¹ Some of the reasons for adopting this premise may be summarized as follows: (a) Latin American nations, like other developing nations, are less constrained in their actions by mass public opinion or by bureaucracies than are developed nations; (b) their leaders usually adopt a distinctive style of policy making that emphasizes the fit between policies and ideology which is often lacking in the incrementalist politics of developed countries; (c) whenever there is a change in leadership in Latin American nations, there is usually a change in domestic and foreign policies which reflects the beliefs of the new leaders. In support of point (a), I would simply argue that there are very few liberal democratic regimes surviving in Latin America at the moment, and that if there were much bureaucratic politics in Latin America, a literature about it would have developed by now. The only works that I know of which discuss the constraining effects of interagency conflict or bargaining on state politics in Latin America are Cleaves, Peter S., Bureaucratic Politics and Administration in Chile (Berkeley: University of California Press 1974)Google Scholar, and Tugwell, Franklin, The Politics of Oil in Venezuela (Stanford: Stanford University Press 1975).Google Scholar The idea, point (b), that beliefs and ideology are particularly important in determining the policies of Latin American elites —because of the lack of public pressures and the distinctive problem-solving style of those elites —was first discussed by Hirschman, Albert O., in Journeys Toward Progress (New York: Twentieth Century Fund 1961), 231.Google Scholar In support of point (c), I could cite the cases of the change in foreign policy which accompanied the elections of Janio Quadros in Brazil, Salvador Allende in Chile, Rafael Caldera in Venezuela, and Juan Peron in Argentina. I could also cite many cases of changes in policy produced by extralegal changes in regimes. This matter has not received sufficient research, however, and it remains to be seen whether changes in policy can best be explained by the changes in belief systems that sometimes accompany changes in government personnel, or by other factors.

² I will define cognitive maps as sets of causal beliefs or assertions, though some scholars use the term to stand for general beliefs about the fundamental characteristics of some aspect of a physical or social environment, whether or not such beliefs are causal. I am indebted to Ole Holsti for this point. Other scholars are interested in sets of causal assertions as they are used in discussions and arguments among groups of policy makers. Such sets are more correctly called “rhetorical maps”, because they do not assume that the individuals believe the causal assertions that they make in order to persuade others to go along with a particular policy. See Axelrod, Robert, “Argumentation in Foreign Policy Decision Making”, paper delivered at the Annual Meeting of the American Political Science Association, Chicago, September 1976.Google Scholar

³ In other words, I am not denying the probable effects of noncognitive variables on the policy choices of individuals; but I am arguing, as Greenstein, Fred does in Personality and Politics (New York: Norton 1975), xvii–xix,Google Scholar that the behavior of political actors cannot always be explained in terms of “situational” determinants and that therefore psychological explanations may sometimes be useful.

⁴ Other types of causal beliefs may be assumed to exist and may be incorporated into the empirical study of cognitive maps. See Hart, , “Comparative Cognition: Politics of International Control of the Oceans”, in Axelrod, Robert, ed., The Structure of Decision (Princeton: Princeton University Press 1976).Google Scholar

⁵ See Margaret Wrightson, “The Documentary Coding Method”, in Axelrod (fn. 4).

⁶ Towards a Great Venezuela: Nationalization of the Oil Industry (Caracas: Oficina Central de Informacion 1975), 11.Google Scholar

⁷ Jervis, Robert, “Consistency in Foreign Policy Views”, in Merritt, Richard L., ed., Communication in International Politics (Urbana: University of Illinois Press 1972)Google Scholar; Axelrod (fn. 4), “The Analysis of Cognitive Maps”.

⁸ See the introductory chapter by Robert Axelrod in Axelrod (fn. 4).

⁹ Hirschman (fn. 1), esp. chap. 4.

¹⁰ See fn. 1.

¹¹ Since the actions of OPEC in 1973, several books have appeared which underline this theme. See, for example, The Americas in a Changing World (New York: Quadrangle Books 1975)Google Scholar; Einaudi, Luigi, ed., Beyond Cuba (New York: Crane Russak 1974)Google Scholar; and Hellman, Ronald and Rosenbaum, H. J., eds., Latin America: The Search for a New International Role (New York: Halsted Press 1975).Google Scholar

¹² Towards a Great Venezuela … (in. 6).

¹³ Ibid., 23.

¹⁴ Ibid., 10.

¹⁵ , in “Panel: Investment Policy”, in Baer, Werner and Kerstenetzky, Isaac, eds., Inflation and Growth in Latin America (New Haven: Yale University Press 1964).Google Scholar

¹⁶ Campos, Reflections on Latin American Development (Austin: University of Texas Press 1976)Google Scholar; Campos, “A Retrospect over Brazilian Development Plans”, in Ellis, Howard S., ed., The Economy of Brazil (Berkeley: University of California Press 1969).Google Scholar

¹⁷ See Handler, Bruce, “Flying High in Rio”, New York Times Magazine, June 8, 1975Google Scholar, pp. 16–17, 87–94; Simonsen, Mario Henrique and de Oliveira Campos, Roberto, A nova economia brasileira (Rio de Janeiro: Livraria Jose Olympio 1974).Google Scholar

¹⁸ Campos (fn. 15), 461.

¹⁹ Marchant, Anyda, “The Political and Legal Framework of Brazilian Life”, in John Saunders, ed., Modern Brazil (Gainesville: University of Florida Press 1971), 115.Google Scholar

²⁰ de Lyra Tavares, Aurelio, “The Brazilian Army and Politics”, in Sigmund, Paul, ed. and transl., Models of Change in Latin America (New York: Praeger 1970)Google Scholar.Tavares's book was originally published in Recife in 1968.

²¹ See Stepan, Alfred, ed., Authoritarian Brazil: Origin, Politics, and Future (New Haven: Yale University Press 1973)Google Scholar; Stepan, , The Military in Politics (Princeton: Princeton University Press 1971).Google Scholar

²² One might want to control for the size of the map by controlling for the number of conceptual variables. Since the maps have roughly the same number of variables (see Table 6), controlling for the number of variables does not radically change the results. One might question the decision to count those goal variables that affect U_BRZ together with those that affect U_ARMY for General Tavares while not counting variables that affect U_LA for President Pérez. It seemed to be appropriate because I was surer of Tavares's equation of the interests of the Brazilian Army with those of Brazil than I was of Pérez's equation of the interests of Venezuela with those of the rest of Latin America.

²³ This should be seen in the light of the “depoliticization” of PETROVEN being attempted by Pérez. Perez appointed General Alfonso Ravard, former head of the Corporación Venezolana de Guayana (and a man known for his integrity), to be the head of PETROVEN. See “Oil, Democracy and Development”, The Economist, December 27, 1975, p. 12 of “Survey on Venezuela”.

²⁴ “Artur da Costa e Silva”, Current Biography Yearbook 1967 (New York: H. W. Wilson 1967), 87.Google Scholar

²⁵ For an example of the predictive power of cognitive maps, see Matthew Bonham and Michael Shapiro, “Explanation of the Unexpected: The Syrian Intervention in Jordan in 1970”, in Axelrod (fn. 4).

²⁶ See ibid., throughout.

²⁷ The comparative foreign policy approach allows for variance in perceptions at the national level in explaining the actions of the state. Yet it does so only in a grudging manner, so that perceptions remain secondary, and “structural” variables are primary. See the introductory chapter by the editor in Rosenau, James N., ed., Comparing Foreign Policies (Beverly Hills, Calif.: Sage 1974).Google Scholar