What is meant by ‘coherence’?  I take it to mean achieving a greater convergence between the following: policies affecting agriculture (including livestock production); crop and livestock producers’ livelihood goals; and supporting research and development efforts. Within this trinity, there is a widespread agreement that producers (and especially small-scale, poor producers) have too often been relegated to third place. The Salzburg report lays a proper emphasis on their full participation, notably in  

• the Institutions and Innovations Working Group’s first recommendation to set up a farmer-owned, farmer-driven fund to direct research, innovation and development towards farmers’ needs;
• the Markets, Trade and Investments Working Group’s third recommendation that smallholders and pastoralists should participate more in policy formation on value chains;
• the Governance and Policy Processes Working Group’s first recommendation for non-state actors to become involved in the policy process;
• the Equity, Rights and Empowerment Working Group’s first recommendation for collaborative partnerships between producers’ organisations, governments, NGOs, banks microfinance and international organisations.

Often repeated is the familiar expression ‘capacity building’

Achieving coherence must go beyond rhetoric. At the project level, communities and research/development agents can and do achieve coherent partnerships but bridging the gap to policy is more difficult. Policy is formulated at a different scale and the policy process (though not necessarily particular policies) has greater continuity than projects. Bridging research/ development and policy calls for continuity, financial resources, a common language, local ownership of the process, enlightened national leadership and stakeholder-based frameworks for negotiation and advocacy. In particular, scale differentiation between local interests (which may or may not achieve consensus) and national policy processes (which have to take account of interests outside agriculture or the rural sector) is a major challenge in trying to bring rural people into policy formation. The new democratic institutions associated with decentralization policies are addressing natural resource management issues (e.g., community forest management). They need to extend their remit into such areas as market regulation, price policies, and input supply. These and other policies determine the incentive structures for producers to invest in increasing output. 

Agricultural development literature has been understandably sectoral in scope, but it should not be assumed that new technologies or management systems can find their way into use without paying proportionate attention to economic or political considerations. In development practice, outside interventionists cannot directly influence policy. This must be attempted by newly-empowered communities within the existing political framework. But the local community will be listened to less, the ‘higher’ up the ladder they aim. However, the research/ development community is no longer necessarily external in personnel (though in funding, it often is). This creates a new opportunity for co-ownership of development initiatives by national or local research or advocacy institutions. 

Depending on specific conditions, mechanisms, frameworks or protocols are needed to link research, development and policy with communities in ways that avoid condescension or patronage towards local communities while at the same time recognising national versus local interests and longer versus shorter time perspectives. It seems less than ideal for this to be undertaken as a specialism rather than fully integrated with the research and development – but such integration calls for an interdisciplinary approach.

In seeking to create mechanisms for policy dialogue, two strategies suggest themselves:

• Forums for dialogue can be convened at district or province level  – this being the ‘highest’ level  where contact between the administration and communities is still immediate, based on touring officers’ itineraries, election campaigns, government programmes and service provision, and cultural affinity between the rulers and the ruled.  Advocacy by community organizations may stand a chance of success at this level, and can be supported by indigenous research/ development organizations. Appropriate signals must, however, be passed to the national policy process. What fine-tuning of governmental procedures is necessary for such an exchange to be effective?.
• Community organizations can combine or aggregate in a hierarchical structure whose top functionaries can press the case for policy priorities directly with the national government. This replicates lobbying by other vested interests, must be resourced and calls for political commitment. To some extent this potential depends on the size and diversity of the country in question. Senegal (for example) is small and well-integrated; Nigeria (for example) is vast and diverse.

Discussions on green revolution seem reluctant to embark on such issues, but both research and developmental experimentation is needed, country by country, to evaluate options for planting structural and institutional frameworks whose continuity can be guaranteed after externally funded projects and researchers withdraw.  The ‘capacity building’ so much favoured in the Salzburg report needs to accomplish a revolution in attitudes among indigenous disciplinary specialists (often trained abroad) who need to assume facilitative or advocacy roles alongside the communities they research or ‘develop’ in the long term.

The case for such institutional frameworks centred on agriculture rests basically in the lack of an alternative, because new democratic processes have not yet proved competence in agricultural issues – more pressing, perhaps, are political negotiations between opposed interests that have hijacked party systems. Rural people still say they feel marginalised; young adults prefer to abandon the village in favour of an insecure but perhaps better rewarded urban life. Agriculture sits uncomfortably on shifting sands of social change. The future performance of the sector, I suggest, is inevitably bound up with political realities and cannot be wholly encapsulated in a world of soil fertility, crop genetics, agronomic technologies, etc. – important though these are;

Much valuable action research has been conducted on new or adapted institutional structures in a context of natural resource management (a significant part of it supported by the Natural Resources Systems Programme of the UK Department for International Development, 1995-2006). A worthwhile objective would be to carry out a synthesis and evaluation of such work, in terms of its applicability to an African Green Revolution.