subsion of its sovereignty offshore; and (2) freedom of the high seas, meaning the freedoms of navigation and fishing in the high sea beyond that offshore coastal area (Joyner, 2000).

     The first principle relies on the comanagement between states and coastal communities in planning, regulating, and conducting resource management (Borgese, 1999). One of the main issues is the obligation for states to maintain or restore populations of harvested fish at levels that produce a "maximum sustainable yield". "Non-exploitive users", i.e., the rest of society's citizens, also have a right of access to the Exclusive Economic Zone for other functions, which include permission to locate aquaculture installations, mineral mining, shipping access, etc. decisions on which remain with government (Caddy, 1999).

      On the second principle, a UN Agreement for the Conservation and Management of Straddling Fish Stocks and Highly Migratory Fish Stocks has been adopted in 1995, mandating states to establish subregional and regional conventions and organizations to facilitate conservation and management of living resources, and an International Seabed Authority for the deep ocean floor and non living marine resources. Except for sedentary species of the sea floor, international fisheries agreements do not speak in terms of ownership of resources but of access rights. This distinction raises the fine point as to the timing of the access and even whether this right could be extended to include the progeny of the resource share in future rights. A corpus of international law has evolved around the 1982 Law of the Sea Convention for protecting and managing the world's oceans (Joyner, 2000), which will likely be extended in the future (Caddy, 1999). Assessments such as the MA, point out that these arrangements are insufficient to avoid a decline of populations of harvested fisheries and 25% of the oceans are overfished, creating problems for both the fish species and the fishermen depending on them. The setting of fishing quotas doesn't take into account the effects of the withdrawal of one species on the functioning of the whole marine ecosystem: it alters not only the targeted fish population but also the other trophic levels concerned by this species as prey or predator. In most situations there is insufficient knowledge on the functioning of marine complex ecosystems to design better management rules.

Challenges for public research and policy options. Scientific knowledge has to help to understand the complexity of such situations, in the oceans as well as on the continents, to formalize these different sets of right regimes and also to design new ways for collective action for the fair implementation of such rights, and reach optimally sustainable management of renewable natural resources. Such knowledge has to guide the design of laws, incentives, contracts, taxes, quotas, permits and licenses that take into account the diversity of situations and that avoid blueprint solutions.

Natural Resources Management Policies. Since state appropriation of NRM based on positive law may coexist with the modalities of local rights systems, which distinguish access to, usage, exploitation, ownership, alienation, exclusion, of "common" goods at a collective level, one option is to recognize that the "law of the land" may further involve land tenure

systems that cannot be reduced to individual ownership. Collective ownership and management of natural resources is protected in Article 10c of the CBD (Sustainable Use of Components of Biological Diversity). Indigenous groups have referenced this Article to help defend their collective rights and NRM practices against governments that would ignore these rights in fulfilling commitments to protect "global" resources. New instruments for collective action have to make explicit and feasible the fair implementation of collective rights and NRM practices in order to obtain the best and sustainable management of renewable natural resources. Formal institutions have to take into account this diversity of NRM knowledge and avoid conforming only to a concept of individual ownership and rights.

7.5 Pro-Poor Agricultural Innovation

7.5.1 Technology supply push and the global agricultural treadmill

The dominant policy model for promoting innovation is called the linear model (Kline and Rosenberg, 1986), or the transfer of technology model (Chambers and Jiggins, 1987; Chapter 2). Also known as "technology supply push," this approach relies on the agricultural treadmill (Cochrane, 1958) i.e., market-propelled waves of technological change that squeeze farm-gate prices, stimulate farmers to capture economies of scale, deliver high internal rates of return to investments in agricultural research (Evenson et al., 1979), but also encourage externalization of significant social and environmental costs (Lal et al., 2005; Mukherjee and Kathuria, 2006).

      While the technology push model provided the basis for the positive impacts of the Green Revolution in favorable areas (Castillo, 1998) and under defined conditions that typically included high subsidies on fertilizers and pesticides (Pontius et al., 2002), it has not served nearly as well as resource-poor areas that are highly diverse, rain fed, and risk prone, and that currently hold most of the world's poor (Anderson et al., 1991; Biggs and Farrington, 1991; Vanlauwe et al., 2006).

      The market-propelled diffusion of innovations called "the agricultural treadmill" (Cochrane, 1958) has been ongoing in developed market economies for 50 years or more. The literature observing the process for hybrid maize in the American Midwest goes back to 1943 (Ryan and Gross, 1943). During these 50 years, farmers in those economies have been able to capture significant economies of scale. The treadmill process in those economies has been heavily supported in terms of public funding of agricultural research, education and extension, credit subsidies, land and irrigation development, supportive legislation, access to inputs, services and markets, and the evolution of farmers' organizations and their lobbies that represent farmers' interests at state and federal or EU levels. One can now speak of a "global treadmill" that allows farmers in developed economies to export their (sometimes subsidized) products to developing countries and compete with local small-scale farmers.

      Value added per agricultural worker in 2003 (constant 2000 US$) in developed market economies was 23,081 with a growth over 1992-2003 of 4.4% (FAO, 2005b). For subsion