land use with deforestation is likely to remain the most economically feasible and politically viable development strategy in the Amazon region because vast areas of cheap land are accessible and markets are distant.
In sum, the causes of Amazon deforestation lie partly in the same frontier conditions that have led to extensive land use in nineteenth century North America and elsewhere. In addition, development policy around the world has supported capital-intensive development of export monocultures. The unique institutional and political history of Brazil has helped determine the particular development pattern there, a pattern significantly different from that of tropical forest development in Zaire or Indonesia (Allen and Barnes, 1985; Brookfield et al., 1991; Lal et al., 1986). A key to the future of the forests lies in policy changes that could limit deforestation and extensive land use while increasing food production from existing agricultural areas. However, the social and economic changes brought about by Amazonian development have created barriers to making and implementing such policies.
The examples above illustrate how the proximate causes of global environmental change result from a complex of social, political, economic, technological, and cultural variables, sometimes referred to as driving forces. They also show that studies of driving forces and their relationships have been and can be done (National Research Council, 1990b; Turner, 1989). However, little of this research has been conducted on a global scale, for at least three important reasons: demand for such studies is a very recent phenomenon; relevant data at the global level are scarce; and social driving forces may vary greatly with time and place. Consequently, much additional work is needed to support valid global generalizations.
We distinguish five types of social variables known to affect the environmental systems implicated in global change: (1) population change, (2) economic growth, (3) technological change, (4) political-economic institutions, and (5) attitudes and beliefs.
Vocal arguments have been made for each of these as the exclusive, or the primary, human influence on global environmental change. In each instance, supportive evidence exists below the global level. Evidence at the global level, however, is generally insufficient either to demonstrate or dismiss claims that a par-