dialogues involving different values of stakeholders have been developed (e.g., Wolfe et al., 2002). Another major barrier for participatory approaches are the limited appre­ciation, rewards and career opportunities for researchers, a limitation which is more pronounced than in the case of interdisciplinarity. A barrier of growing significance, specific for participatory approaches is the "digital divide" (i.e., the difference in access to information technology) between the developed and the developing world and between the rich and the poor (Rao, 2005; Britz et al., 2006; Chetty et al., 2006). It has contributed to inequity and inefficient use of AKST (Bouma et al., 2007).
     The expectations for integrative scientific approaches and the practical preconditions offered by the performance of the knowledge, science and technology generation sys­tem often seem to be in conflict and it has been suggested that for integrated approaches to be feasible and to become more commonplace, institution-level changes in curricula, incentives, evaluation criteria and accountability would be required (Lele and Norgaards, 2005).

4.4.4 Risks associated with integration
Although interdisciplinarity has been increasingly consid­ered the ideal of research, increasing the level of integration has so far been a rocky path in some countries in NAE. The barriers described above are not the only challenges as there are also risks associated with increasing the extent of integration. These risks need to be minimized and managed carefully to ensure that integrative approaches help rather than hinder achievement of the goals of this assessment.
     Interdisciplinary research relies heavily on high-quality disciplinary research. However, many of the changes imple­mented in recent years particularly in Western Europe in the name of integration, streamlining and quality control have resulted in cuts in funding of disciplinary research. This re­search has long provided essential knowledge for AKST and gradual cuts have caused confusion and disillusionment of scientists involved in such research. This development has resulted in fragmentation and loss of continuity of the sci­ence base, weaker links between science and application and less security for the future (OSI, 2006). It might also limit the capacity to respond adequately to current as well as future challenges facing agrifood systems. It has been recommended that the costs and time needed for rebuild­ing expertise be included in evaluations of area of research considered for discontinuation. Finding the optimal balance between integrated approaches and disciplinary approaches has been (and will continue to be) an important challenge. The strategic planning of public sector funding organizations needs to be better joined up at a national level to help main­tain crucial scientific expertise and facilities (OSI, 2006). There are also initiatives to improve strategic planning at an international level to avoid duplication of effort at a time of increasing funding constraints (EURAGRI, 2005).
     Balancing the influence of stakeholders in the develop­ment of AKST agendas to ensure that funds are focused on the areas most relevant to society and the environment, has been a challenge (see also 4.5.3.3 and 4.5.5). Despite much progress in theoretical work there is still little agreement amongst social scientists regarding the best methodologies to be used for citizen participation (Pidgeon et al., 2005).

Analytic-deliberative processes that can accommodate  a very wide plurality of views in public policy discourses and decisions have been recommended (Pidgeon et al., 2005). New technologies represent particular challenges in terms of citizen participation. The problems the general public faces in judging the potential risks and benefits associated with biotechnology are one recent example. Research sug­gests that in general, people rely on the judgment of trusted others rather than making choices vis-à-vis technologically complex new products in a rational fashion (Grove-White et al., 2000). It is, however, noteworthy that choices of citi­zens are also contributed to by their value systems, where scientists are no experts.
     Media have so far preferred to exploit and heighten pub­lic fears of certain new technologies although hope has been expressed that they can change "to encourage mature dis­cussion of the implications of uncertainties and unknowns surrounding new technologies and their insertion into ev­eryday life—as necessary for constructive public debate" (Grove-White et al., 2000). The same encouragement can be addressed at other organisations the general public uses as trustworthy sources of information. An important aspect is also thought to be the need to pay more attention at the earliest development stages to the social constitutions (i.e., the particular social values and assumptions) new technolo­gies are perceived to have (Grove-White et al. 2000).
     Following 15 to 20 years of evolution, participatory techniques are now accepted as part of the mainstream sci­ence for agricultural development, especially in developing countries. Participation is an inherent part in "innovation systems". The difference between one-directional mediation of information and creation of multidirectional, interactive knowledge networks is fundamental (Table 4-2) (Buhler et al., 2002). On the other hand, it has been argued that the more traditional approaches (e.g., technology transfer) have in places been very successful, providing the appropriate in­frastructure was present and that increased use of participa­tion techniques as a research tool has not had a clear impact (Bentley,   1994).  Real impact would require more than short-term technology development efforts (Humphries et al. 2000). Seeing farmer participation in research primarily as a route to the empowerment of local populations and al­most independent of any eventual research outputs has been questioned (Sumberg et al., 2003).
     A more integrated approach and multi-disciplinary re­search programs should not lead to less disciplinary research and a depletion of agricultural research but should be seen as a reinforcement of agricultural research. The integration of different structures carries the risk of increasing the ad­ministrative burden and wasting funds where it has led lead to an additional layer of bureaucracy. Approaches in inte­gration that do not increase the layers of bureaucracy may be a challenge but would be a more efficient use of limited resources.

4.5 Development of Structures, Funding and Agenda of AKST

4.5.1 Establishment of structures
Much of the invention and technological improvement in NAE agriculture before 1840 and to a lesser extent up to