576 International Organization organization to other states in which the conditions that ordinarily would prompt such demands did not exist. A supply-driven explanation Since demand-side explanations for the creation of science policy organizations appear to be on weak ground for most states,the next logical question is,what if these bureaucratic innovations are not demanded inside the state but supplied from outside?In fact,most(roughly 70 percent)of these science policy organizations were created between 1955 and 1975.16 Beginning in the early 1950s two international organizations,UNESCO and the Organization for Economic Cooperation and Development (OECD),began actively to promote this science policy innovation among their member states.17 In this section I describe the promotion activities of one of these organizations, UNESCO,and present evidence that its activities were the impetus for widespread adoption of science policy.Such an explanation allows us to make sense of the data presented earlier since it is consistent with both (1)the low-end clustering on the histograms,indicating that many countries adopted these bureaucracies at very low levels of the indicator variables,and(2)the timing of adoption across states,specifically the rapid increase immediately after the international organizations begin to promote the science policy innovation. Specifically,I will argue that from its inception UNESCO as an organization has had to address two constituencies-the states comprising its membership and the professional experts in its substantive areas of concern.As will be described below,the relationship between these two constituencies within the organization has changed over time.These shifts in turn have been responsible for changes in UNESCO's programs,particularly the rise of science policy as an area of concern.18 Origins of UNESCO's interest in science As originally conceived,UNESCO was to be named the United Nations Educational and Cultural Organization.Science was understood to be part of culture.The notion that science was qualitatively different from other aspects 16.By extending the period by five years to include the years 1976-80,the percentage of adopting states rises to 84.4. 17.For more on the science policy promotion activities of the OECD see Martha Finnemore, "Science,the State,and International Society,"Ph.D.diss.,Stanford University,1991,chap.3. 18.The following account of UNESCO activities is based on research done at the library and archives of UNESCO's Paris headquarters.While these contain a wide variety of documents authored by national governments,it must be acknowledged that carrying out the research at the international organization's headquarters does run the risk of bias in favor of an international organization-driven explanation at the expense of a national one.A research design in which the science policy archives were consulted at a variety of national capitals in countries of different regions and development levels and facing different security situations would be superior. However,such a design was not feasible in this case.576 International Organization organization to other states in which the conditions that ordinarily would prompt such demands did not exist. A supply-driven explanation Since demand-side explanations for the creation of science policy organizations appear to be on weak ground for most states, the next logical question is, what if these bureaucratic innovations are not demanded inside the state but supplied from outside? In fact, most (roughly 70 percent) of these science policy organizations were created between 1955 and 1975.16 Beginning in the early 1950s two international organizations, UNESCO and the Organization for Economic Cooperation and Development (OECD), began actively to promote this science policy innovation among their member states.17 In this section I describe the promotion activities of one of these organizations, UNESCO, and present evidence that its activities were the impetus for widespread adoption of science policy. Such an explanation allows us to make sense of the data presented earlier since it is consistent with both (1) the low-end clustering on the histograms, indicating that many countries adopted these bureaucracies at very low levels of the indicator variables, and (2) the timing of adoption across states, specifically the rapid increase immediately after the international organizations begin to promote the science policy innovation. Specifically, I will argue that from its inception UNESCO as an organization has had to address two constituencies—the states comprising its membership and the professional experts in its substantive areas of concern. As will be described below, the relationship between these two constituencies within the organization has changed over time. These shifts in turn have been responsible for changes in UNESCO's programs, particularly the rise of science policy as an area of concern.18 Origins of UNESCO's interest in science As originally conceived, UNESCO was to be named the United Nations Educational and Cultural Organization. Science was understood to be part of culture. The notion that science was qualitatively different from other aspects 16. By extending the period by five years to include the years 1976-80, the percentage of adopting states rises to 84.4. 17. For more on the science policy promotion activities of the OECD see Martha Finnemore, "Science, the State, and International Society," Ph.D. diss., Stanford University, 1991, chap. 3. 18. The following account of UNESCO activities is based on research done at the library and archives of UNESCO's Paris headquarters. While these contain a wide variety of documents authored by national governments, it must be acknowledged that carrying out the research at the international organization's headquarters does run the risk of bias in favor of an international organization-driven explanation at the expense of a national one. A research design in which the science policy archives were consulted at a variety of national capitals in countries of different regions and development levels and facing different security situations would be superior. However, such a design was not feasible in this case