570 International Organization success,these wartime institutions were then redeployed by states to meet peacetime objectives.7 Robert Gilpin makes a more detailed and broader security argument based on his investigations of French science.8 He argues that France's creation of science policy organizations was the direct result of a perceived threat to French influence and independence from a preponderance of U.S.power immediately following World War II.At one level,this threat was understood militarily and led the French to use their science community to upgrade their defense establishment,notably to establish a separate nuclear strike force.But threats to influence and security in the French view were not limited to the military sphere.The French were also concerned about loss of economic dominance.U.S.economic strength following World War II was viewed with trepidation,and direct U.S.investment in France was viewed as a form of imperialism by a foreign power.During that time the French spoke of a "technology gap"that they must bridge by harnessing French science in the service of French industry to protect French economic independence and integrity.9 Security understood in this sweeping way,as any threat to influence and independence,could operate in so many arenas that developing tidy objective indicators to test for its presence is probably impossible.10 The narrower arguments about security threats understood in a military context are some- what easier to uncover.If armed conflict or the threat of armed conflict is critical,indicators of perceived military threat,such as defense spending as a percentage of GNP,should be correlated with the creation of science policy organizations.States perceiving military threats should be among the first to adopt science policy;conversely,relatively secure states should be clustered among the late adopters. Testing the demand-driven explanations Each of these explanations posits a material condition that then sparks a demand for the state to adopt new tasks and to create new bureaucracies to carry out those tasks.While it would be impractical to investigate the actual 7.See the following chapters in Ina Spiegel-Rosing and Derek de Solla Price,eds.,Science, Technology and Society:A Cross-disciplinary Perspective (London:Sage,1977):Jean-Jacques Salomon,"Science Policy Studies and the Development of Science Policy,"pp.43-70;Sanford Lakoff,"Scientists,Technologists,and Political Power,"pp.355-92;and Harvey Sapolsky, "Science,Technology,and Military Policy,"pp.443-72. 8.Robert Gilpin,France in the Age of the Scientific State (Princeton,N.J.:Princeton University Press,1968). 9.Ibid. 10.For example,in the French case,threats to influence and independence extended to cultural matters and led France to pursue a number of foreign policy initiatives aimed at preserving and extending French language and culture in other states
570 International Organization success, these wartime institutions were then redeployed by states to meet peacetime objectives.7 Robert Gilpin makes a more detailed and broader security argument based on his investigations of French science.8 He argues that France's creation of science policy organizations was the direct result of a perceived threat to French influence and independence from a preponderance of U.S. power immediately following World War II. At one level, this threat was understood militarily and led the French to use their science community to upgrade their defense establishment, notably to establish a separate nuclear strike force. But threats to influence and security in the French view were not limited to the military sphere. The French were also concerned about loss of economic dominance. U.S. economic strength following World War II was viewed with trepidation, and direct U.S. investment in France was viewed as a form of imperialism by a foreign power. During that time the French spoke of a "technology gap" that they must bridge by harnessing French science in the service of French industry to protect French economic independence and integrity.9 Security understood in this sweeping way, as any threat to influence and independence, could operate in so many arenas that developing tidy objective indicators to test for its presence is probably impossible.10 The narrower arguments about security threats understood in a military context are somewhat easier to uncover. If armed conflict or the threat of armed conflict is critical, indicators of perceived military threat, such as defense spending as a percentage of GNP, should be correlated with the creation of science policy organizations. States perceiving military threats should be among the first to adopt science policy; conversely, relatively secure states should be clustered among the late adopters. Testing the demand-driven explanations Each of these explanations posits a material condition that then sparks a demand for the state to adopt new tasks and to create new bureaucracies to carry out those tasks. While it would be impractical to investigate the actual 7. See the following chapters in Ina Spiegel-Rosing and Derek de Solla Price, eds., Science, Technology and Society: A Cross-disciplinary Perspective (London: Sage, 1977): Jean-Jacques Salomon, "Science Policy Studies and the Development of Science Policy," pp. 43-70; Sanford Lakoff, "Scientists, Technologists, and Political Power," pp. 355-92; and Harvey Sapolsky, "Science, Technology, and Military Policy," pp. 443-72. 8. Robert Gilpin, France in the Age of the Scientific State (Princeton, N.J.: Princeton University Press, 1968). 9. Ibid. 10. For example, in the French case, threats to influence and independence extended to cultural matters and led France to pursue a number of foreign policy initiatives aimed at preserving and extending French language and culture in other states
UNESCO 571 demand-making process over a large number of countries having very different political systems,it is quite simple to check for the existence of conditions said to be prerequisite to those demands.The test reported here compiled and analyzed quantitative indicators of domestic conditions that might prompt creation of a science policy bureaucracy in a sample of forty-four countries chosen to be globally representative in terms of both geography and develop- ment levels.As suggested in the foregoing discussion,these were percentage of GDP spent on R&D;proportion of scientists and engineers in the population; per capita GDP;and percentage of gross national product(GNP)spent on defense.A complete description of the indicators used and the method of compiling them can be found in the appendix. Testing a global sample of states raises issues about comparability among the units of analysis,particularly comparability of developing and industrialized states.Cameroon and the United States,for example,are so different on so many measures that one may question whether the units of analysis are sufficiently alike to make comparison appropriate and meaningful. In this case,comparability of the units is ensured by the research questions being asked.The hypotheses being tested concern the behavior of states as a political and organizational form:What prompts states to adopt new tasks and construct new apparatuses to carry out those tasks?The hypotheses do not carry with them caveats about degrees of stateness,state capacity,or other potentially limiting characteristics.Instead they make arguments about the behavior of states qua states.Cameroon and the United States may be very different states,but they are both states nonetheless.In fact the article will suggest that what is going on in this case is a redefinition of the state as a political and organizational form;that is,a redefinition of what is necessary and appropriate behavior for a state. Figures 1-4 show the distribution of values for each of the indicators of state conditions at the time science policy bureaucracies were created in the countries studied.A quick look at figures reveals that none of the patterns corresponds to the expected patterns described above.If any of these conditions were both necessary and sufficient,there would be a large cluster of adoptions on the histogram at that necessary and sufficient value.Instead,the adoptions appear to occur at a very wide range of values for all four of the variables.No single value of any variable appears likely as a necessary and sufficient condition for adoption. In fact,countries adopted these science bureaucracies at wildly different levels of each of these domestic conditions.Some elaboration from the raw data will make the extremely wide range of variation in values even clearer: (1)Countries created these bureaucracies when they had as few as nine scientists employed in R&D(e.g.,Congo)or as many as half a million (e.g.,the United States and the Soviet Union). (2)R&D spending as a percentage of GDP ranged from 0.01 percent at the time of adoption (Bangladesh)to 1.5 percent(France). (3)Per capita GDP in constant U.S.dollars ranged from a low of $118/year
UNESCO 571 demand-making process over a large number of countries having very different political systems, it is quite simple to check for the existence of conditions said to be prerequisite to those demands. The test reported here compiled and analyzed quantitative indicators of domestic conditions that might prompt creation of a science policy bureaucracy in a sample of forty-four countries chosen to be globally representative in terms of both geography and development levels. As suggested in the foregoing discussion, these were percentage of GDP spent on R&D; proportion of scientists and engineers in the population; per capita GDP; and percentage of gross national product (GNP) spent on defense. A complete description of the indicators used and the method of compiling them can be found in the appendix. Testing a global sample of states raises issues about comparability among the units of analysis, particularly comparability of developing and industrialized states. Cameroon and the United States, for example, are so different on so many measures that one may question whether the units of analysis are sufficiently alike to make comparison appropriate and meaningful. In this case, comparability of the units is ensured by the research questions being asked. The hypotheses being tested concern the behavior of states as a political and organizational form: What prompts states to adopt new tasks and construct new apparatuses to carry out those tasks? The hypotheses do not carry with them caveats about degrees of stateness, state capacity, or other potentially limiting characteristics. Instead they make arguments about the behavior of states qua states. Cameroon and the United States may be very different states, but they are both states nonetheless. In fact the article will suggest that what is going on in this case is a redefinition of the state as a political and organizational form; that is, a redefinition of what is necessary and appropriate behavior for a state. Figures 1-4 show the distribution of values for each of the indicators of state conditions at the time science policy bureaucracies were created in the countries studied. A quick look at figures reveals that none of the patterns corresponds to the expected patterns described above. If any of these conditions were both necessary and sufficient, there would be a large cluster of adoptions on the histogram at that necessary and sufficient value. Instead, the adoptions appear to occur at a very wide range of values for all four of the variables. No single value of any variable appears likely as a necessary and sufficient condition for adoption. In fact, countries adopted these science bureaucracies at wildly different levels of each of these domestic conditions. Some elaboration from the raw data will make the extremely wide range of variation in values even clearer: (1) Countries created these bureaucracies when they had as few as nine scientists employed in R&D (e.g., Congo) or as many as half a million (e.g., the United States and the Soviet Union). (2) R&D spending as a percentage of GDP ranged from 0.01 percent at the time of adoption (Bangladesh) to 1.5 percent (France). (3) Per capita GDP in constant U.S. dollars ranged from a low of $118/year
572 International Organization 24 22 20 18- 16- 14 12- 10 6 2 0 0.0-0.30.3-0.60.6-0.90.9-1.21.2-1.51.5-1.81.8-2.12.1-2.42.4+ Percentage of gross domestic product spent on R&D FIGURE 1.Research and development(R&D)spending at the time of science policy adoption (Pakistan)to a high of more than $9,000/year(Denmark)at the time these bureaucracies were created. (4)Defense spending as a percentage of GNP ranged from 0.7 percent (Mexico,Sri Lanka)to more than 10 percent at the time of adoption(France, Iraq,Jordan,and the Soviet Union). The range of variation on the defense variable is more than a factor of ten; the range of variation on all of the other variables is a factor of one hundred or more.Ranges of variation this large do not readily suggest any causal connection between sufficient state conditions and the adoption of science bureaucracies. Similarly,Figures 1-4 provide little support for the necessary condition hypothesis,that is,that there is some minimum threshold value of these variables that triggers demand for the bureaucracy.If such a value existed,we should see very few (or no)adoptions at the low end of the value range for one or more of these variables;all values would be spread across the upper end of the range at or above the necessary condition level. Again,the far-flung distribution of values revealed in these figures and elaborated in the text above does not readily support this proposition.Rather than clustering at the upper end of the value ranges,there seems to be a concentration of values at the low end,particularly on the science variables. This is clearly not a bunching that would support the existence of a necessary
572 International Organization 24—i 22- 20 — 18 — 16 — 14— 6 — 4 — 0.0-0.3 0.3-0.6 0.6-0.9 0.9-1.2 1.2-1.5 1.5-1.8 1.8-2.1 2.1-2.4 2.4+ Percentage of gross domestic product spent on R&D FIGURE l. Research and development (R&D) spending at the time of science policy adoption (Pakistan) to a high of more than $9,000/year (Denmark) at the time these bureaucracies were created. (4) Defense spending as a percentage of GNP ranged from 0.7 percent (Mexico, Sri Lanka) to more than 10 percent at the time of adoption (France, Iraq, Jordan, and the Soviet Union). The range of variation on the defense variable is more than a factor of ten; the range of variation on all of the other variables is a factor of one hundred or more. Ranges of variation this large do not readily suggest any causal connection between sufficient state conditions and the adoption of science bureaucracies. Similarly, Figures 1-4 provide little support for the necessary condition hypothesis, that is, that there is some minimum threshold value of these variables that triggers demand for the bureaucracy. If such a value existed, we should see very few (or no) adoptions at the low end of the value range for one or more of these variables; all values would be spread across the upper end of the range at or above the necessary condition level. Again, the far-flung distribution of values revealed in these figures and elaborated in the text above does not readily support this proposition. Rather than clustering at the upper end of the value ranges, there seems to be a concentration of values at the low end, particularly on the science variables. This is clearly not a bunching that would support the existence of a necessary
UNESCO 573 24- 22 18- 16- 12- 4- 2、 0 0.0-0.20.2-0.40.4-0.60.6-0.80.8-1.01.0-1.21.2-1.41.4-1.61.6+ Scientists and engineers per 1,000 population FIGURE 2.Proportion of scientists and engineers in the population at the time of science policy adoption and sufficient condition,since the bunching in each case is accompanied by a large number of data points at the high end of each scale.Instead,it appears to be a strong negation of any necessary-but-not-sufficient-condition argument.If arriving at some minimum threshold level of these variables is supposed to trigger demand for a science policy bureaucracy,that threshold must be so low as to have very little explanatory power. In fact a large number of small,poor,technologically unsophisticated,and militarily unthreatened countries created these bureaucracies in the 1950s and 1960s.It is this group that accounts for the clustering of data points at the low end of Figures 1-4.Guatemala,for example,created its Consejo Nacional de Investigacionnes Cientificas y Tecnicas in 1966 when it reported having only fourteen scientists employed in R&D jobs,spent only 0.01 percent of GDP on research,had a GDP per capita of $806,and,since it faced no serious military threats,spent only 1.07 percent of GNP on defense.Cameroon and the Congo were equally unlikely candidates for a science bureaucracy. 11.The Congo created its Conseil National de la Recherche Scientifique in 1963 when it reported having only nine scientists engaged in R&D jobs and when spending on R&D was only 0.11 percent of GDP.Measured in U.S.dollars,GDP per capita was only $253 that year,and military spending accounted for only 2.04 percent of GNP.Cameroon created its Office National de la Recherche Scientifique et Technique in 1965 when it reported employing only eighty scientists in research jobs and spending only 0.16 percent of its GDP on research.Per capita GDP was $334 for that year,and the country spent only 2.3 percent of its GNP on defense
UNESCO 573 24—I 22— 20 — 18 — 16 — 55 14— £ 6 — 4 — 2 — 0.0-0.2 0.2-0.4 0.4-0.6 0.6-0.8 0.8-1.0 1.0-1.2 1.2-1.4 1.4-1.6 1.6+ Scientists and engineers per 1,000 population FIGURE 2. Proportion of scientists and engineers in the population at the time of science policy adoption and sufficient condition, since the bunching in each case is accompanied by a large number of data points at the high end of each scale. Instead, it appears to be a strong negation of any necessary-but-not-sufficient-condition argument. If arriving at some minimum threshold level of these variables is supposed to trigger demand for a science policy bureaucracy, that threshold must be so low as to have very little explanatory power. In fact a large number of small, poor, technologically unsophisticated, and militarily unthreatened countries created these bureaucracies in the 1950s and 1960s. It is this group that accounts for the clustering of data points at the low end of Figures 1-4. Guatemala, for example, created its Consejo Nacional de Investigacionnes Cientificas y Tecnicas in 1966 when it reported having only fourteen scientists employed in R&D jobs, spent only 0.01 percent of GDP on research, had a GDP per capita of $806, and, since it faced no serious military threats, spent only 1.07 percent of GNP on defense. Cameroon and the Congo were equally unlikely candidates for a science bureaucracy.11 11. The Congo created its Conseil National de la Recherche Scientifique in 1963 when it reported having only nine scientists engaged in R&D jobs and when spending on R&D was only 0.11 percent of GDP. Measured in U.S. dollars, GDP per capita was only $253 that year, and military spending accounted for only 2.04 percent of GNP. Cameroon created its Office National de la Recherche Scientifique et Technique in 1965 when it reported employing only eighty scientists in research jobs and spending only 0.16 percent of its GDP on research. Per capita GDP was $334 for that year, and the country spent only 2.3 percent of its GNP on defense
574 International Organization 24- 22- 20 18- 16 14- 12- 10- 6 4- 2- 0-750750-15001500-3.0003.0004.5004.500-6.0006.000-7.5007500-9.0009,000+ GDP per capita in constant SUS(1980) FIGURE 3.Gross domestic product (GDP)per capita at the time of science policy adoption At the same time,the histograms do show that some countries create science bureaucracies at reasonably high levels of all the indicator variables.Signifi- cantly,the first instances of this science bureaucracy creation occur among this group,suggesting that demand-driven explanations may fit some of the earliest adopters of science policy.Britain,the first adopter,clearly created its Department of Science and Industrial Research in 1915 for security reasons to counter German advances in chemicals and machinery that were directly supporting the German war effort.12 The establishment of the National Science Foundation in the United States in 1950 was explicitly related to concerns about military and industrial competitiveness and was strongly influenced by the creation of the atom bomb.13 French science policymaking,as chronicled by 12.Alter,The Reluctant Patron.See also Roy McLeod and E.Kay Andrews,"The Origins of the D.S.I.R.:Reflections on Ideas and Men,1915-1916,"Public Administration,vol.48,no.1,1970,pp. 23-48;and Ian Varcoe,"Scientists,Government,and Organized Research in Great Britain 1914-1916,"pp.192-216.The United Kingdom is not included in the quantitative analysis above because science data for that country for 1915 are unavailable. 13.Dickson,The New Politics of Science.See also J.Merton England,A Patron for Pure Science: The National Science Foundation's Formative Years,1945-1957 (Washington,D.C.:National Science Foundation,1982);N.Dupree,Science in the Federal Government;Bruce Smith,American Science Policy Since World War II (Washington,D.C.:Brookings Institution,1990);and U.S. Congress,House Committee on Science and Technology,Task Force on Science Policy,A History of Science Policy in the United States,1940-1985,Science Policy Study Background Report,no.1, 99th Congress,2d sess.,1986,serial R
574 International Organization 8 3 24 22 20 18 16 14— I 8 - 6 — 4 — 2 — 0 0-750 750-1,500 1,500-3,000 3,000-4,500 4,500-6,000 6,000-7,500 7,500-9,000 9,000+ GDP per capita in constant $US (1980) FIGURE 3. Gross domestic product (GDP) per capita at the time of science policy adoption At the same time, the histograms do show that some countries create science bureaucracies at reasonably high levels of all the indicator variables. Significantly, the first instances of this science bureaucracy creation occur among this group, suggesting that demand-driven explanations may fit some of the earliest adopters of science policy. Britain, the first adopter, clearly created its Department of Science and Industrial Research in 1915 for security reasons to counter German advances in chemicals and machinery that were directly supporting the German war effort.12 The establishment of the National Science Foundation in the United States in 1950 was explicitly related to concerns about military and industrial competitiveness and was strongly influenced by the creation of the atom bomb.13 French science policymaking, as chronicled by 12. Alter, The Reluctant Patron. See also Roy McLeod and E. Kay Andrews, "The Origins of the D.S.I.R.: Reflections on Ideas and Men, 1915-1916," Public Administration, vol. 48, no. 1,1970, pp. 23-48; and Ian Varcoe, "Scientists, Government, and Organized Research in Great Britain 1914-1916," pp. 192-216. The United Kingdom is not included in the quantitative analysis above because science data for that country for 1915 are unavailable. 13. Dickson, The New Politics of Science. See also J. Merton England, A Patron for Pure Science: The National Science Foundation's Formative Years, 1945-1957 (Washington, D.C.: National Science Foundation, 1982); N. Dupree, Science in the Federal Government; Bruce Smith, American Science Policy Since World War II (Washington, D.C.: Brookings Institution, 1990); and U.S. Congress, House Committee on Science and Technology, Task Force on Science Policy, A History of Science Policy in the United States, 1940-1985, Science Policy Study Background Report, no. 1, 99th Congress, 2d sess., 1986, serial R