conditions under which the involvement of the state may facilitate or impede local management. It has also explored several institutional forms with the potential to improve cross-scale linkages, noting the fluid and diverse nature of these forms. In fact, a rich diversity of institutional forms exists for linking local-level or community-based institutions with those at the regional, national, or international levels. As shown by the Bangladesh fisheries example, these institutional forms are highly dynamic, changing from area to area and year to year.
In addition to management regimes involving these institutional forms, cross-scale linkages also may be enhanced through the use of certain research and management approaches. Of the four such approaches considered in Table 9-5, adaptive management is of particular interest because of its explicit attention to scale and dynamics and because of its potential as a tool for linking social systems and natural systems. This section develops the contention that the adaptive management approach, with a consideration of resilience, is useful for both the theory and practice of cross-scale linkages.
Adaptive management was designed to integrate uncertainty into the decision-making process, and to ensure that policy makers and managers could learn from their successes as well as failures. As a resource management approach and planning tool, it was initially more technocratic than participatory (Holling, 1978). According to Lee (1999), it still “appears to be a ‘top down’ tool useful primarily when there is a unitary ruling interest able to choose hypotheses and test them.” But because it emphasizes learning by doing, feedback relations, and adaptive processes, it has become a particularly promising approach to study the dynamics of systems, both natural and social. Initially concerned with the dynamics of ecosystems, adaptive management also has been applied to the study of the dynamics of linked social and natural systems (Berkes and Folke, 1998).
As used by Hilborn and Walters (1992), adaptive management requires the following six components: (1) identification of alternative hypotheses; (2) assessment of whether further steps are needed to estimate the expected value of additional information; (3) development of models for future learning about hypotheses; (4) identification of policy options; (5) development of performance criteria for comparing options; and (6) formal comparison of options. Together, these steps provide the tools to deal with uncertainty and lay a foundation for learning. We deal with each in turn.
Steps (1), (2), and (4) explicitly require the manager to integrate uncertainty into the management strategy. This is a distinct break from the notion that science can deliver the information needed for resource management, simply and unambiguously. Adaptive management assumes inherent uncertainty in ecosystems and recognizes the limits of knowledge. There are scientific uncertainties that are too