easily. Before we only received residential training and had to stay overnight three days or more, making it hard for those with babies and children."

All seed committee members also received training on seed business and marketing. When Paul Van Mele asked her what major insights she got from that training, Bulbuli replied: "Before I didn't consider it, but now I realise that the customer is my God. We should respect their choice (of variety)." Assess market demand for quality seed When establishing a new seed producer group in a region, it is very important to know what the market demand is: how much of what variety is needed at exactly what time of the year? Only then the group can establish an annual seed production plan and place specific requests for breeder seed to BRRI. The regular federation meetings, with representatives from the different groups, enable the seed committee to quickly gather information from its on average 1,000 members. For sales outside the federation, initially information about local preferences is collected with surveys during weekly markets. Making use of a place where farmers are already gathered saves transaction costs and gives a more or less unbiased selection of farmers to gauge demand. A similar approach called Going Public, uses two-way communication to not only gather but

248


also share information is described in Chapter 9 and by Bentley et al. (2003). Once local preferences are known, the seed committee prepares a seed production plan and discusses it with the seed producers. After harvest, they actively market their seed to non-federation members within the union in an innovative way (see Box 20.1), resulting in people placing demands for the next season. The RDRS federation model also allows for efficient exchange of information and seed between federations. Lalmonirhat district covers nine unions, each with one federation. Over about three years, all the farmers in the area know about the quality seed from the Mohendra Nagar federation. Due to increased awareness, demand is increasing steadily every season. For the coming season, this federation targets to produce 10 tons of truthfully labelled seed, about double what they produced this season. To produce 5 tons they had to procure 150 kg of seed from BRRI. The long-term target is to produce 40 tons of seed. Establish federation-based processing, storage and packaging facilities Appropriate facilities at the federation level are necessary to ensure delivery of quality seeds. Federations with a grain storage facility can easily convert it for seed storage. The minimum requirement is a cement drying floor, while some federations have a seed drying machine obtained through a previous project funded by DFID. These seed dryers are particularly useful in a community-based seed system, whereby large quantities of seed have to be dried. They could potentially be shared between federations. PETRRA funded the purchase of two small organic cocoons (see photo), each with a capacity of storing five tons of seed and costing Tk 100,000 (US$ 1,750). Three others have been purchased through a GTZ project and eight more through revolving funds. Alternative community storage structures are being explored with a local plastic-producing company. By August 2004, two other federations used RDRS "We hang banners at local bazaars, markets, banks, the Union Parishad, anywhere people gather. We let everybody know that quality seed of BR 14, BRRI dhan 28 and BRRI dhan 29 is available from our federation," explains Ms. Bulbuli Rani during one of the meetings at the Mohendra Nagar federation in Lalmonirhat district in April 2004. As vice-chairman, she is a well-respected community member. Mr. M. G. Neogi, the senior programme manager agriculture from RDRS replies, "This is like Going People." Four days earlier he had heard Paul Van Mele's presentation on Going Public, during a national workshop in Dhaka where he, among many other NGO and government extension service providers, learnt about a range of innovative uptake and extension methods developed under PETRRA. Driving back to the RDRS campus, Neogi is still pondering: "Since our childhood, we see the medicine men using the hat (weekly market) to sell their products, but we never thought we could use this approach in our own profession." In September 2004, Neogi informs Paul that RDRS will mainstream Going Public in their agricultural programme from 2005 onwards.

Box 20.1 Hanging Banners

249


revolving funds to transform existing rice storage into seed storage facilities, bringing the total of well-equipped federations to 15. Create incentives for all actors The incentive mechanism for farmer trainers has been explained earlier. In this part, we will elaborate on the innovative incentive mechanisms for the other actors.

Organic cocoons are made of white PVC with an airtight zip and reduce post-harvest losses to less than 1%. The one depicted in the photo can store 5 tons of rice seed. Trained farmers have overcome the psychological barrier of storing seed at the community. But as cocoons are expensive, alternative solutions are being explored with local manufacturers.

First of all, the clients of quality seed need to be considered, as community demand for quality seed will be the first incentive for seed producers. By using quality seed of modern varieties, farmers' rice yield has increased by more than 50%, reaching 5.7 tons per hectare in boro season (November - May) and 4.9 tons per hectare in aman season (July - November) leading to a big improvement of food security for resource-poor farmers. The clients of truthfully labelled seed clearly reap benefits and get a quick return on investment. Once demand is ensured, at what profit will federation seed producers remain interested? And how much is needed to recover costs for service provision such as marketing support? The federation pays their seed producers Tk 1.5 (US$ 0.025) per kg of seed over and above the grain price. After drying, storing and packaging, truthfully labelled seed is sold to customers at double the market price of grain. The federation members, seed committee and RDRS share the profit on a 20:40:40 basis. All agreed on this profit-sharing mechanism; the share of RDRS goes into the revolving fund and has already enabled RDRS to further expand the model to other federations. The seed innovation system created employment in other sectors. Under their income-generating activities programme, another federation now produces polythene-lined jute bags to package seed, whereas initially bags were purchased from Dhaka. This integration of services strengthens the overall community seed system. The federation model allows for poor people to help one another to improve their livelihood. Define clear responsibilities for quality control and marketing Having a share of the profit isn't a gift; it comes with specific responsibilities. The federation provides physical assets and dries, packages, and stores seed. The seed committee inspects the fields of the seed producers and conducts a seed germination and moisture test after harvest. Once all seed is harvested, they communicate the level of seed production to members of their own and other

250


federations. Demands for the next season are gathered. As opposed to the Grameen seed and Farmseed models, where farmers sell their own truthfully labelled seed or quality seed (see previous chapters), here the seed committee members are in charge of the sales, reducing the transaction costs for the seed producers. Production and marketing planning takes place at the union level. RDRS helps the federation's seed committee to prepare a yearly action plan and a budget, to assess the required seed quantity, and develop a marketing strategy. Expand coordination of seed production and marketing to regional level RDRS can facilitate seed exchange between federations, but to get more actors involved to buffer against sudden seed shortages in the region, a different structure was needed. Therefore, a focal area forum was established for Northwest Bangladesh with representatives from various government institutions, resourcepoor farmers, private sector and NGOs. The forum helps to provide poor farmers access to consistent information, quality seed and rice technologies even after the PETRRA project. The forum improves quality seed demand assessment at the regional level and acts as a decentralised hub. On a seasonal basis, they place their demand for breeder seed of specific varieties to the rice seed network coordinator at BRRI, as described in Chapter 17. Through a series of workshops, the Northwest focal area forum so far approved the varieties BRRI dhan 28 and BRRI dhan 29, and confirmed the RDRS federation seed system as suitable for scaling-up in Northwest Bangladesh. Members decided to rotate the chair periodically and oversee the planning and implementation of rice-based activities. They committed resources to make this happen. Sustainability of the forum was boosted after the endorsement by the state minister for agriculture (see Box 21.1).

KEYS FOR SUCCESS

"If the customer is dissatisfied, that's the end of my business." Vegetable sellers in the street know this all to well. And those who sell quality seed are vulnerable in the same way, presuming they do not have a monopoly in the region. PETRRA has enabled many poor seed producers to emerge across Bangladesh.

Building on community institutions (federations), farmer seed producers and farmer trainers increases efficiency of human, social and physical capital. RDRS is open to new ideas and ready to share this with federations (see Box 20.2). Ownership is ensured by giving the civil society a voice in project decision-making. Creating a conducive learning environment for women gives them more voice in rice seed cultivation and agriculture. 251


Capacity building of federation seed committee members addresses technical and marketing aspects. Training-of-trainers for extensionists and farmer trainers. Incentives for all stakeholders are based on profit-sharing. Revolving funds pay for farmer trainers and covers training costs of seed producers. Revolving funds allows RDRS to cover more federations and crops. Box 20.2 Bagging Seed

"Initially we used polythene bags to package truthfully labelled seed, but the bags got easily perforated by the seed, allowing moisture to seep in, so we now use a polythene-lined jute bag," explains Mr. M.G. Neogi from RDRS. When visiting the GKF seed processing centre and shown the bags used by them, Neogi asks Mr. Jabbar, the manager of the centre, if he can take one as an example: "These bags are better than ours and I will show them to those federations that produce seed. I cannot make any decisions on what type of bags they use, only the federations can, but at least I can introduce it to them."

DIFFICULTIES, RISKS AND ASSUMPTIONS Timing of seed supply by BADC is unpredictable from one season to the other and the seed by BADC is subsidised. "If we sell our seed at a higher price than BADC, they sometimes dump seed in our area and we cannot sell the seed from the federation," informs Neogi, hitting the nail on the head. The subsidised government seed system remains a challenge for these bottom-up innovations. Another risk is related to scaling-up and maintaining quality insurance. The more the federation expands their market to other unions, the further the distance becomes between seed producers and clients, and the higher the need will be to strengthen quality control. Apart from the more technical and marketing issues, mentality changes are required. The partnership between DAE and RDRS (and NGOs in general) is still in its infancy. Over the years, DAE has developed a culture of working on the basis of projects and getting financial resources to do a particular job. Although the project trained their staff as master trainers, DAE still asked the project to pay for using their staff as trainers. Appreciating the opportunity to work in partnership requires a mental shift or perhaps a change in operational procedures endorsed by higher officials. The endorsement by the state minister for agriculture actually made this happen.

SCALING UP From 2000 to 2004, PETRRA funded training activities at federation level and 252


helped to purchase two organic cocoons. Three more cocoons were purchased by a GTZ project. In the scaling-up process, eight others were acquired through revolving funds. As cocoons were too costly, however, two other federations opted to use RDRS revolving funds to transform existing rice storerooms into seed storage facilities. To optimise resources, these are shared among 28 seed-producing federations. By August 2004, revolving funds had expanded the rice seed production to 28 federations; five of them equally embarked on wheat, potato and mustard seed production. During this period the latest high-yielding varieties BRRI dhan 28, 29, 33 and 39 had been extended to more than 14,000 farmers through the federation seed production system, of which 60% were federation members, the others being neighbours. RDRS partnered with BRRI to conduct action research in farmers' fields to validate varieties new to the region or unreleased ones, as described in Chapter 16. As such, the federations started to serve as a forum to quickly gather feedback, narrowing the gap between research and adoption of new varieties. Consequently, RDRS started experimenting with action research on a broader range of rice-related issues and involving students from the Bangladesh Agricultural University, resulting in the signing of a memorandum of understanding between the two institutions for conducting action research jointly over a period of 10 years (see Box 20.3). April 21, 2004. Mr. Imul Mosadek is working in the rice field when we arrive in Neezpara village. Although the sun is out again and temperature is at its peak, he is still wearing his green raincoat and collecting data in his notebook: "I am very pleased to conduct my fieldwork on hybrid rice seed production with Bulbuli. She is highly motivated and stimulates me a lot." To the question why she gets involved in research, she promptly replies: "I only know if something is a good technology when I have seen it with my own eyes. Hybrid rice is said to give much higher yields than modern varieties, and if it works I will adopt it and promote it through the federation. And if I can produce the seed myself that would be even better." Hybrid rice seed production is not an easy technology and new to Bangladesh, but Bulbuli is determined to give it a try.

Box 20.3 Institutionalising Participatory Research under PETRRA

PETRRA emphasised having farmers participate in setting research priorities and developing technologies, as well as conducting research on technology uptake. This approach encouraged poor farmers to adopt research findings, such as transplanting 2-3 seedlings per hill compared to farmer's traditional practice of using 5-6 seedlings per hill. By 2004, several students from the Bangladesh Agricultural University (BAU) had engaged in this new way of doing research, as part of their MSc degree. Eight worked on crop management, eight on social aspects, two on gender and two on livelihoods. Following initial success, the executives of BAU and RDRS signed a memorandum of understanding for implementing a students' internship programme and exchanging staff for a ten-year period.

253


CONCLUSION Innovations at RDRS emerged at various levels and include: a pay-for-service mechanism for training farmer groups; sharing profit on seed sales and reinvesting this through revolving funds; using creative marketing strategies; establishing a focal area forum; and institutionalising action research with two leading governmental research and education institutes. The social capital built up over the years has offered a fertile ground for rice-based innovations to sprout.

REFERENCES AND RECOMMENDED READING Bentley, J., Boa, E., Van Mele, P., Almanza, J., Vasquez, D. and Eguino, S. (2003) Going Public: a new extension method. International Journal of Agricultural Sustainability 1(2), 108-123. RDRS (1999) Operational Framework (Manual) for Management of the Union Federations and Secondary Groups. Rangpur Dinajpur Rural Service, Rangpur, Bangladesh. RDRS (undated) Union Federation Bye-laws. Rangpur Dinajpur Rural Service, Rangpur, Bangladesh. Wallace, I. (1994) Creating learning networks between formal agricultural institutions and rural people: the potential role of local non-governmental organisations as intermediaries. Journal of Agricultural Education and Extension 1(2), 1-14.

254

Part VI: Synthesis

20 People and Pro-Poor Innovation Systems

Paul Van Mele Ahmad Salahuddin Noel P. Magor

"Improving access to technologies and services for resource-poor farmers, men and women, is one thing; considering them as partners, not just as recipients or beneficiaries, yet another."

EXTENSION, COMPLEXITY AND POVERTY Innovations in Rural Extension shows that extension is about working with multiple actors, each with their personal and institutional histories, norms, values and interests. It is about getting the technologies right, improving access to inputs, knowledge and markets within existing policies, and stimulating learning and experimentation. These dimensions, and the level to which they are addressed in an integrated manner, determine the success of extension. Adding a poverty aspect further complicates each of the above mentioned dimensions. The need for farmer participation becomes more stringent when developing and promoting pro-poor technologies and markets. But private businesses, scientists and governmental extension agents often have little or no experience in working with the poor, especially with women. Illiteracy rates are higher among poorer people, their personal networks are less elaborate and transaction costs (which represents time and costs to access information, services, markets and technologies, negotiate contracts, and so on) are comparatively higher compared to better-off farming families. High transaction costs not only affect the poor in getting access to support, but also affect service delivery agents who want to target the poor while developing extension or business models. To add to the complexity of reaching the one billion rural poor in this world, one has to consider the diversity of poverty itself (Berdegué, 2000). The diversity of strategies people use to cope with poverty adds to the need for plurality in extension. 257


Participatory technology development is an integral part of the innovation system and has been addressed as such in the various chapters. Reader-friendly overviews can be found in books by Ashby et al. (2000) and Bentley and Baker (2002). The multi-faceted needs of poor farmers and the multiple demands on their precious time influences our choice of methods for situation analysis, communication and training. This calls not only for diversity in extension mechanisms, but equally for a flexible use of multiple communication and learning tools fine-tuned to the specific client group, and building on the strengths of the range of service providers available in the system. This innovation systems approach not only moves away from the idea of a one-size-fits-all technology, but also of an ideal blue-print extension method (Biggs, 2004). This chapter synthesises lessons learnt from the PETRRA project and ventures into some new areas. We will first describe the influence of policy on people driving the innovation system, followed by a discussion on the dynamic roles that multiple actors play in pro-poor extension and business development, and how actors interact in the 'theatre of agricultural innovation', to quote Röling and Jiggins (1998: 304). We further discuss the concept of transaction cost theory, illustrated with experiences from PETRRA. We then explore some of the promising innovations that emerged, followed by suggestions for future research. From 1999 to 2004, PETRRA inspired partners to innovate not only with technologies, but also with farmer education, communication, organisational and institutional models in delivering pro-poor services and inputs. Innovations in Rural Extension offers us a rich menu for the reader to select their own dish. Ingredients can be replaced, spices added.

PEOPLE MATTER People are the drivers of change. This is true for policy-makers, donors, service providers and clients, and hence justifies having a closer look at the human dimensions, social contexts and organisational cultures of these actors. In what follows, we use the term service in its broadest sense, including advice, training, technologies and anything that brings benefits to the intended target group. Extension policy and public sector In Bangladesh, changes in policy created an enabling environment for innovations to

258

People and Pro-Poor Innovation Systems

emerge in seed systems (see Part V of this volume) and more broadly in the extension system. In what follows, we will give a brief overview of how large projects influenced policy at the Department of Agricultural Extension (DAE), followed by the various ways PETRRA interacted with DAE. During the 1980s the World Bank funded DAE to implement the top-down training and visit (T&V) system of extension. The field extension agents or block supervisors visited mainly better-off farmers and hoped that technologies would spread spontaneously to other layers of the farming community. To trigger desired changes in the T&V model, in 1992 a first DAE reform initiative took place through the Agriculture Support Services Project (ASSP), funded by the World Bank, DFID and the Government of Bangladesh. By 1996, the DAE had a new agricultural extension policy (NAEP), which also embraced the livestock, fisheries and forestry departments (Hassanulah, 2002). DFID further strengthened this initiative through the Agricultural Services Innovations and Reform Project (ASIRP) from 1999 to 2003. Major outputs of this project were a mission statement and a strategic plan to help motivate change in the organisation. "The Department of Agricultural Extension's mission is to provide efficient and effective needs based extension services to all categories of farmer, to enable them to optimise their use of resources, in order to promote sustainable agriculture and socio-economic development." DAE, 1999 With its large bureaucracy and roughly 24,000 staff members, the largest resource of extension staff in the country, the challenges to induce institutional change were enormous. Stakeholders within and beyond DAE felt that donors drove the agenda and pushed for the quick achievement of outputs in some areas, resulting in reduced internal ownership (Pasteur, 2002). In an interview with the New Agriculturalist in April 2000, Donal Brown, a former natural resources advisor for DFID in Bangladesh, confirmed this: "One could try and impose [changes] but, if one imposes, the long-term sustainability of these activities is just not going to happen." PETRRA's modus operandi was guided by principles that stood in contrast to this charge of 'lack of ownership'. From its very inception it nurtured a 'learning by doing' environment. PETRRA developed ideas jointly with their partners through personal or group interactions, and helped them to reflect on their own comparative advantages, their strengths and experiences, as such cultivating local ownership. Many technologies and extension methods developed or fine-tuned under PETRRA became mainstreamed in their respective partner organisations. PETRRA worked with multiple service providers at the field level, while maintaining good links with policy makers and DAE senior management. DAE block supervisors were invited to participate in field activities in most of its sub-projects. This shift from DAE contracting out others to deliver services, as was the case

259


under ASSP and ASIRP, to NGOs asking DAE to partner, indicates a move towards better balanced partnerships and power, as was recommended by World Bank expert Gary Alex (2001). He also indicated that the mechanisms set up under the ASIRP project, although they supported decentralisation and improved extension support to farmers, failed to strengthen research-extension linkages. This shortcoming partly explains why in August 2004 the state minister for agriculture so strongly endorsed the newly established focal area forums that bring representatives of poor farmers, researchers, private sector and intermediaries together (see Box 21.1). Donors and flexibility Creating a learning system requires commitment, flexibility and fundamental changes in norms and values, not only within implementing organisations (Pretty and Chambers, 1994; Röling and Wagemakers, 1998), but equally within the donor community. "For far too long, the heart of development practice has been characterized by an irony which saps the energies and motivations of even the most enthusiastic practitioner: those very institutions that are established to facilitate societal change at one moment, invariably become its next constraint." Bawden, 1994: 258 A project-wise and planned approach with logical frameworks or logframes is often proposed as the most appropriate way to organise innovations and development (Leeuwis, 1995). This philosophy, however, presumes that people proceed based on rationally organised decision-making and learning, which goes at the expense of creativity and scope to respond to new learning and unpredictable change. "Funding agencies of innovation and development activities usually wish to know in advance which goals have been set and how these goals will be realized… thereby the capacity to learn, in intervention processes can be severely hampered." Leeuwis, 1995 "Institutional innovation itself needs to be recognised as an important and valid (if difficult) research subject and output." Dorward et al., 2000 "Some of these [donor programme management systems] will require a long time scale and a process approach, chipping away at problems, and being willing to be opportunistic and flexible." Duncan et al., 2002 The new challenge for donors and implementing agencies alike is to develop mechanisms that allow one to capitalise on the diversity of perspectives, ideas and opportunities that arise when implementing a project. This points us to the principles of change management and organisational learning, which has been present in business literature for decades, but which has only been widely recognised more recently (Easterby-Smith and Araujo, 1999). 260


"The challenge for development agencies is whether they want to provide the incentives to encourage a learning and change culture and incorporate professional people with these skills into their staff and development projects." Biggs and Smith, 2003 Organisational learning at the donor level also requires regular consultation with those implementing the projects and the clients to develop evidence-based policy, while at the same time, for gender for instance, gender-sensitive and genderknowledgeable people in decision-making positions will be needed at both donor and project level. "While there are some positive developments in donors' policies and practice, the key challenge to gender mainstreaming occurs at the implementation stage." Macdonald, 2003 'Strength in diversity' has strong resonance in development circles (Chambers et al., 1989; Hall et al., 2003b; Biggs, 2004), and more recently in donor thinking about rural poverty alleviation (Berdegué and Escobar, 2001; Farrington et al., 2002a). Donors have a large responsibility in stimulating local innovations, but their support to mainly the largest NGOs with heavy management structures may push the development landscape into the other direction. Vertical integration has its limitations, especially for development organisations. Small, flexible and professional NGOs are often ignored despite their ability to quickly respond to emerging local needs and mobilise the poor, irrespective of their membership of microcredit programmes. To unlock the potential of more local actors, donors could support innovation systems research to identify champions, and to unravel their personal, historical and institutional contexts that shaped them. Projects, service providers and potential champions Innovations require more than creative capacity to invent new ideas; they require managerial skills and talent to transform good ideas into practice (Van der Ven et al., 1989 in Ayas, 1995). To this, we would like to add the need for motivation and a long-term vision. "Unlike buying stocks, it is hard work to put ideas into practice. And no one can do everything." Nalebuff and Ayres, 2003: 10 Ways to identify potential champions among project partners and to nurture their commitment deserve equal emphasis to the policies and regulations shaping institutional change. In their report for DFID on drivers of pro-poor change, Duncan et al. (2002) say that reform can be stimulated in two ways: by promoting broader processes of social and economic change (such as education, in particular of women); and through identifying and supporting champions of change (including

261


NGOs, community-based organisations, reform-minded elements of the political parties and of the civil service, the media, the private sector, professional associations, the research community and the Bangladeshi diaspora). The cases presented in this book highlight some of these champions. We believe that committed people are the glue that make partnerships successful and drive institutional change. Understanding the historical context and personal characteristics of those people shaping innovation systems is crucial, yet often ignored. Short-term projects like PETRRA may be criticised for not having changed the institutional context in which scientists work or for not having brought in enough international extension experts. But one could argue that values, once experienced, become part of people's personal history that will remain within the system. All four top management officials from BRRI interviewed and more than 80% of the people involved in PETRRA sub-projects actually improved their knowledge, attitude and practices with regard to value-based, demand-led research (Solaiman et al., 2004). As for bringing in experts, the way in which new ideas are introduced and their modus operandi are at least as important as their actual technical or methodological expertise. Creating local ownership and empowering project staff are key to the sustainability of induced change. Although these are popular contemporary advocacies, they can easily fall to pieces in one's hands. "Project cycle planning and management could be improved a great deal if it was acknowledged that all parts of projects are carried out by people working in social contexts, with all the features of social relationships that are present in human interactions." Biggs and Smith, 2003 Professional pride and personal satisfaction after having worked through a problem with farmers can become major motivational factors for researchers and extensionists alike. But often scientists and governmental extension agents lack the opportunities of getting heart-warming feedback from resource-poor farmers, policy-makers and donors alike. It is with this in mind that PETRRA created an enabling environment for government, non-government and private sectors to experiment and develop or test new technologies and methods with farmers, together. Nurturing a shared hope for change was a prerequisite for PETRRA and its partners to walk the extra mile. "Hope, as an ontological need, demands an anchoring in practice. Hopelessness and despair are both the consequence and the cause of inaction or immobilism." Freire, 2003: 9 A first experience is a lesson for life. Through effective partnerships that build on complementary skills and mutual benefits, the chance of having a rich first experience increases. Besides, professional pride and ownership is shared from the on-set and boosts scaling-up, as witnessed by the video sub-project (Chapter 5) and

262


several others (Solaiman et al., 2004). Innovation systems research, which addresses organisational culture, can help in bringing 'like-minded' organisations or individuals closer together and ensure a higher partnership performance. We believe that winwin situations could occur both more frequently and with better planning. "A more systematic consideration of organisational culture issues within project planning and management is likely to improve the effectiveness of development interventions." Biggs and Smith, 2003 Cultivating local ownership is important, no doubt, but professional pride can also close one's ears for criticism and stifle one's creativity. We also experienced that ownership can lead to protectionism. Occasionally, PETRRA had to intervene to overcome partners' apparent resistance in bringing their innovations into the public domain, as was the case when trying to scale up a new rice-duck farming system. Involving additional partners and shifting responsibilities offered solutions (see Chapter 12). Appropriate incentives and communication mechanisms are important for lifting motivation of staff who are involved in developing innovations to a higher level of organisational pride and ownership. Some sub-projects had inadequate communication between local, regional and national offices and lacked clarity about mandate and decision-making power at each level, as such undermining staff motivation. While we recognise professional pride, personal satisfaction and heart-warming feedback as important incentives for people to engage in participatory research, there also exists the risk of them sticking to their newly acquired comfort zone. We believe that researchers and extensionists need to be stimulated more to continuously challenge their own work, get out of their professional comfort zone pro-actively, and change their culture of non-listening to farmers. Reaching rural women: policy and reality Resource-poor farmers, women in particular, are extremely motivated to receive training in all aspects of agriculture (see Part II on gender). A selection of quotes presented by Orr et al. (2004a) from women and men who participated in PETRRA sub-projects illustrates the impact training has had on gender and livelihoods. "When we used to fail to preserve good quality seeds, husbands used to quarrel with us." "When you are poor, you don't want to consult with your wife or family members." "We are not interested to sharecrop anymore, we want to work with our own agricultural land." But also links to scientists and government extension agents improved: "Before we were afraid of the Rural Development Academy, it is a wellprotected area and big officers may not talk with us. Now we are proud to talk to scientists." "Now block supervisors come to us and even ask us for solutions." But women's involvement in training programmes is still largely determined by donor policies. Reviewing EU and UK development co-operation, Khan (2003)

263


mentions that gender remains a low priority despite policy commitments to the contrary. She suggests donors to increase their collaboration with civil society organisations and to open up their dialogue with multiple actors. Gender studies are useful for analysis of separate household needs, responsibilities and roles, but it has also led to new knowledge being compartmentalised. Perceived wisdom often reinforces existing gender discrimination in access to information: "More attention must be paid to traditional women's roles, such as post-harvest activities and livestock care, as well as to new off-farm livelihood activities." Gill et al., 2004 This stands in sharp contrast to PETRRA's findings, which promote women to get training on all aspects of agriculture, also on those areas where they do not necessarily do the work. Limiting women's training to their traditional roles excludes them from household decision-making about agriculture and inhibits empowerment. "Women can be empowered by giving them equal access as men in training and extension programmes." Hossain et al., 2004 Under PETRRA, resource-poor women, once trained, emerged as strong advocates. In some cases also women solidarity was a driving force for female farmer extension agents to establish new groups in new villages and promote low-cost agricultural technologies. Tools for identifying these champions among the rural poor as important actors in the innovation system, not just as beneficiaries, need further attention. The transaction cost theory offers additional insights into forces that shape the innovation system, and into how access to technologies, services and markets can be improved for the poor.

TRANSACTION COSTS: BRINGING PEOPLE INTO ECONOMICS In 1985, Williamson articulated the evolution of modern institutions as a key contributor to the theory of new institutional economics (NIE), which tries to apply economics in the real world where people and organisations engage in both transformation (production) and transaction (contracting and exchange) activities. As people are given a more central position, the theory borrows liberally from social science disciplines. More recently, it is finding its way into development and rural extension literature (Dorward et al., 2000; Morrison et al., 2000; Farrington et al., 2002a). The seeds of awareness and practice can be seen in the concept of institutional intermediation as used by the Bangladesh Rural Advancement Committee (BRAC) for developing a pro-poor poultry business model that reduced linkage weaknesses (Lovell, 1992). Similarly, in Magor (1996) the enterprise web

264


adds understanding as to why some promising technologies failed to be extended. Political economy aspects of research and development, as addressed by Biggs (1978, 1992) and Biggs and Farrington (1990), also relate to new institutional economics. Transaction costs in pro-poor service delivery The concept of transaction cost theory, which is central to new institutional economics, is described in Box 11.2. We do not claim to be specialists in new institutional economics, but in this book we have tried to use transaction cost theory as a pragmatic tool for analysing uptake pathway models for specific technologies. Below, we expand this line of thinking to a more generic level, namely to innovation systems and how transaction costs are affected when bringing in a pro-poor agenda. It is important to note that transaction costs only mean anything in the social system in which they are analysed: the purpose of the system determines what is defined as a good transaction cost minimisation and what is not (Biggs, personal communication). For example, the highly socially differentiated agrarian sector of Bihar, India, used different modes of transactions in the labour, land and credit markets that were very transaction cost efficient. However, that was in a social system that maintained poverty and social exclusion over time (Biggs, 1978).

Establishing contacts and capacity Service providers require contacts with multiple institutes and farmer groups, and they need time to source information or technologies and fine-tune them to the needs of their clients. In some cases they may also require training to upgrade their skills in order to perform. Basically, all these make up the transaction costs that by and large determine whether an actor will embark on providing a certain service to their clients or not; or whether a partnership or a network will be established to fulfil specific tasks. The PETRRA sub-projects described in this book illustrate the underlying principles. To reduce transaction costs for pro-poor agricultural development, PETRRA facilitated the establishment of networks and partnerships between scientists, NGOs, and private sector entrepreneurs, from technology development and validation, all the way to promotion and developing communication materials. Establishing initial contacts between actors requires a facilitator and in the case of pro-poor agricultural development initial public or private investment is needed. Experimentation with institutional and organisational innovations does not happen spontaneously, nor does it happen overnight. The majority of the twenty subprojects on uptake and extension were led by NGOs; seven established partnerships with community-based organisations. Once capacity is built, other incentives take over. NGOs and private entrepreneurs were trained by national and international experts, a major motivational factor at the

265


early stages. But as the sub-projects gained more experience of working with poor farmers, NGOs started to realise how well-suited agriculture was for poverty alleviation and for integrating it with their other on-going programmes. After the Rangpur Dinajpur Rural Service (RDRS) embarked on seed production, other federations started producing polythene-lined jut bags as part of their incomegenerating activities (Chapter 20).

Sourcing and validating information Sourcing and validating information may be a major constraint for service providers to embark on new areas. Bangladesh counts thousands of NGOs, yet only a few have agricultural expertise. Lack of technical capacity and information being a global issue for NGOs (IIRR, 1999), what would motivate them to engage in agriculture, where could they get relevant agricultural advice and technologies, and at what cost? Under PETRRA, NGOs established technical links with government research and extension, and learnt to commit financial resources for tapping into this expertise. Links were established through the uptake forum, focal area forums or specific partnerships. The focal area forums provide a mechanism for a wide range of actors to get continuous access to sources of technical expertise and streamline their efforts in validating technologies for local suitability, feasibility and acceptability (see Box 21.1). This breakthrough not only helps to optimise use of human and financial resources between actors from the government, NGOs and private sector, it also allows for local innovations to enter the formal research, extension and education systems. The Bangladesh Agricultural University and the NGO RDRS signed a memorandum of understanding for students to conduct action research with poor farmers. The role of NGOs as intermediaries between formal educational institutions and the rural people remains an area of great potential for developing innovations (Wallace, 1994), and may help to institutionalise participatory approaches in higher education.

Establishing farmer groups Working with the poor may initially increase costs. The case on the aromatic rice value chain (Chapter 14) raises the issue of cost-effectiveness in establishing producer groups for domestic and export markets. There is an extra cost involved in organising a larger number of poor farmers compared to working with a few welloff farmers, but in reality this approach has the potential to create a substantial volume of rural employment (Farrington et al., 2002a). Public fund allocation strategies need to take these implications for the labour market into account. Working with the poor also allows for economies of scale, especially when NGOs are involved to coordinate their members or those of local NGOs and communitybased organisations (see Chapters 18-20).

266


But organisations that have the necessary skills and contacts, such as commercial businesses, researchers and governmental extension officers, often lack the knowledge or confidence to facilitate the establishment of groups of poor farmers, despite them constituting the largest part of the farming community. So, one of the questions that arise is: Do I organise farmer groups myself or establish a strategic partnership with someone else who can facilitate this more efficiently than me? In communities where various groups already exist, building on these rather than establishing new ones limits transaction costs. Vertical integration or strategic networks For establishing contacts, building capacity, sourcing and validating information, and establishing farmer groups, an actor can opt to go solo or link up with others. The governmental research and extension institutes along with many NGOs are generally used to doing things solo; only since the reshuffle of public funds, the need has arisen to start thinking about strategic partnerships. Also, international companies operating in developing countries often lack the usual infrastructure and support system: market intelligence, manufacturing capabilities, or distribution channels. So, they have much to gain from tapping into local networks and local knowledge (WBCSD, 2004). A strategic network is a way to lower transaction costs without having to vertically integrate (Jarillo, 1995). As the different partners remain independent, there is more flexibility, but also the need for building trust becomes more prevalent. Under PETRRA, especially the smaller NGOs chose to build strategic networks with local organisations. By having worked with over 150 local NGOs and community-based organisations in various projects, the Agricultural Advisory Society (AAS) has filtered out the 'opportunist feeders'. They now have a rich source of sincere organisational relationships on which they can build, as and when they see fit. Strategic networks may be formed based on economic considerations, although organisational history and personal contacts often play an equally important role in selecting partners, as indicated by several case studies in this book. It is our experience that the success of a partnership between NGOs and community-based organisations, for instance, is determined by the size, history and organisational culture of the partners, along with the influence sphere of the individuals leading the partnership. Partnerships are dynamic and context-specific: multiple scenarios are possible depending on the diversity and density of service providers, their intrinsic strengths, the type of service to be delivered, the intended client group, and so on.

DIVERSITY OF SERVICE PROVIDERS We use the term diversity to indicate both the number of different actors and their relative abundance or density in a given area. These dimensions affect choice and

267


quality of service delivery, from the community level up to the national and international level. Multiple actors: competing or complementing? Who has what role to play in developing pro-poor technologies and establishing pro-poor markets? We will address how different actors may compete with or complement one another in the delivery of quality seed supply, complex technologies, and training and advice. These key 'commodities' will help to clarify the need for diversity in service providers.

Seed suppliers According to Tripp and Pal (2000) plenty of private sector seed enterprises have emerged in developing countries, but there are few examples of those embarking on public crop varieties, such as self-pollinated rice and wheat that are not hybrids. Also, NGO and private seed enterprises not only compete with public sector seed providers and farm-saved seed, but also between themselves (Almekinders and Louwaars, 1999). Despite this, a number of innovations in the rice seed system emerged under PETRRA. As the rice seed market in Bangladesh is far from saturated and poor farmers are eager to get access to quality seed, we anticipate that more competition will strengthen self-imposed quality control mechanisms (see Part V of this volume). Most of the private seed enterprises in India offer few economies of scale, but high economies of scope as they can expand into other seed crops (Tripp and Pal, 2001). This may be only partly true for Bangladesh. Some NGOs embarked on wheat, mustard, potato and onion seed production, after having learnt about rice seed production. But as the rice seed market is far from saturated in Bangladesh, economies of scale are still possible. Syngenta started producing rice seed in the late 1990s, and is gradually increasing their production while they gather experience and explore the market. But the incentives are not merely economic. While for seed-producing agribusinesses it offers an option to diversify their income and strengthen their customer base, for poor farmer seed entrepreneurs it more often is an end to a means. Rice seed production offers a pathway out of poverty and a pathway into community respect. "I no longer have to buy, but can actually sell seed," says Shamima Akhter during a village fair in Kishoreganj, "My husband, mother-in-law and neighbours respect me much more now." In brief, governmental organisations, agribusinesses, NGOs and farmers each have a role to play in the production and supply of quality seed. While most actors reach their clientele through an existing distribution network, small-scale farmer seed entrepreneurs diversify the outlets for seed in the community exponentially. By 2004, the awareness of quality seed was still growing, leading to increased demands.

268


Public funds are especially required in the initial phase to build capacity among NGOs and small-scale, private seed entrepreneurs. But once capacity is built, one should be able to produce and trade seed on a full commercial basis, in absence of market distorting policies.

Suppliers of complex technologies Do the same principles in developing pro-poor seed businesses hold for the dissemination of a complex technology, a new farming system or a value chain for exporting aromatic rice? Often these innovations are non-existent at the time of intervention, resulting in a higher perceived potential to position oneself in these new markets. But these innovations are intrinsically complex; as more side conditions need to be fulfilled, initiatives by individuals or small-scale enterprises are less likely to take place. To disseminate or establish complex technologies, also larger organisations or businesses need to make crucial decisions on opting for vertical integration or strategic partnerships, on addressing all required activities themselves or outsourcing some. Partners are selected based on their competitive strength, interest in participation and for a variety of motivational and personal reasons. Establishing contacts and trust between the various actors is part of the initial transaction costs. To help organisations in this decision-making process, Magor introduced the enterprise web as an analytical tool for strategic planning (see Part IV). Irrespective of the level of market integration, public funds are likely to be required to help disseminate complex pro-poor technologies. Once networks and necessary conditions are fulfilled, the system should be self-sustaining.

Suppliers of training and advice In a synthesis of a six-country study on extension, Farrington et al. (2002b) recommended to create and support opportunities for the poor, not just as producers and labourers, but also as consumers. However, they fail to acknowledge the active role poor farmers can play in delivering services, advice and technologies, themselves. Pioneering large businesses already started to blend social and financial values under the umbrella of corporate social responsibility, and involve the poor in their markets, as customers and entrepreneurs (WBCSD, 2004). Training the poor is considered a necessary investment. Cases presented in this book support the need to consider poor farmers as partners, not just as recipients or beneficiaries. Giving them the opportunity to play a role in delivering services themselves, as a means to social and economic empowerment, opens up a whole new debate on public fund allocation. Several interesting concepts and experiences have been presented recently (Katz, 2002; Rivera and Zijp, 2002; Scheuermeier, 2003). Rather than channelling money through service providers, for

269


instance, public funds could be assigned to farmer groups who then decide how to best use it. This book gives examples of poor farmers, men and women, taking on the role of seed producers, sellers and marketing agents, but also of extension agents. Once trained, they quite easily established new groups of poor farmers in other villages and taught them about rice and seed production, as well as soil fertility management. RDRS federations started to use communicative female farmers as resource persons to train other groups, paying them Tk 50 (US$ 0.90) per session. Public funds allocation is needed in well-integrated areas for the delivery of services related to health, safety and the environment, whereas substantial support will remain crucial for agricultural extension related to subsistence crop and for those areas where access to information, advice and markets is weak (Farrington et al., 2002a and b). Local government One would be tempted to think that local governments are more aware of people's needs, constraints and opportunities, and should be better able to respond to these than the central government. Although they can play a significant role in community initiatives for agricultural development, local governments are not a necessary or sufficient condition (Tendler, 1997; Bentley and Boa, 2004). So far, in Bangladesh social development organisations have been much more pro-active in involving local government than actors working in the field of agriculture have been. Shifts may gradually occur with some NGOs (re)discovering the importance of agriculture in rural development. Although clear benefits could be reaped, most PETRRA partners did not establish links with the local government or Union Parishad. For the rice-duck and mobile pump sub-projects, both complex enterprises with clear impacts on the wider community, local government support was a prerequisite. Also, the Rural Development Academy (RDA) in Bogra built strong links with the Union Parishad, whose chairman was well known to the deputy director agriculture at RDA. The latter involved the local government in organising various awareness and scaling-up activities under the Seed Health Improvement sub-project. Good human relationships are the corner stone for a successful collaboration. Learning networks and forums Arising from the need to provide farmers with consistent information, PETRRA sub-projects started to interact more at the regional level among themselves and with other projects, NGOs, farmer representatives, governmental organisations and commercial businesses with an interest in rice. Two focal area forums, namely in the Northeast and the Northwest, emerged as multiple actor platforms for: (i) channelling the voice of poor farmers, men and 270


women; (ii) establishing a network that facilitates quality control and dissemination of quality seed in the region; (iii) screening, validating and transforming information into consistent advice, as well as for (iv) pooling resources in training farmers (Samsuzzaman and Mazid, 2004). The focal area forums are a practical example of decentralised decision-making in agricultural research and extension. Currently, two farmers are members of the Northwest focal area forum, along with representatives of governmental institutes (BRRI, BARI, BINA and BADC), NGOs (RDRS and GKF), the government extensions service (DAE) and private companies. Mrs. Bulbuli Rani, vice chair of one of the RDRS federations, was elected as a farmer representative. Over the years, she has established contacts with multiple organisations and interacted in action research with scientists and university students. She now critically assesses new technologies in her own field, helps to coordinate seed production at the community level and set up a small tailor workshop at her house, where she teaches young ladies from the neighbourhood. RDRS, stimulated by PETRRA, started the initiative for a focal area forum in August 2002. Although they asked the DAE numerous times to sign a memorandum of understanding to become a formal member of the Northwest focal area forum, their initial reaction was one of reservation and hesitation. Several events brought the various actors closer together, but it wasn't until the state minister for agriculture provided his support during a policy dialogue in 2004 that the director-general of DAE came on board (Box 21.1). Immediately after, the DAE block supervisors were asked to collect the meeting times of all RDRS federations, and received instructions to meet the farmer groups at times that these already gather for other activities. As such, DAE saves considerable time by not having to organise group meetings, and farmers save time by having to interrupt their schedule only once a week. "I could not believe that the minister would accept the idea so strongly; he was brilliant," said Dr. Syed Samsuzzaman, one of the focal area forum initiators from RDRS, immediately after the policy dialogue. "Honouring an agreement is a strong motivator to behave in the collective interest." Kerr and Kaufman-Gilliland, 1994 in Röling, 1996 Clearly the new agricultural extension policy enabled an initiative like the focal area forum to emerge and crystallise, but until this event, the country lacked good examples of how partnerships and decentralisation in research and extension could take shape on the ground. The focal area forums became a reality, and the endorsement by the minister a historic event, probably as significant as the establishment of the new agricultural extension policy itself. Mechanisms of costsharing were discussed from the early on-set and ensured that this platform got a life-span that transcended the PETRRA project.

271


Box 21.1 Minister Endorses Focal Area Forums

August 3, 2004 was a great day for all Northwest focal area forum members as the state minister for agriculture and other distinguished guests participated in a policy dialogue with them and other agricultural players in the region. The top decision-makers not only expressed their appreciation of the concept, activities and progress made, the minister also instructed all to immediately take necessary action to formalise the forum and replicate it. seholds with 3-8 months food security from own rice production, with some flexibility depending on region, actor and technology "Congratulations to the organisers who have invited me to such an enthusiastic meeting. I had been thinking over this issue for a long time. My experience with farmers was that there is a gap between scientists and farmers; I failed to see hope. But today I see some light and hope for the first time that it can work. The focal area forum concept has come to us as a big opportunity and the director-general DAE should go for signing a memorandum of understanding involving all relevant DAE offices. We should try to replicate it all over Bangladesh." Mr. Mirza Fakhrul Islam Alamgir M.P., state minister for agriculture "Advice of the minister is very vital to sign a memorandum of understanding with relevant partners in the focal area forum. We will start the revolution. We start with rice but will expand to various other crops. PETRRA has made a revolutionary contribution to all this. … Focal area forum activities are like a one-stop service. … We have reduced the gap between the different actors." Mr. Tariq Hassan, director-general DAE "We talked about research linkage and its importance. The focal area forum showed the pathways as to how it can be done. … Coordination, capital and credit can play a very important role. … The focal area forum is a model that can be replicated all over Bangladesh if encouraged and supported by the government." Dr. A. R. Gomosta, director research, BRRI "We started with rice because it is very important and it still needs continued development. The Focal area concept very much matches with the new agricultural extension policy. Within the focal area forum we are not only governmental organisations and NGOs, but also private sector. We are supporting farmer groups organised by RDRS. We are doing it in addition to, but not hampering our regular programme, rather strengthening it." Mr. Elias Hossain, additional director DAE, Rangpur (Salahuddin, 2004)

Extension services can tap into multiple resources of actors, methods and tools. Under PETRRA a vast range of methods and tools were developed, tested and validated in order to make the learning environment more accommodating for the poor, women in particular. Each of the examples, or elements out of them, can be used by any service provider depending on the situational context, as such adding further to the desired diversity in extension and pro-poor business development. 272


MULTIPLE EXTENSION AND LEARNING METHODS Which extension method is the best and which one do we promote, is a question often asked. But does it make sense to promote a single method? By proceeding under the perspective of the 'pipeline' model of linear transfer of methods (in analogy to the concept of linear transfer of technologies), many opportunities to reduce poverty in a cost-effective way are missed (Biggs, 2004). The scope for local actors to innovate with extension methods and institutional models is reduced from the very beginning. Clearly new thinking is required as to how to reach more people more quickly (IIRR, 2000). Promoting diversity and cross-fertilisation between various extension, farmer education and organisational development methods point the way ahead (Hagmann et al., 1998; Braun et al., 2000; Van Mele and Braun, 2004; this volume). A lesson for donors and decision-makers, therefore, is to avoid endorsing extension monocultures. Techniques from anthropology and other social sciences allow us to prioritise problems of communities (or groups within), learn about areas where new knowledge is likely to result in innovations, as well as what opportunities exist to build learning methods into existing organisational structures. There is no single extension method that reaches all farmers, neither is there a service delivery system that works under all conditions. In what follows, we will first discuss how transaction costs influence poor farmers' access to information and education. We then consider farmer field schools (FFS) as one of the main innovations in farmer education, followed by a range of other methods and tools tested under PETRRA, and which we believe have great potential to complement farmer field schools. Transaction costs in receiving extension services Under PETRRA, a number of transaction cost reducing innovations emerged in terms of capacity building and awareness raising. Demonstration plots were no longer in the fields of better-off farmers, but in poor farmers' fields. Women received training in their courtyard or in buildings of community institutions, rather than having residential training sessions (see Part II). Through partnerships with community-based organisations poor farmers, men and women, easily engaged as group coordinators and in some cases as extension agents in their own and neighbouring villages. For access to technologies such as seed, transaction costs for the clients is lowest when the retailers are actually farmers within their own community. To remain workable, these systems rely on regular supply of foundation seed, resolved by the rice seed network, NGOs and strategic networks with community-based organisations (see Part V). The potential benefits of decentralising a system lie in the strengths of its local institutions. 273


Bringing multiple services together through carefully identified local champions significantly reduces transaction costs for the poor. This concept builds on the onestop shop. The idea is definitely not to go for one model, one method or one service provider, but to offer multiple services in the same person, locality or facility. To give some examples, the NGO AAS trained resource-poor farmers to become seed producers, so people in the community know where to get good quality seed. But AAS also trained the same people as village soil fertility management experts, as such bringing multiple services together in the same persons (Chapter 8). Going Public, to interact with people where they already gather such as in market places, or linking agricultural extension to traditional entertainment brings multiple services together in the same locality (Chapters 9 and 10). The NGO Shushilan, on the other hand, uses their facility as a one-stop shop. They sell quality seed, trustworthy fertilisers, vaccines for livestock among other inputs, while farmers can also bring in samples of soil, water or diseased plants and consult the small library and field workers at their Agricultural Service Centre. When we asked the librarian, Ms. Suriya Sultana, how she would like to see her small rural library evolve, she mentioned that pictorial children books would be a good addition, as women tend to bring their children when visiting the library. Under PETRRA, many of the NGOs moved towards inclusion of agricultural programmes. RDRS organised weekly training sessions on rice-potato-rice cropping for groups of women in the village. But these were generally held the day after they had gathered for their credit programme. As women already meet on a weekly basis, this opportunity could be grasped to identify their interests and needs, and give them access to other services, be it public health, information or markets. Apart from reducing transaction costs by improving access to information and technologies, service providers need to assess the critical amounts of information that farmers need in order to trigger local innovation. By capturing this critical amount in farmer education programmes, impact can be realised more efficiently and at lower cost. Learning from farmer field schools The idea to replace recommendations with education based on experiential learning has brought about a major paradigm shift in extension (Kenmore et al., 1987; Röling and Pretty, 1997; Röling and Wagemakers, 1998), with farmer field schools being 274


one of the best documented examples (see Box 21.2). The farmer field school uses experiential learning to improve farmers' agroecological knowledge, as well as their experimentation and decision-making skills (van de Fliert, 1993; Gallagher, 2003; Winarto, 2004). A field school usually comprises season-long regular group meetings with a set pattern of activities. This includes agroecosystem analysis whereby farmers visit their field on a regular basis, observe the crop, its pests, natural enemies and environment, after which they return and draw what they just observed on a large poster paper. The whole exercise involves measurement, analysis, peer review and experimentation. But a field school also involves presentations and special topics along with group building activities. Farmer field schools, which were initially developed to tackle the brown plant hopper problem in rice in the late 1980s, are now promoted in various agricultural, fisheries, livestock and forestry programmes. For examples see LEISA Magazine March 2003 at www.leisa.info.

Box 21.2 Farmer Field Schools at a Glance

In rice-based cropping systems, the immediate benefits of field schools continue to be closely linked to the use of inputs, especially insecticides. According to Bartlett (2004), farmer field schools are not designed for rural families with no access to land, and there are fewer immediate benefits for poor farmers who have not been using high levels of purchased inputs. Also, women from poor households often sell their labour and find it difficult to participate in regular training sessions, whereas the better the economic position in society, the stricter the form of purdah or seclusion that women in Bangladesh practice (Banu and Bode, 2002). They may avoid contact with men with whom they have no direct kinship relation or simply avoid public places altogether. What scope is there to strengthen cultural and social sensitivity of extension methods? We believe that the shift to learner-centred approaches in extension is one of the better evolutions over the last two decades, but at the same time we want to pose a challenge: that farmer field schools and other learner-centred approaches should be promoted as part of a broader framework of farmer and community development, complemented by other methods, and based on local institutional strengths. Rice farmer field schools have been developed longest, yet we still lack evidence of them experimenting with or being complemented by small or mass media to reach those millions of farmers that haven't been lucky to be part of a field school. How to reach more farmers with quality education remains an issue (Heong et al., 1998). Bangladesh has roughly 12 million farm families of which 9.4 million are small farm holdings with less than one hectare (BBS, 2004). By 2001 and under various projects, the Department of Agricultural Extension established 6,200 farmer field schools across Bangladesh; roughly 157,000 farmers received direct training in integrated pest management (IPM) in rice (see www.communityipm.org). Even if all rice farmer field school efforts undertaken over the past 15 years across the world were to have been concentrated in Bangladesh, still only two million farmers would have been reached. 275


Even if one takes an optimistic view of how farmers might use their field school education to offer this to other farmers, and to develop community-based organisations that undertake progressively more ambitious self-directed development, the impact would not meet the need. Are farmer field schools designed for and suited to become a mass education approach? If not, how might we redefine their role? Barzman and Desilles (2002) pointed to an excessive preoccupation of their farmer field school programme to train a certain number of farmers annually, under pressure of donors, and at the expense of quality of the learning process. This was confirmed later by a report for CARE Bangladesh by Andrew Bartlett (2004) who mentioned that in the scaling-up, project staff became stuck in delivery mode. In an early review of farmer field schools in Asia, van de Fliert (1993) mentioned that training quality and intensity deteriorated as the programme scaled up. So how can quality be maintained in farmer education methods when going to scale? We consider the key objective and strength of farmer field schools to lie in its focus on learning, not on reaching large numbers. Understanding ecological relationships, and changing learning and experimental behaviour can be achieved through participatory learning approaches, such as farmer field schools, which offer great opportunities to develop, validate and select the most relevant learning exercises that trigger experimentation and innovation. But additional value could be obtained, once these methods, exercises and materials are developed, if they were incorporated into other learner-centred methods such as video, entertainmenteducation, or mass media (see Part III on learning with rural communities), used by champions positioned in organisations outside the field school. Other methods: going to scale An overarching factor stimulating creativity was the competitive tender mechanism that PETRRA used in approving sub-projects. As none of the extension methods were imposed, but built on the organisations' strengths and philosophies, most innovations became mainstreamed in the respective organisations that researched them. Ownership was cultivated through a learning by doing culture and a flexible management system. By adding a certain element of competition and stimulating cross-fertilisation between methods, PETRRA speeded up the innovation processes. During regular uptake forum meetings, each partner had to present their methods to other subprojects, DAE staff and other non-participating NGOs. A knowledge, attitude and practice study revealed that out of 27 sub-projects covered by the study, findings of 21 were used by a wide range of governmental and non-governmental organisations (Solaiman et al., 2004). Folk songs, for instance, were readily taken up by other organisations, indicating that extension methods should not only be appropriate and attractive to the client group, but equally to those implementing it.

276


External reviewers asked us to make comparisons between methods, but this would mean taking methods out of their context: any method may have a high or a low impact, be cost-effective or not, depending on those implementing the method, the learning content, and the characteristics of the clients and communities. Nevertheless, we have tried to extract some generic characteristics in terms of investment requirement and anticipated outputs (see Table 21.1). A service provider who wants to try out any of these methods could use this as a decision-making tool. Table 21.1 Qualitative assessment of extension methods INPUT

METHOD

OUTPUT

FACILITATION SKILLS

MONEY

TIME TO ORGANISE

HUMAN CAPITAL

SOCIAL CAPITAL

High

High

High

High

High

Farmer-to-farmer extension

Medium

Medium

High

Medium

Medium

Video-supported learning

Low

Medium

Medium

High

Low1

Medium

Low

Low

Medium

Low

Low

Medium

Medium

Medium

Medium

Farmer field schools

Going Public Entertainmenteducation2 1

Will be high if objective of video is social mobilisation. 2Can be live shows or programmes on radio or TV.

Women-led extension approach A recent FAO survey showed that female farmers receive only five percent of all agricultural extension services worldwide and that only 15% of the world's extension agents are women (FAO, 2004). Women farmers in Nigeria were more satisfied with the quality of the services delivered by female than by male extension agents (Lahai et al., 2000). That more women should be recruited by service providers is well known, but often social, cultural or institutional barriers have hampered this. Under PETRRA, trained village women who displayed a high level of solidarity and commitment became extension agents; they organised events in their neighbouring villages once a month (Chapter 3). Working through communitybased organisations and having flexible employment formulas for village women extension agents may help to reduce the gender imbalance in extension services and increase women's access to extension.

Family approach in training This approach showed that training husbands and wives together (with or without children) improved intra-household decision-making and community respect 277


(Chapter 4). The family approach helps to reduce social and cultural barriers; it enables women to get access to services delivered by outsiders more easily. The method has been pioneered in Bangladesh by CYMMIT for wheat post-harvest (Meisner et al., 2004), and has been expanded under PETRRA by giving women access to information about all agricultural topics.

Farmer-to-farmer extension Farmer-to-farmer extension can be very powerful, especially when linked to experiential learning and participatory rural appraisal (PRA) techniques, such as village soil fertility maps (Chapter 8). Building a vast network of local NGOs and community-based organisations may lead to a more efficient use of social capital, and allow a more coordinated approach in training farmer extension agents.

Video-supported learning Unexpectedly, comparative analysis revealed that women learnt more things from meticulously designed videos on post-harvest technologies, including insect and disease management, than from farmer-to-farmer extension (Chapter 7). The videos resulted in higher levels of experimentation and adoption of new technologies. Over the last couple of decades we have seen many changes in the use and role of media for communication in development (Norrish, 1998). The potential of using video within the framework of an interactive dialogue, and still having a video product at the end is an enormous advantage when it comes to scaling-up. Especially for quality maintenance of methods like farmer field schools, videos can add tremendous value, as the messages can be carefully engineered and remain the same. The instant playback feature of video enables continuous participation and immediate feedback. Besides, images have a high credibility and can easily motivate people (Dagron, 2001). Apart from the potential of video adding value to farmer field schools, field-based experiential learning methods can also provide useful inputs for making mass media farmer education programmes (see also Chapter 6).

Going Public A method whereby extensionists or scientists go to public places, such as markets, to interact with farmers was developed earlier by CABI Bioscience in another project in Bolivia (Bentley et al., 2003), and tested in Bangladesh with AAS, BRRI and the Rural Development Academy at Bogra. To address the criticism that farmer field school graduates hardly share their learning with the wider community, as was the case in the Philippines (Rola et al., 2002), Going Public offers one of the possible solutions. Van Mele and Zakaria (2004) invited trained farmers to man a stand at a weekly hat or market and to share their 278


newly acquired seed health expertise with interested visitors. And because women in Bangladesh remain mainly confined to their homestead, Going Public was further modified to reach more women by going to the uthan or courtyard (Chapter 9).

Entertainment-education When Shushilan embarked on their sub-project to test improved seed uptake pathways, they started with the more familiar field demonstration days. Through the regular uptake forum meetings, organised by PETRRA, they started to discover their own organisational strengths. The picture songs emerged as a jewel in the crown. With their vast experience in using traditional media such as drama and songs, Shushilan developed a new cultural programme with agricultural messages. Music, lyrics and paintings all came nicely together in the picture songs, which turned out to be a culturally appropriate way to reach large numbers of rural women (Chapter 10). Traditional media, such as folk songs, drama and puppet shows were, for instance, proposed in Sri Lanka to complement group training in integrated pest management (IPM) (van de Fliert and Matteson, 1989). But FAO perceived multimedia strategic extension campaigns as only suitable for awareness raising. Soon afterwards, they piloted farmer field schools in Indonesia and since this was perceived a more effective approach to promote IPM, it replaced all other IPM extension approaches in Sri Lanka. Only in the mid 1990s, and under supervision of Dr. KL Heong from IRRI, entertainment-education was successfully applied to address pesticide misuse by Vietnamese rice farmers. Because farmers depend on local radio broadcasts as their primary source of information, the researchers placed the farmers' ever-present radios at the heart of a media campaign. "We got a group of actors to play out a series of brief comedies, relating solid scientific facts through rustic situations to make the audience laugh," Dr. Heong explained. "We found these simple, humorous messages fixed themselves in the minds of thousands of farmers." Entertainment-education refers to "the process of purposely designing and implementing a media message to both entertain and educate, in order to increase audience knowledge about an educational issue, create favourable attitudes, and change overt behaviour" (Singhal and Rogers, 2003). Considering that entertainment-education is a major approach used to trigger behavioural change on public health issues, it is quite remarkable how little it is used in agricultural development. A quick search on the internet yielded 94,300 results for entertainment-education and health, compared to only 4,130 when combined with agriculture, indicating the huge potential for agricultural extension to draw from cross-sector experiences. Likewise, experiences presented in this book may have practical applications for fisheries, forestry, public health and other sectors.

279


Primary school and college education Reaching farmers through their children is a very powerful extension approach (Nathaniels, 1998; CIP-UPWARD, 2003; Arnst et al., 2004). Although several national NGOs in Bangladesh (e.g. BRAC, PROSHIKA, FIVDB and RDRS) develop their own non-formal education curricula and learning tools, integrating these with their agricultural development programmes has so far remained underexplored. NGOs in Nepal, such as the Centre for Agro-Ecology and Development (CAED), have been using schools for years with very promising outcomes. In Table 21.2 we present the potential circumstances in which each of the methods described can be used. As stated earlier, these offer some broad guidance only. Making small modifications of education curricula can be a real challenge, even if the environment seems conducive at first. In 2002, Van Mele tried to introduce some of the seed health exercises in the non-formal primary education programme of the NGO BRAC, the largest non-formal education system in the world (Mednick, 2004). With a group of women teachers the curriculum for biology Table 21.2 Suggested use of extension methods

280

METHOD

CIRCUMSTANCES UNDER WHICH METHOD MAY BE USED

Women-led group extension

Requires communities where a certain critical mass of social capital is already in place

Family approach in training

Is applicable for any community, irrespective of the level of social capital

Farmer-to-farmer extension

Requires solid organisational support for it to be effective and will work best if implemented alongside other rural development activities

Farmer field schools

Requires skilled facilitators and high initial investment cost. Ideally used in pilot phases to develop and test learning tools that can be incorporated in all other methods. Principles and processes could be built into curriculum of wide range of service providers

Video-supported learning

Requires multidisciplinary approach in developing scripts. Adds value to any other method. Can be effective to educate farmers in remote areas without the need for well-trained facilitators. May need adjustment to fit regional or local culture

Going Public

Can be tried by any service provider with little preparation. Lends itself well to reach people in remote areas where general organisational support may be weak

Entertainment-education

Requires multidisciplinary approach in developing scripts. If no use is made of radio or TV, the method is limited to areas where live performers operate. Highly appropriate to reach rural women

Primary school and college education

Requires flexibility of education system and teachers' corps to include processes and tools of farmer field schools, or to organise video or agricultural entertainment shows. Children welcome this as a shift from sterile teaching methods in most rural areas


classes was assessed: as sowing seed, observing plant growth, and weekly drawing was already part of their curriculum, only minor modifications were needed. Children would bring a small amount of rice seed from their home, manually clean it in the classroom, and sow the spotted and irregular seeds in a separate pot from the healthy seeds. Within one season and after having trained the teachers, more than 2,000 children in 70 schools learnt about rice seed health. Children brought the message back home and stimulated parents to test the importance of seed health for themselves. The monthly parents meetings revealed an increased awareness and improved practice. The experience looked promising: as BRAC operates thousands of schools across the country and educates mainly girls, the potential was enormous. But it was never mainstreamed, illustrating one of the challenges of vertical scaling-up (between programmes) within large rural development organisations. Multiple learning tools Sometimes discussions arose in meetings between PETRRA sub-projects as to whether video is a tool or a method. Basically, the process of developing a video can be empowering in terms of the experiential learning that occurs among those involved in its development; the end-product, the video tape or DVD, is the tool that can be used to share information contained in the video with many others. The method deals with how this end-product is used in training and determines the quality of the learning that takes place among those watching it. Building a common understanding of terminologies used in extension method research was one of the hurdles PETRRA had to tackle during various uptake forum meetings. All its 20 sub-projects on uptake and extension involved partnerships with NGOs and various other actors. Seven of them were led by researchers, the others by NGOs mostly in partnership with researchers and DAE staff. Especially those sub-projects that were approved at the earlier stages of PETRRA embarked on doing extension, rather than on extension method research. This challenge was addressed in an iterative and interactive way, as part of the project learning cycle.

Discovery learning exercises Discovery learning relies on engaging people in experimentation, observation, measurement and so on, activities which allow people to draw their own 281


conclusions. Creating tools for discovery learning has emerged as an important challenge for scientists (Röling and Jiggins, 1998). Before developing discovery learning exercises, testing scientists' perceptions about local knowledge is required. As farmer field schools have not included issues like seed storage management (Bjoernsen Gurung, 2003), seed health discovery learning exercises were developed with Bangladeshi scientists from national research institutes and universities under PETRRA (Van Mele, 2002). According to one of the senior entomologist who participated in the workshop, "Farmers don't know the exact role of seed moisture content on the development of storage insect pest." A discovery learning exercise was developed to address this knowledge gap on the life cycle of storage insect pests. However, in-depth knowledge analysis carried out for the video project on post-harvest about one year later (see Chapters 5 and 7) revealed that women knew all too well that high seed moisture resulted in higher insect infestation (without knowing about increased insect fecundity rates). Addressing the issue of insect life cycle was useless in this case: the missing knowledge was that moisture was carried by air through the pores of the earthen storage pots. Porosity had to be addressed, not insect life cycles. Scientists' perceptions about local knowledge shouldn't be taken for granted when developing farmer education curricula and tools. Overall, creativity and flexibility are needed to develop conditions in which these discovery learning exercises can be used. Exercises developed in farmer field schools are currently being used in the formal education system (CIPUPWARD, 2003; Arnst et al., 2004). As a learning platform, Going Public also allows similar exercises to be used, but only those that allow people to observe or experience something in a short time, let's say 10 minutes, rather than exercises requiring weekly or season-long observations (see also Bentley et al., 2003).

Visual aids All cases described in this book developed or incorporated visual aids for various purposes. Tools in themselves play a flexible role in extension and farmer education; they can be used or modified as one sees fit and, depending on how they are used, can have a greater or lesser impact. Let us take the example of photos. They were used to stimulate creative thinking in group discussions (Chapters 3 and 7). They also cultivated pride among farmer innovators in various sub-projects (Orsini and Jahn, 2004; Van Mele and

282


Zakaria, 2004), and helped communities to learn about the social dynamics in the adoption of new technologies (Van Mele and Zakaria, 2002). But photos were also used as learning tools in training-of-trainers sessions. A4-sized laminated photographs of farmer interviews and focus group discussions confronted project staff with multiple scenarios (Van Mele et al., 2002). It helped young researchers to gain a better eye for details and to be aware of social dynamics when conducting farmer interviews or focus group discussions with a community. Shushilan combined songs and dance with large paintings depicting major rice pests and natural enemies, how to use organic fertiliser, and so on (Chapter 10). This case, as in the video project, shows the necessity to involve multiple disciplines and farmers in developing messages for rural communities. Scientifically validated information should form the basis of learner-centred farmer education. The NGO SAFE used agroecosystem analysis in farmer field schools to visualise and evaluate the effect of herbicides on rice plants, earth worms and other living organisms. The tool helped farmers make better-informed decisions; initially they feared herbicides would "poison" the soil or reduce soil fertility if used continuously on the same field (Chowhan et al., 2004). In another sub-project, villagers drew soil fertility maps that helped them in testing and improving their soil fertility management (see Chapter 8). During a PETRRA workshop on communication material development in April 2004, scientists worked alongside non-formal education specialists and graphic designers to produce diagrams of their uptake and extension methods. Earlier on, the same mix of people had developed extension materials with extensionists and farmers. But communication is not only about making things visible and easily accessible to a client group. Coordinated efforts are needed to make optimal use of the diversity of information sources, communication tools and learning methods. The way this is shaped is context-specific and depends on the resources available in the innovation system, such as money, motivation, moral support, experience, enthusiasm, knowledge, creativity and collaborative spirit.

The Bangladesh Rice Knowledge Bank: public knowledge organised PETRRA helped to sustain the research findings in the public domain and to increase user access to updated knowledge and technology beyond projects and organisations by catalysing the Bangladesh Rice Knowledge Bank. This linked to a regional initiative of IRRI to establish digitised, country-specific rice information systems. Extension service providers are the direct beneficiaries, as both English and Bengali versions of technical information, leaflets and posters can be downloaded for printing. It is regularly updated with an emphasis on low-cost technologies, and is available on CD-ROM, in print and online (www.knowledgebank-brri.org).

283


The institutional home of the knowledge bank is the Bangladesh Rice Research Institute, linked to the focal area forums, and with back up support from IRRI to ensure long-term sustainability.

INNOVATION SYSTEMS RESEARCH Innovation systems research emphasises the relationship between innovations and its evolving political, economic and social context. It provides a framework for (i) exploring patterns of partnerships; (ii) revealing and managing the institutional context that governs these relationships and processes; (iii) understanding research and innovation as a social process of learning; and (iv) thinking about capacity building in a systems sense (Hall, 2002). The success of an innovation system depends on its capacity to change in ways that are positive in a development sense. Although organisations are important, it is often individuals rather than organisations that are critical (Clark et al., 2003). Röling and Jiggins (1998) have argued for some time that more professionalism is needed in thinking about people if sustainable development is to be reached. Learning about people helps to manage institutions that drive innovation systems, and may require certain tools to facilitate this. To give an example, Van Mele and Zakaria (2002) developed a new tool, namely the Innovation Tree, to visualise and analyse the way an innovation spreads over time between community members. Learning about local innovators led to changed behaviour of staff at the Rural Development Academy, as reflected in the way subsequent activities and project proposals were developed. During the documentation of the cases presented in this book, which was considered an integral part of the institutional learning process, we used narratives, enterprise webs, photographs, actor linkage maps, innovation systems research methods (Hall et al., 2003b; Matsaert et al., 2004) and various other social science methods. We agree with Biggs and Smith (2003) that more tools are needed to analyse organisational cultures and personal behaviours, but at the same time we recommend a wider use of tools for stimulating creative thinking and local ownership (see also Box 21.3).

UNFINISHED BUSINESS We wanted to encourage partnerships that equally and effectively combined strengths and eliminated weaknesses of different groups of people and their organisations. This doesn't happen often enough, nor is it the accepted norm among government organisations, NGOs or the private sector. We saw changes in behaviour and better joint working practices, though we're still not sure how wider improvements can be stimulated. Will other NGOs and government organisations 284


change the way they work and collaborate after observing partnerships forged through PETRRA? Mechanisms to stimulate wider changes are still required. 1.

Avoid funding or promoting a single blue-print extension method

2. a

Use actor analysis to analyse organisational cultures, strengths, ambitions and weaknesses in engineering partnerships

3.

Apply innovation systems research in planning projects and identifying local innovations

4. a

Create early, low-budget opportunities for multiple actors to interact and learn to work with each other

5. a

Train people involved in community needs assessment to distinguish between implicit and explicit demand

6. a

Link agricultural R&D activities, whether by government, non-government of private sector, more closely to the established education system

7. a

Incorporate communication specialist and broad-based professionals with experience in learning approaches from the beginning of the project

8.

Build adult learning and discovery learning principles into mass media programmes

9. a

Increase understanding of institutional elements that are important in developing local ownership over technologies and extension methods

Box 21.3 Suggestions for Successful Innovation Systems

10. Develop mechanisms to increase creative thinking capacity among all actors 11. Introduce new ideas in the system as early as possible in a subtle way 12. Allow for a flexible management structure that can be responsiveness to opportunities 13. Support institutional learning continuously.

The emerging practice of NGOs in Bangladesh to link agriculture to their social development programmes is encouraging but still in its infancy. Better promotion of links with agribusiness and cross-fertilisation between extension and education, whether formal or non-formal, would benefit from 'innovation'. The explicit policy of pro-poor development is already stimulating new ideas, though that must be matched by a flexibility and commitment to change in institutes and organisations. Policies themselves need refining as evidence of success is gathered. Partnerships and learning networks help to share ideas and create new ones. But ideas need to be tested, to branch out, amplified and be modified if they are to benefit the millions of poor farmers. We do not have enough experience to confidently mix and match extension, education and communication methods and tools. More experimentation is needed and a willingness to accept that not everything works the first time round. Information sources, such as the Bangladesh Rice Knowledge Bank, are just becoming part of decentralised information hubs. But creating the trough at which the horse can 285


drink is not enough. How will service providers gain access to these hubs? Validating and incorporating local knowledge and innovations is yet another challenge. We found many local organisations doing exciting work, but why is this ignored so often? The simple answer is poor documentation. Writing things down takes time, a certain creativity and persistence. It also has to be seen as rewarding in its own right. We hope that the efforts made in writing this book help to shine the light on the forgotten heroes of local development, and that the chapters are seen as a warm tribute in part to local creativity and methodological diversity. At the end of PETRRA can we say that we've answered all the questions? The short answer is that this is never going to be possible. Development doesn't start and stop. It keeps on refining, applying, going back and then going forward, providing solutions and doing new things alongside old improved things. This book is our way of documenting what PETRRA and its partners have done and achieved. It points to things that still need to be done. Above all else, this book is a testament to the innovations produced by committed champions for pro-poor development in Bangladesh. We hope it suggests how we can each become one and provides the inspiration for you to have a go yourself.

REFERENCES AND RECOMMENDED READING Alex, G. (2001) Review of decentralized EPICC structures and Upazila Partnership Initiative Fund Projects. Consultancy Report on Agricultural Extension for DFID Bangladesh, October 21-31, 2001. Almekinders, C. and Louwaars, N. (1999) Farmers' Seed Production. New Approaches and Practices. Intermediate Technology Publications, London. Arnst, R, Senerath, H. B., Geeves, R., Jatiket, M. and Winearto, N. (2004) Starting early: IPM in elementary school teaching. In: Eveleens, K. Jiggins, J. and Lim Guan Soon (eds) A History of IPM in Asia and Beyond. FAO, Bangkok, in press. Ashby, J. A. and Sperling, L. (1994) Institutionalising participatory, client-driven research and technology development in agriculture. ODA Network Paper 49. Ashby, J. A., Braun, A. R., Gracia, T., Guerrero, M. d. P., Hernandez, L. A., Quiroz, C. A. and Roa, J. I. (2000) Investing in Farmers as Researchers: Experience with Local Agricultural Research Committees in Latin America. CIAT, Cali, Colombia. Ayas, K. (1999) Project design for learning and innovation: Lessons learned from action research in an aircraft manufacturing company. In: Easterby-Smith, M., Burgoyne, J. and Araujo, L. (eds) Organizational Learning and the Learning Organization. Developments in Theory and Practice. Sage Publications, London, pp. 176-193. Banu, L. J. and Bode, B. (2002) CARE Bangladesh's FFS approach: new frontiers in farmer empowerment. CARE, Bangladesh. Bartlett, A. (2004) Entry Points for Empowerment. CARE Bangladesh, Dhaka.

286


Barzman, M. S. and Desilles, S. (2002) Diversifying rice-based systems and empowering farmers in Bangladesh using the farmer field school approach. In: Uphoff, N. (ed.) Agroecological Innovations: Increasing Food Production with Participatory Development. Earthscan Publications, London. Bawden, R. (1994) Creating learning systems: a methaphor for institutional reform for development. In: Scoones, I., Thompson, J. and Chambers, R. (eds) Beyond Farmer First. Rural People's Knowledge, Agricultural Research and Extension Practice. Intermediate Technology Publications, London, pp. 258-263. BBS (2004) Statistical Pocketbook of Bangladesh 2002. Bangladesh Bureau of Statistics (BBS), Ministry of Planning, Dhaka. Bebbington, A. J., Merril-Sands, D. and Farrington, J. (1994) Farmer and community organisations in agricultural research and extension: functions, impacts and questions. ODA Network Paper 47. Bentley, J. (1989) What farmers don't know can't help them: the strengths and weaknesses of indigenous technical knowledge in Honduras. Agriculture and Human Values 6, 25-31. Bentley, J. (1994) Facts, fantasies, and failures of farmer participatory research. Agriculture and Human Values 11(3), 140-150. Bentley, J. (2000) The mothers, father and midwives of invention: Zamorano's natural pest control course. In: Stoll, G. (ed.) Natural Crop Protection in the Tropics: Letting Information Come to Life. Agrecol, ICTA, Margraf Verlag, pp. 281-289. Bentley, J. W. and Baker, P. S. (2002) Manual for Collaborative Research with Smallholder Coffee Farmers. CABI Bioscience, Egham, UK. Bentley, J. W. and Boa, E. (2004) Community Plant Health Clinic: An Original Concept for Agriculture and Farm Families. CABI Bioscience, Egham, UK. Bentley, J., Boa, E., Van Mele, P., Almanza, J., Vasquez, D. and Eguino, S. (2003) Going Public: a new extension method. International Journal of Agricultural Sustainability 1(2), 108-123. Berdegué, J. A. and Escobar, G. (2001) Agricultural knowledge and information systems and poverty reduction. World Bank AKIS Discussion Paper, Washington. Berdegué, J. A. and Escobar, G. (2002) Rural diversity, agricultural innovation policies and poverty reduction. AgREN Network Paper 122. Available at: http://www.odi.org.uk/agren/ Besette, G. (2001) Facilitating Stakeholder Participation Through Communication in Participatory Natural Resource Management Research. A Guide to Participatory Development Communication for Research Teams and Practitioners. International Development Research Centre, Ottawa. Biggs, S. D. (1978) Planning rural technologies in the context of social structures and reward systems. Journal of Agricultural Economics 29(3), 257-271. Biggs, S. D. (1990) A multiple source of innovation model of agricultural research and technology promotion. World Development 18 (11), 1481-1499.

287


Biggs, S. D. (1992) Farming systems research and rural poverty: A political economy perspective on institutionalisation. Institutionalisation of a Farming Systems Approach to Development. Proceedings of Technical Discussions. FAO, Rome, pp. 65-84. Biggs, S. D. (2004) An actor innovations systems approach to scaling-up of agricultural technologies. In: Proceedings of a Workshop on Uptake Pathways and Scaling-up of Agricultural Technologies to Enhance the Livelihoods of Nepalese Farmers. Hill Agriculture Research Project, Kathmandu, pp. 35-54. Biggs, S. D. and Farrington, J. (1990) Farm systems research and the rural poor: The historical, institutional, economic, and political context. Paper presented at the 10th Annual Association for Farming Systems Research-Extension Symposium. Michigan State University, Oct. 14-17, 1990. Biggs, S. D. and Matsaert, S. H. (1999) An actor-oriented approach for strengthening research and development capabilities in natural resource systems. Public Administration and Development 19, 231-262. Biggs, S. D. and Matsaert, S. H. (2004) Strengthening poverty reduction programmes using an actor-oriented approach: examples from natural resources innovation systems. AgREN Network Paper 134. Biggs, S. D. and Smith, S. (2003) A paradox of learning in project cycle management and the role of organizational culture. World Development 31(10), 1743-1757. Bjoernsen Gurung, A. (2003) Experiments in store: participatory action research with women farmers of Nepal. Beraterinnen News 1, pp. 31-38. Boa, E., Bentley, J. W. and Stonehouse, J. (2001) Standing on all three legs: the técnico as a cross-cultural occupational group. Economic Botany 55(3), 363-369. Bode, B. (2002) In Pursuit of Power: Local Elites and Union-Level Governance in Rural Northwestern Bangladesh. CARE, Bangladesh. Available at: http://www.careinternational.org.uk/ Braun, A. R., Thiele, G. and Fernandez, M. (2000) Farmer field schools and local agricultural research committees: complimentary platforms for integrated decision-making in sustainable agriculture. AgREN Network Paper 105. Bunch, R. (1989) Encouraging farmers' experiments. In: Chambers, R., Pacey, A. and Thrupp, L. A. (eds) Farmer First. Farmer Innovation and Agricultural Research. Intermediate Technology Publications Ltd, London, pp. 55-61. Chambers, R. (1997) Whose Reality Counts? Putting the First Last. Intermediate Technology Publications, London. Chambers, R., Pacey, A. and Thrupp, L. A. (eds) (1989) Farmer First. Farmer Innovation and Agricultural Research. Intermediate Technology Publications, London. Chapman, R. and Tripp, R. (2003) Changing incentives for agricultural extension - A review of privatised extension in practice. AgREN Network Paper 132. Chowhan, G., Orr , A. and Riches, C. (2004) Partnerships for information flow: herbicides in Bangladesh. Proceedings of National Uptake Workshop. Department of Agriculture

288


Extension, Farmgate, Dhaka held April 17-18, 2004. IRRI, Dhaka, Bangladesh. Christoplos, I. (1996) Poverty, Pluralism and Extension Practice. IIED Gatekeeper Series No. 64. CIP-UPWARD (2003) Farmer Field Schools: Emerging Issues and Challenges. International Potato Center-Users' Perspectives with Agricultural Research and Development. Los Baños, Philippines. Clark, N., Hall, A., Sulaiman, R. and Naik, G. (2003) Research as capacity building: the case of an NGO facilitated post-harvest innovation system for the Himalayan hills. World Development 31(11), 1845-1863. Coldevin, G. and FAO (2001) Participatory communication and adult learning for rural development. Journal of International Communication 7(2), 51-69. Cornwall, A., Guijt, I. and Welbourn, A. (1994) Acknowledging process: challenges for agricultural research and extension methodology. In: Scoones, I., Thompson, J. and Chambers, R. (eds) Beyond Farmer First. Rural People's Knowledge, Agricultural Research and Extension Practice. Intermediate Technology Publications, London, pp. 98-117. DAE (1999) Strategic Plan 1999-2002. Department of Agricultural Extension, Dhaka, Bangladesh. Dagron, A. G. (2001) Making Waves - Stories of Participatory Communication for Social Change. Rockefeller Foundation, New York. DFID (2000) Poverty Elimination and the Empowerment of Women. Strategies for Achieving the International Development Targets. Department for International Development, London. Dixon, J., Gulliver, A. and Gibbon, D. (2001) Farming Systems and Poverty. Improving Farmers' Livelihoods in a Changing World. FAO, Rome. Doner, R. F. and Schneider, B. (2000) The new institutional economics, business associations and development. Business and Society Programme Discussion Papers Series No. 110, International Institute for Labour Studies (IILS), Geneva, Switzerland. Available at http://www.ilo.org/public/english/bureau/inst/ Dorward, A., Kydd, J., Morrison, J., Poulton, C. and Smith, L. (2000) New institutional economics: insights on innovation dissemination and uptake. In: Hainsworth, S. D. and Eden-green, S. J. (eds) Sustaining Change. Proceedings of a Workshop on the Factors Affecting Uptake and Adoption of Research Outputs. Natural Resources International Limited, Chatham, UK, pp. 97-103. Duncan, A., Sharif, I., Landell-Mills, P., Hulme, D. and Jayanta Roy (2002) Bangladesh: Supporting the Drivers of Pro-Poor Change. DFID, London. Available at http://www.lcgbangladesh.org Easterby-Smith, M. and Araujo, L. (1999) Organizational learning: current debates and opportunities. In: Easterby-Smith, M., Burgoyne, J. and Araujo, L. (eds) Organizational Learning and the Learning Organization. Developments in Theory

289


and Practice. Sage Publications, London, pp. 1-21. Engel, P. G. H. (1997) The Social Organization of Innovation: A Focus on Stakeholder Interaction. Royal Tropical Institute, Amsterdam, The Netherlands. FAO (2004) Gender and Food Security. Education, Extension and Communication. Available at http://www.fao.org/Gender/en/educ-e.htm Farrington, J. and Bebbington, A. J. (1994) From research to innovation: getting the most from interaction with NGO's. In: Scoones, I., Thompson, J. and Chambers, R. (eds) Beyond Farmer First. Rural People's Knowledge, Agricultural Research and Extension Practice. Intermediate Technology Publications, London, pp. 203-213. Farrington, J., Christoplos, I., Kidd, A. D. and Beckman, M. (2002a) Can extension contribute to rural poverty reduction? Synthesis of a six-country study. AgREN Network Paper 123. Farrington, J., Christoplos, I., Kidd, A. D., Beckman, M. and Cromwell, E. (2002b) Creating a policy environment for pro-poor agricultural extension: the who? what? and how? ODI Natural Resource Perspectives 80. Feder, G., Willett, A. and Zijp, W. (1999) Agricultural extension: Generic challenges and some ingredients for solutions. World Bank Policy Research Working Paper 2129. Fraser, C. (1987) Pioneering a New Approach to Communication in Rural Areas: the Peruvian Experience with Video for Training at Grassroots Level. FAO, Rome. Freire, P. (2003) Pedagogy of Hope. Reliving Pedagogy of the Oppressed. Continuum, London. Fujisaka, S. (1995) Incorporating farmers' knowledge in international rice research. In: Warren, D. M., Slikkerveer, L. J. and Brokensha, D. (eds) The Cultural Dimension of Development: Indigenous Knowledge Systems. Intermediate Technology Publications, London, pp. 124-139. Gallagher, K. (2003) Fundamental elements of a Farmer Field School. LEISA Magazine 19(1), 5-6. Available at http://www.leisa.info/ Garforth, C. J. and Norrish, P. (2000) Communication perspectives on uptake and adoption. In: Hainsworth, S. D. and Eden-green, S. J. (eds) Sustaining Change. Proceedings of a Workshop on the Factors Affecting Uptake and Adoption of Research Outputs. Natural Resources International Limited, Chatham, pp. 105-146. Gill, G. J., Aziz, M. A., Saleha Begum, Bevan, C., Hossain, S. M. A. and Meadows, K. (2004) Ex-Post Evaluation of the NOPEST and LIFE Projects GCP/INT/758/EC-BGD (21). FAO, Dhaka, Bangladesh. Grameen (2004) Grameen Telecom. Available at http://www.grameen-info.org/grameen Guijt, I., Berdegué, J. A., Loevinsohn, M. and Hall, F. (eds) (2000) Deepening the Basis of Rural Resource Management. ISNAR, RIMISP. Guijt, I. and Shah, M. K. (eds) (1998) The Myth of Community. Gender Issues in Participatory Development. Intermediate Technology Publications, London.

290


Gündel, S., Hancock, J. and Anderson, S. (2001) Scaling-up Strategies for Research in Natural Resources Management: A Comparative Review. University of Greenwich, Greenwich, UK. Hagmann, J., Chuma, E., Murwira, K. and Connolly, M. (1998) Learning Together Through Participatory Extension: A Guide to an Approach Developed in Zimbabwe. Harare: AGRITEX / GTZ / ITDG-Zimbabwe. Hagmann, J., Chuma, E., Murwira, K. and Connolly, M. (1999) Putting process into practice: operationalising participatory extension. AgREN Network Paper 94. Hall, A., Sivamohan, M. V. K., Clark, N., Taylor, S. and Bockett, G. (2001) Why research partnerships really matter: innovation theory, institutional arrangements and implications for developing new technology for the poor. World Development 29(5), 783-797. Hall, A. (2002) Innovation systems and capacity development: an agenda for North-South research collaboration? International Journal of Technology Management and Sustainable Development 1, special edition. Hall, A. J., Yoganand, B., Sulaiman, R. V. and Clark, N. G. (2003a) Innovations in innovation: partnership, learning and diversity in the generation, diffusion and use of new knowledge. In: Hall, A. J., Yoganand, B., Sulaiman, R. V. and Clark, N. G. (eds) PostHarvest Innovations in Innovation: Reflections on Partnership and Learning. DFID Crop Post-Harvest Programme, South Asia and NR International Limited, Kent, pp. 1-15. Hall, A. J., Yoganand, B., Sulaiman, R. V. and Clark, N. G. (eds) (2003b) Post-Harvest Innovations in Innovation: Reflections on Partnership and Learning. DFID Crop Post-Harvest Programme, South Asia and NR International Limited, Kent. Harris, R. (2001) Telecentres in Rural Asia: Towards a Success Model. International Conference on Information Technology, Communications and Development, Kathmandu. Available at http://unpan1.un.org/intradoc/groups/public/documents/apcity/unpan006304.pdf Hassanullah, M. (2002) Agricultural extension of Bangladesh: A case of reform initiatives. Paper presented at a workshop Extension and Rural Development: A Convergence of Views on International Approaches?, held November 12-15, 2002 in Washington DC. Heeks, R. (1999) Information and communication technologies, poverty and development. Development Informatics Working Paper Series 5. Heong, K. L. and Esacalada, M. M. (1997) Perception change in rice pest management: a case study of farmers' evaluation of conflict information. Journal of applied Communications 81(2), 3-16. Heong, K. L., Esacalada, M. M., Huan, N. H. and Mai, V. (1998) Use of communication media in changing rice farmers' pest management in the Mekong Delta, Vietnam. Crop Protection 17(5), 413-425. Hossain, M., Paris, T. R., Bose, M. L. and Chowdhury, A. (2004) Nature and impact of women's participation in economic activities in rural Bangladesh. CPD-IRRI Policy 291


Brief 7. IRRI, Dhaka, Bangladesh. IIRR (1999) Research Partnerships: Issues and Lessons from Collaborations of NGO's and Agriculturs Research Institutions. International Institute of Rural Reconstruction, Cavite, Philippines. IIRR (2000) Going to Scale: Can We Bring More Benefits to More People More Quickly? International Institute of Rural Reconstruction, Cavite, Philippines. Jarillo, J. C. (1995) Strategic Networks: Creating the Borderless Organisation. ButterworthHeinemann, Oxford. Katz, E. (2002) Innovative Approaches to Financing Extension for Agriculture and Natural Resource Management. LBL, Swiss Center for Agricultural Extension, SDC, Swiss Agency for Development and Cooperation, Lindau. Kenmore, P., Litsinger, J. A., Bandong, J. P., Santiago, A. C. and Salac, M. M. (1987) Philippine rice farmers and insecticides: thirty years of growing dependency and new options for change. In: Tait, J. and Napompeth, B. (eds) Management of Pests and Pesticides. Farmers' Perceptions and Practices. Westview Press, Colorado, pp. 98-115. Kerr, N. L. and Kaufman-Gilliland, C. M. (1994) Communication, commitment, and cooperation in social dilemmas. Journal of Personality and Social Psychology 66, 513-529. Khan, Z. (2003) Closing the Gap: Putting EU and UK Gender Policy into Practice - South Africa, Nicaragua and Bangladesh. One World Action, London. Available at http://www.oneworldaction.org Klein, P. G. (1999) New Institutional Economics. Available at http://encyclo.findlaw.com Kolb, D. A. (1984) Experiential Learning. Experience as the Source of Learning and Development. Prentice Hall, New Jersey. Lahai, B. A. N., Goldey, P. and Jones, G. E. (2000) The gender of the extension agent and farmers' access to and participation in agricultural extension in Nigeria. Journal of Agricultural Education and Extension 6(4), 223-233. Leeuwis, C. (1995) The stimulation of development and innovation: reflections on projects, planning, participation and platforms. Journal of Agricultural Education and Extension 2(3), 15-28. Lovell, K (1992). Breaking the Cycle of Poverty - The story of BRAC. Kumarian Press, New York. Macdonald, M. (2003) Closing the Gap: Bangladesh. One World Action, London. Available at http://www.oneworldaction.org Magor, N. (1996) Empowering Marginal Farm Families in Bangladesh. PhD dissertation. Adelaide University, Adelaide, South Australia. Matsaert, H., Ahmed, Z., Salam, S. A., Aziz, M., Rahman, M. M., Hussain, F. and Islam, N. (2004) An Actor Oriented Analysis of Chilli and Beef Production in the JamalpurChars of Bangladesh. Final Report, draft 5 May 2004, of the DFID Crop Post Harvest Programme. 292


Mednick, F. (2004) The Rise http://www.newhorizons.org/

of

Non-Formal

Education.

Available

at

Meisner, C. A., Sufian, A., Baksh, E., Smith O'Donoghue, M., Razzaque, M. A. and Shaha, N. K. (2004) Whole family training and adoption of innovations in wheat-producing households in Bangladesh. Journal of Agricultural Education and Extension, in press. Morrison, J., Dorward, A., Kydd, J., Poulton, C. and Smith, L. (2000) Sustainable Livelihoods and New Institutional Economics. Available at http://www.livelihoods.org/pip Nalebuff, B. and Ayres, I. (2003) Why Not? How to Use Everyday Ingenuity to Solve Problems Big and Small. Harvard Business School Press, Boston. Nathaniels, N. Q. R. (1998) Self help, cashew trees, and learning in rural primary schools, Nachingwea district, Tanzania. Forests, Trees and People Newsletter 35. Available at http://www-trees.slu.se/newsl. Norrish, P. (1998) Radio and video for development. In: Richardson, D. and Paisley, L. (eds) The First Mile of Connectivity. FAO, Rome. Nuruzzaman, M. and Qazi Wadud Newaz (2004) Woman-led cultural extension by Shushilan. Proceedings of National Uptake Workshop. Department of Agriculture Extension, Farmgate, Dhaka held April 17-18, 2004. IRRI, Dhaka, Bangladesh. OECD (1997) National Innovation Systems. Organization for Economic Co-operation and Development (OECD), Paris. Okoth, J. R., Khisa, G. S. and T., J. (2003) Towards self-financed farmer field schools. LEISA Magazine 19(1), 28-29. Orr, A., Seema, F., Nabi, S. A. and Peter, J. I. (2004a) From technology to livelihoods. PETRRA Policy Brief 1. IRRI, Dhaka, Bangladesh. Available at http:// www.petrrairri.org Orr, A., Seema, F., Nabi, S. A. and Peter, J. I. (2004b) From Technologies to Livelihoods: Case Studies of Four PETRRA Sub-Projects. Consultancy Report, IRRI, Dhaka, Bangladesh. Orsini, J. and Jahn, G. (2004) Livelihood Improvement Through Ecology: Scaling-up environmentally safe rice production. Proceedings of National Uptake Workshop. Department of Agriculture Extension, Farmgate, Dhaka held April 17-18, 2004. IRRI, Dhaka, Bangladesh. Pasteur, K. (2002) Changing organisations for agricultural extension in Bangladesh: strategies for change. Lessons for Change in Policy and Organisations 3. Institute of Development Studies, Brighton. Available at: http://www.livelihoods.org/info Pound, B., Snapp, S., McDougall, C. and Braun, A. (eds) (2003) Managing Natural Resources for Sustainable Livelihoods: Uniting Science and Participation. Earthscan Publications, London. Pretty, J. N. and Chambers, R. (1994) Towards a learning paradigm: new professionalism and institutions for agriculture. In: Scoones, I., Thompson, J. and Chambers, R. (eds) Beyond Farmer First. Rural People's Knowledge, Agricultural Research and

293


Extension Practice. Intermediate Technology Publications, London, pp. 182-202. Quizon, J., Feder, G. and Murgai, R. (2001a) Fiscal sustainability of agricultural extension: the case of the farmer field school approach. Journal of International Agricultural and Extension Education 8(1), 13-24. Quizon, J., Feder, G. and Murgai, R. (2001b) Fiscal sustainability of agricultural extension: the case of the farmer field school approach: supplementary remarks. Journal of International Agricultural and Extension Education 8(2), 73-76. Rivera, W. M. and Zijp, W. (eds) (2002) Contracting for agricultural extension. International case studies and emerging practices. CAB International, Wallingford. Rola, A. C., Jamias, S. B. and Quizon, J. B. (2002) Do Farmer Field School graduates retain and share what they learn? An investigation in Iloilo, Philippines. Journal of International Agricultural and Extension Education 9(1), 65-76. Röling, N. G. (1996) What to think of extension? A comparison of three models of extension practice. Extension workshop. Alternative mechanisms for funding and delivering extension. AERDD Bulletin, University of Reading, UK. Röling, N. G. and van de Fliert, E. (1994) Transforming extension for sustainable agriculture: the case of integrated pest management in rice in Indonesia. Agriculture and Human Values 11, pp. 96-108. Röling, N. and Pretty, J. N. (1997) Extension's role in sustainable agricultural development. In: Swanson, B. E., Bentz, R. P. and Sofrancko, A. J. (eds) Improving Agricultural Extension. A Reference Manual. FAO, Rome. Röling, N. G. and Jiggins, J. (1998) The ecological knowledge system. In: Röling, N. G. and Wagemakers, M. A. E. (eds) Facilitating Sustainable Agriculture: Participatory Learning and Adaptive Management in Times of Environmental Uncertainty. Cambridge University Press, Cambridge, pp. 283-311. Röling, N. G. and Wagemakers, M. A. E. (eds) (1998) Facilitating Sustainable Agriculture: Participatory Learning and Adaptive Management in Times of Environmental Uncertainty. Cambridge University Press, Cambridge. Salahuddin, A. (2004) focal area forum - Dream Translated into Reality. Draft paper. IRRI, Dhaka, Bangladesh. Samsuzzaman, S. and Mazid, M. A. (2004) focal area forum in Northwest Bangladesh: Evolution, Approach and Practices. RDRS, Rangpur, Bangladesh. Sarker, A. E. (2001) The secrets of success: the Grameen Bank experience in Bangladesh. Labour and Management in Development Journal 2(1), 1-17. Scheuermeier, U. (2003) Summary of LBL e-mail discussion on rural business development services. Beraterinnen News 1, pp. 14-18. Schwartz, L. A. (1994) The role of the private sector in agricultural extension: economic analysis and case studies. AgREN Network Paper 48. Singhal A. and Rogers, E. (2003) Entertainment-Education: A Communication Strategy for Social Change. Lawrence Erlbaum Associates, London. 294


Solaiman, M., Bhattacharjee, M. K. and Rahman, A. (2004) Knowledge, Attitude and Practice (KAP) Study on Value-Based, Demand-Led Research System of PETRRA. IRRI, Dhaka, Bangladesh. Stroud, A. (2003) Transforming institutions to achieve innovation in research and development. In: Pound, B., Snapp, S., McDougall and Braun, A. (eds) Managing Natural Resources for Sustainable Livelihoods: Uniting Science and Participation. Earthscan Publications, London, pp. 88-112. Swanson, B. E., Bentz, R. P. and Sofrancko, A. J. (eds) (1997) Improving Agricultural Extension. A Reference Manual. FAO, Rome. Taher Mia, M. A. (2004) SHIP: Completion Report of Rice Seed Health Improvement SubProject. IRRI, Dhaka, Bangladesh. Tendler, J. (1997) Good Government in the Tropics. Johns Hopkins University Press, Baltimore. Thiele, G., Nelson, R., Ortiz, O. and Sherwood, S. (2001) Participatory research and training: ten lessons from the Farmer Field Schools (FFS) in the Andes. Currents 27, 4-11. Tripp, R. (2001) Seed Provision and Agricultural Development: The Institutions of Rural Change. ODI, London. Tripp, R. and Pal, S. (2001) The private delivery of public crop varieties: rice in Andhra Pradesh. World Development 29(1), 103-117. van de Fliert, E. (1993) Integrated pest management: farmer field schools generate sustainable practices. A case study in Central Java evaluating IPM training. Wageningen Agricultural University Papers (93)3. PUDOC, Wageningen, the Netherlands. van de Fliert, E. and Matteson, P. C. (1989) Integrated Pest Control. Channels for Extension in Sri Lanka. Journal of Extension Systems 5(1), 33-47. van de Fliert, E., Pontius, J. and Röling, N. G. (1995) Searching for strategies to replicate a successful extension approach: training of IPM trainers in Indonesia. Journal of Agricultural Education and Extension 1(4). Available at http://library.wur.nl/ejae van den Ban, A. (2000) Different ways of financing agricultural extension. AgREN Network Paper 106b. van der Veen, R. (2000) Learning natural resource management. In: Guijt, I., Berdegué, J. A., Loevinsohn, M. and Hall, F. (eds) Deepening the Basis of Rural Resource Management. ISNAR, RIMISP, pp. 15-22. Van Mele, P. (2002) Discovery Learning for Seed Health. Report of Seed Health Improvement sub-project (PETRRA) in Bangladesh. CABI Bioscience, UK. Van Mele, P. (2003). Promising practices for knowledge generation on tree and forest management. In: Van Mele, P. (ed.) 2003. Way Out of the Woods: Learning How to Manage Trees and Forests. CPLPress, Newbury, UK, pp. 135-143. Van Mele, P. and Braun, A. (2004) The importance of methodological diversity in research and development innovation systems. In: Participatory Research and Development 295


for Sustainable Agriculture and Natural Resource Management. A Sourcebook. CIP/UPWARD, Philippines, in press. Van Mele, P. and Zakaria, A. K. M. (2002) The Innovation Tree: a new PRA tool to reveal the innovation adoption and diffusion process. PLA Notes 45, 54-58. Van Mele, P. and Zakaria, A. K. M. (2004) From concept to impact: developing and communicating multipurpose seed drying tables in Bangladesh. In: Participatory Research and Development for Sustainable Agriculture and Natural Resource Management. A Sourcebook. Vol. 3 A. CIP/UPWARD, Philippines, in press. Van Mele, P., Enamul Huda and Guan Soon, L. (2002) Farmer Participatory Research and Technology Development. Report of Seed Health Improvement sub-project (PETRRA) in Bangladesh. CABI Bioscience, UK. van Steenbergen, F. (2003) Creating markets with the poor: selling treadle pumps in India. IIED Gatekeeper Series No. 107. van Veldhuizen, L., Water-Bayers, A. and de Zeeuw, H. (1997) Developing Technology with Farmers: A Trainer's Guide for Participatory Learning. ETC, Leusden, the Netherlands. Waisbord, S. (2001) Family Tree of Theories, Methodologies and Strategies in Development Communication: Convergences and Differences. Rockefeller Foundation, New York. Wallace, I. (1994) Creating learning networks between formal agricultural institutions and rural people: the potential role of local non-governmental organisations as intermediaries. Journal of Agricultural Education and Extension 1(2), 1-14. World Business Council for Sustainable Development (WBCSD) (2004) Doing Business with the Poor. A Field Guide. Available at http://www.wbcsd.org Williamson, O. E. (1985). The Economic Institutions of Capitalism. The Free Press, New York. Winarto, Y. T. (2004) Seeds of Knowledge. The Beginning of Integrated Pest Management in Java. Yale University Southeast Asia, Connecticut.

296