number of research institutions, technologies released, or number of scientists or extension workers, but also in terms of how they relate to other actors in the wider innovation system. Evaluation parameters in research and extension agencies also need re-examination. Evaluating performance based on number of technologies released has restricted sci­entists from engaging in other equally important aspects such as technology adaptation and problem-solving. Similarly exclusive focus on technology dissemination has restricted extension from engaging in other important institutional in­novations that are required for raising farm incomes. The role of social scientists also needs to change from measuring impacts to experimenting with new institutional arrange­ments and learning from them.
     The need for partnering with the various other orga­nizations involved in agricultural development has been evident in many ESAP nations since the 1990s. There have been increasing calls for public-private partnerships in agri­cultural development in the last decade and several efforts were made to promote this approach. Several innovative institutional arrangements involving a wide range of part­ners emerged in response to the realization that agricultural development involves interaction among a wide range of ac­tors. Fostering such interaction and increased collaboration among multiple partners will require the identification and elimination of the mistrust between potential partners and organizations in both, public and private sectors (Box 5-1).
     The increasing complexity of agricultural development and the rapidly changing external environment necessitates that all actors in the agricultural innovation system includ­ing those directly dealing with AKST embrace partnerships as an organizational principle. It is increasingly clear that there cannot be a blueprint for promoting partnerships, but development of partnership arrangements could be facili­tated through funding arrangements designed to promote and support stakeholder meetings and handholding devel­opment of joint collaborative activities. These need to be supplemented with efforts to reflect on partnership progress and lesson-learning to direct the much-needed institutional changes among different actors in the innovation system. Some  of the  key  recommendations  that  have  emerged through a joint analysis by different stakeholders who have participated in four NRM projects in India are relevant to those interested in promoting partnerships in RDTE (Box 5-2). The projects examined include: (1) integrated man­agement of land and water resources (DFID/NRSP-ICAR); (2) improved livelihoods through a consortium approach (ICRISAT); (3) promotion of zero-tillage (Rice-Wheat Con­sortium) and (4) community development (Aga Khan Rural Support Project).
     Agricultural innovation occurs when different actors in the innovation systems interact and share knowledge and work in partnerships (Figure 5-1). While understand­ing and planning agricultural development interventions, it is worthwhile to use the conceptual framework of "in­novation system". Its attraction is that it recognizes that innovation is not a research driven process simply relying on technology transfer. Instead innovation is a process of generating, accessing and putting knowledge into use and is complicated and context-embedded. Consequently, its main determinants are the interactions of different people, the

ideas they have and the social setting of these interactions and relationships. Its other important insight, which is now widely recognized in the development sector, is that institu­tions really matter. Thus, the attitudes, habits, practices and ways of working that shape how individuals behave have an enormous impact on whether or not innovation takes place (Hall, 2006). Addressing these issues related to gov­ernance and partnerships in AKST assumes primary impor­tance in programs aimed at strengthening AKST in the ESAP region.
     Conventional approaches to strengthening capacity in agriculture focused only on science and technology. This is important and will continue to be important especially for countries with limited science and technology capacity. Emerging frontiers of new knowledge will necessitate orga­nizing special training programs in such select areas. Knowl­edge and information exchange among different countries is required to bridge the gaps in capacity to develop and ap­ply new knowledge. CGIAR centers and international and regional donors play important roles and their efforts will need to be strengthened. However, science and technology capacity alone will not enough to bring about better knowl­edge uptake and use; applying new knowledge will neces­sitate the development of several kinds of capacities among several actors. Capacity to develop and implement policies, experiment and evaluate new approaches and address issues related to quality, standards and markets will all need to be upgraded throughout the ESAP region.
     To attain the development and sustainability goals of AKST, organizations require a wide range of capacity— broadly called innovation capacity. Innovation capacity is the ability of the network of actors in an innovation sys­tem to address problems and to identify, test and imple­ment solutions—in other words, to innovate. It comprises the context-specific range of scientific and other skills and information held by individuals and organizations, practices and routines (institutions) and the patterns of interaction and policies needed to create and put knowledge into pro­ductive use in response to an evolving set of challenges and opportunities (Hall, 2007).
     Options for action are as follows:
•   Capacity development will involve diagnosing the exist­ing innovation system, including exploring the actors, their knowledge and skills, roles, patterns of interac­tion, habits and practices and the policy environment. An innovation systems framework could be used as a diagnostic tool to understand the existing innovation system and also as a framework for planning interven­tions (World Bank, 2006). Learning from the emerging institutional arrangements in the region necessitates a detailed analysis of cases where the various actors in specific contexts come together and collaborate to solve particular problems or address new challenges. What kinds of changes were made, how were they implement­ed, were they sustained at the end of the specific initia­tive and why?
•   Not many organizations have a culture of learning. Op­portunities will need to be created and specifically fund­ed to bring in this change of culture. It will be useful to bring staff together to reflect on lessons learned and discuss how goals could be better achieved. The concept