¹ I consider the pioneering work in this field to be Deutsch's, Karl W.Political Community at the International Level: Problems of Definition and Measurement (Garden City, N.Y: Doubleday & Co., 1954)Google Scholar . Deutsch raised all the major questions and introduced many of the concepts that still preoccupy and guide the research effort.

² For an identification of highly or closely linked national polities with the notion of “integrated international system” see O'Leary, Michael K., “Linkages Between Domestic and International Politics in Underdeveloped Nations,” in Rosenau, James (ed.), Linkage Politics: Essays on the Convergence of National and International Systems (New York: The Free Press, 1969), pp. 324–346Google Scholar. See also the argument of W. F. Hanrieder to the effect that regional integration movements provide a key way of isolating linkages. Hanrieder, Wolfram F., “Compatibility and Consensus: A Proposal for the Conceptual Linkage of External and Internal Dimensions of Foreign Policy,” American Political Science Review, 12 1967 (Vol. 61, No. 4), pp. 972–973CrossRefGoogle Scholar.

This approach is rejected by Rosenau, James N., “Compatibility, Consensus, and an Emerging Political Science of Adaptation,” American Political Science Review, 12 1967 (Vol. 61, No. 4), pp. 983–988CrossRefGoogle Scholar. He argues that the questions addressed in integration frameworks of study cannot be combined with questions which hinge on adaptation. This is a more restrictive definition of adaptation and of integration than is necessary or desirable.

³ I wish to acknowledge my great debt to Ivan Vallier in seeking to come to grips with these issues even though I fall short of attaining his ideal.

⁴ For a representative agenda of topics and articles illustrating the rubrics regionalism, regional cooperation, regional organization, regional systems, etc., see Nye, Joseph S. Jr, (ed.), International Regionalism: Readings (Boston: Little, Brown and Co., 1968)Google Scholar .

⁵ This stab at a definition differs appreciably from the working definition of integration involving a shift in the loyalties of political actors with which I had worked previously. In abandoning the earlier definition I am expressing my agreement with the criticism Nye leveled at my own and at Deutsch's definitions. See Nye, Joseph S., “Comparative Regional Integration: Concept and Measurement,” International Organization, Autumn 1968 (Vol. 22, No. 4), pp. 856—858CrossRefGoogle Scholar .

⁶ These definitions are too well known to require extensive recapitulation. For discussions see Etzioni, Amitai, Political Unification: A Comparative Study of Leaders and Forces (New York: Holt, Rinehart and Winston, 1965), chapter 1Google Scholar ; die process of “political unification” is described in chapter 2. Jacob, Philip E. and Toscano, James V. (ed.), The Integration of Political Communities (Philadelphia: Lippincott, 1964)Google Scholar , chapter 1; chapters 2 and 3 of this volume contain an admirable summary by Karl Deutsch of the communications-transactions approach to integration. The federalist case is made in a sophisticated fashion by Sidjanski, Dusan, Dimensions européennes de la science politique: Questions méthodologiques et programme de recherches (Paris: Librairie géneYale de droit et de jurisprudence, 1963)Google Scholar.

⁷ Schmitter, Philippe C., “La dinámica de contradicciones y la conducción de crisis en la integración centroamericana,” Revista de la integratión, 11 1969 (No. 5), pp. 140–147Google Scholar, offers the notion of self-encapsulation which highlights the problem of identifying organizational success with progressive integration.

⁸ Discussion of the “Inter-American system,” for instance, usually includes descriptions of all regional organizations, programs, and commitments of the member states of the Organization of American States (OAS) as well as of the relations between separate organizations. The “European system” is sometimes taken as a descriptive term for the totality of Western European organizations and programs without regard to any impact on the reciprocal integration of the member nations.

The level of political integration is the main characteristic that distinguishes political communities from other political systems. System is the more encompassing concept, indicating that changes in the action of one (or more) unit(s) affect actions in one (or more) other units, and that these latter changes in turn have repercussions on the unit or units in which or from which the change was initiated.… In short, units of systems are interdependent; members of communities are integrated. [Etzioni, p. 6.]

To confuse matters further the notion of “system” is sometimes assimilated into that of a “model,” a recurrent and/or typical way in which integration is thought to go forward. This usage is reserved for the discussion of how variables are to be organized, not for the delimitation of the field.

⁹ Binder, Leonard, “The Middle East as a Subordinate International System,” World Politics, 04 1958 (Vol. 10, No. 3), pp. 408–429CrossRefGoogle Scholar. Brecher, Michel, “International Relations and Asian Studies: The Subordinate State System of Southern Asia,” World Politics, 01 1963 (Vol. 15, No. 2), pp. 213–235CrossRefGoogle Scholar.

¹⁰ This is shown in Russett, Bruce M., International Regions and the International System: A Study in Political Ecology (Chicago: Rand McNally & Co., 1967)Google Scholar. See the comment on such treatments of regionalism in Young, Oran R., “Professor Russett: Industrious Tailor to a Naked Emperor,” World Politics,04 1969 (Vol. 21, No. 3), pp. 486–511CrossRefGoogle Scholar.

¹¹ The list of studies dealing with aspects of law, economics, military strategy, current diplomacy, and ideology is immense. For an excellent selection of recent studies see Nye, pp. 73–74, 145–146, 282–283, 428–429. For Europe in particular also see European Community Institute for University Studies (Brussels), University Studies on European Integration, 1969 (No. 5)Google Scholar. For an exhaustive list of documents relating to economic cooperation in the third world see Wionczelc, Miguel S. (ed.), Economic Cooperation in Latin America, Africa, and, Asia (Cambridge, Mass: M.I.T. Press, 1969)Google Scholar. The following studies, in addition to the ones included in these bibliographies, are highly relevant to the conclusions here reported: Friedrich, Carl J. (ed.), Politische Dimensionen der europäischen Gemeinschaftsbildung (Cologne: Westdeutscher Verlag, 1968)CrossRefGoogle Scholar; Zellentin, Gerda (ed.), Formen der Willensbildung in den europäischen rganisationen (Frankfurt: Athenäum Verlag, 1965)Google Scholar; Hammerich, Kai Ewerlöf, L'union des industries de la Communauté européenne dans le Marché commun (Stockholm: Federation of Swedish Industries, 1969)Google Scholar; Feld, Werner, The European Common Market and the World (Englewood Cliffs, N.J: Prentice-Hall, 1967)Google Scholar; Study Group of the Graduate Institute of International Studies (Geneva), The European Free Trade Association and the Crisis of European Integration(London: Michael Joseph, 1968)Google Scholar; Aubrey, Henry G., Atlantic Economic Cooperation (New York: Frederick A. Praeger, 1967)Google Scholar; Gregg, Robert W. (ed.), International Organization in the Western Hemisphere (Syracuse, N.Y: Syracuse University Press, 1968)Google Scholar; Sidjanski, Dusan, Dimensiones instituciqnales de la integración latino, americana (Buenos Aires: Instituto para la Integración de América Latina, 1967)Google Scholar; Pruque, Armando, Siete años de actión de la ALALC (Buenos Aires: Instituto para la Integration de América Latina, 1968)Google Scholar; Kramer, Francisco Villagrán, Integración económica centroamericana: Aspectos sotiales y politicos (Estudios Universitarios, Vol. 4) (Guatemala City: University of San Carlos, 1967)Google Scholar; Hansen, Roger D., Central America: Regional Integration and Economic Development (National Planning Association Studies in Development Progress, No. 1) (Washington: National Planning Association, 1967)Google Scholar; Rothchild, Donald (ed.), Politics of Integration: An East African Documentary (EAPH Political Studies 4) (Nairobi: East African Publishing House, 1968)Google Scholar.

¹² This distinction was first clearly elaborated by Schmitter in the article cited in footnote 7. It may be observed also in Africa and in Eastern Europe. For a European example see Pfaltzgraff, Robert L. and Deghand, James L., “European Technological Collaboration: The Experience of the European Launcher Development Organization (ELDO), Journal of Common Market Studies, 09 1968, (Vol. 7, No. 1), pp. 22–34CrossRefGoogle Scholar.

¹³ The definitive study probing the ability of NATO to trigger integrative results outside the narrow military field is Francis Beer, A., Integration and Disintegration in NATO: Processes of Alliance Cohesion and Prospects for Atlantic Community(Columbus: Ohio State University Press, 1969)Google Scholar. Modelski makes the same point about SEATO and Slater about the OAS (see Nye for full citation). See also Korbonski, Andrzej, “The Warsaw Pact,” International Conciliation, 05 1969 (No. 573)Google Scholar, and Korbonski's contribution to this volume.

¹⁴ Gruhn, Isebill, Functionalism in Africa: Scientific and Technical Cooperation (unpublished Ph.D. dissertation, University of California, Berkeley, 1967)Google Scholar. Sewell, W. R. Derrick and White, Gilbert F., ”The Lower Mekong,” International Conciliation, 05 1966 (No. 558)Google Scholar; Rosberg, Carl G. Jr, with Segal, Aaron, “An East African Federation,” International Conciliation, 05 1963 (No. 543)Google Scholar. Jalloh, Abdul A., The Politics and Economics of Regional Political Integration in Central Africa(unpublished Ph.D. dissertation, University of California, Berkeley, 1969)Google Scholar.

¹⁵ In Western Europe the Greek response to the strong condemnation on the part of the Council of Europe's organs confirms this finding. McNulty, A. B., “Stock-Taking on the European Convention on Human Rights,” Council of Europe document DH(68)7, 1968Google Scholar. The Inter-American Commission on Human Rights has been active and innovating in the face of governmental indifference and is now presenting the OAS Council with a final draft for a stronger and more permanent hemispheric machinery. Scheman, L. Ronald, “The Inter-American Commission on Human Rights,” American Journal of International Law, 04 1965 (Vol. 59, No. 2), pp. 335–343CrossRefGoogle Scholar; Cabranes, José A., “The Protection of Human Rights by the Organization of American States,” American Journal of International Law, 10 1968 (Vol. 62, No. 4), pp. 889–908CrossRefGoogle Scholar.

¹⁶ The distinction between “scope” and “level” in the discussion o£ the spillover mechanism is first made by Schmitter, Philippe C. in “Three Neo-Functional Hypotheses about International Integration,” International Organization, Winter 1969 (Vol. 23, No. 1), pp. 161–166CrossRefGoogle Scholar.

¹⁷ See especially Inglehart, Ronald, “An End to European Integration?” American Political Science Review, 03 1967 (Vol. 61, No. 1), pp. 91–105CrossRefGoogle Scholar. Inglehart's analysis received powerful confirmation from a poll of French opinion conducted in 1968; see “Les Français et l'unification politique de l'Europe d'après un sondage de la SOFRES,” Revue Française de science politique, 02 1969 (Vol. 19, No. 1), pp. 145–170Google Scholar.

¹⁸ For examples see Lagos, Gustavo, “The Political Role of Regional Economic Organizations in Latin America,” Journal of Common Market Studies, 07 1968 (Vol. 6, No. 4), pp. 291–309CrossRefGoogle Scholar.

For an argument that distribution crises (in East Africa and Central America) must be expected to arise in such settings, thus inevitably politicizing negotiations and institutions dealing with economic welfare, see Hansen, Roger D., “Regional Integration: Reflections on a Decade of Theoretical Efforts,” World Politics, 01 1969 (Vol. 21, No. 2), pp. 257—270CrossRefGoogle Scholar. This feature, among others, leads Hansen to conclude that the neo-functional approach cannot be used for the analysis of regional integration among less developed countries. Hansen finds the tendency toward distribution crises to be an instance of “instant politicization” even though in Central America it did not occur until the integration process had gone on for seven years. In Europe such a development was avoided, he says, because the dominance of the hidden hand of automatic market forces assured more or less equitable distribution of benefits (an argument also made by Krause, Lawrence, European Economic Integration and the United States [Washington: Brookings Institution, 1967])Google Scholar. The makers of monetary policy, those who fashion subsidies for miners, railroads, and farmers, and the businessmen who conclude specialization agreements would be interested to learn this! On the contrary, serious distribution crises were avoided because of a consistent bargaining style stressing package deals permitting the continuation of special benefits not to be expected if an automatic and general free market were to operate. While Hansen is quite right in calling attention to the limits on the spillover from economic policymaking into political unification (see proposition 5 above), he carries the criticism much too far by downgrading the importance of “expansive” bargaining styles and the institutions in which the styles take shape, whether supranational or not.

¹⁹ Kaiser, Karl, “The U.S. and the EEC in the Atlantic System: The Problem of Theory,” Journal of Common Market Studies, 06 1967 (Vol. 5, No. 4), pp. 388–425CrossRefGoogle Scholar, and Kaiser, Karl, “The Interaction of Regional Subsystems: Some Preliminary Notes on Recurrent Patterns and the Role of Superpowers,” World Politics, 10 1968 (Vol. 21, No. 1), pp. 84–107CrossRefGoogle Scholar. See Etzioni, pp. 44–50, for a related treatment of “external elites.” Denham, Robert E., “The Role of the U.S. as an External Actor in the Integration of Latin America,” Journal of Common Market Studies, 03 1969 (Vol. 7, No. 3), pp. 199–216CrossRefGoogle Scholar. Andrén, Nils, “Nordic Integration,” Cooperation and Conflict, 1967 (Vol. 1), pp. 1–25CrossRefGoogle Scholar.

²⁰ For the classic statements of this position see Hoffmann, Stanley, “Discord in Community: The North Atlantic Area as a Partial International System,” International Organization, Summer 1963 (Vol. 17, No. 3) PP. 521–549CrossRefGoogle Scholar, and Hoffmann, Stanley, “Obstinate or Obsolete? The Fate of the Nation-State and the Case of Western Europe,” Daedalus, Summer 1966 (Vol. 95, Nos. 3–4), pp. 862–916Google Scholar. Hansen treats the exogenous variable in this fashion, considering it the most “compelling” of all and using it as the most salient for predicting a reasonably rosy future for economic unions of less developed countries. He also explicitly accepts the dichotomy of “high” versus “low” politics, treating economic welfare issues in the third world as an instance of “high” politics. Hansen, , World Politics, Vol. 21, No. 2, pp. 268–271, 246–250Google Scholar.

²¹ Illustrative of this practice is the massive work by Bowie, Robert R. and Friedrich, Carl J. (ed.), Studies in Federalism (Boston: Little, Brown and Co., 1954)Google Scholar, which contains research and theorizing done on behalf of the then functioning European ad hoc Parliamentary Assembly.

²² Taylor, Paul, “Concept of Community and the European Integration Process,” Journal of Common Market Studies, 12 1968 (Vol. 7, No. 2), pp. 83–101CrossRefGoogle Scholar. This argument misses the essential distinctions between ends and means and between activist/actor and theorist/observer. To illustrate: Jean Monnet and Walter Hallstein are “federalists” in the sense that they hope to create a united Europe with more or less federal institutions; they are “functionalists” in the sense that they do not believe in constitutional conventions and elaborate institutional schemes because they prefer to initiate common policies and arouse new client groups which will eventually result in a federal regime. Federalist theorists believe in the federal end but may be willing to use functional means; federalist activists despise the recourse to functional means. Neo-functionalist theorists are concerned with the end postulated by the actors only to the extent that they are preoccupied with understanding why and how actual integrative outcomes occur. Neo-functional theory and federal theory, therefore, are by no means similar.

²³ As examples of ideological federalists we cite the names of Denis de Rougemont, Hans Nord, Henri Brugmans. See also Marc, Alexandre, Europe: Terre décisive (Paris: La Colombe, 1959)Google Scholar. Another group of federal activists, in disappointment, has turned to a species of neo-functionalism for ideological sustenance; see Olivi, Bino, L'Europa difficile (Milan: Edizioni di Communitá, 1964)Google Scholar, and Spinelli, Altiero, The Eurocrats: Conflict and Crisis in the European Community (Baltimore: Johns Hopkins Press, 1968)Google Scholar. A new twist is given to this complex relationship by the emergence of a young group of federalists anxious to work with functional institutions to achieve the radical democratization of European society. They publish the monthly journal Agenor.

²⁴ This passage from Friedrich's most recent work speaks for itself:

The review of selected issues in contemporary federal relations has… shown that federalism is more fully understood if it is seen as a process, an evolving pattern of changing relationships, rather than a static design regulated by firm and unalterable rules. This finding ought not to be misunderstood as meaning that the rules are insignificant; far from it. What it does mean is that any federal relationship requires effective and built-in arrangements through which these rules can be recurrently changed upon the initiative and with the consent of the federated entities. In a sense, what this means is that the development (historical) dimension of federal relationships has become a primary focal point, as contrasted with the distribution and fixation of jurisdictions (the legal aspect). In keeping with recent trends in political science, the main question is: What function does a federal relationship have?—rather than: What structure?

Friedrich, Carl J., Trends of Federalism in Theory and Practice (New York: Frederick A. Praeger, 1968), p. 173)Google Scholar. For a similar treatment of federalism in Africa, Malaysia, and the West Indies see Franck, Thomas M. (ed.), Why Federations Fail (New York: New York University Press, 1968)Google Scholar.

²⁵ The work of Donald Puchala, as represented by his contribution to this volume, though part of this approach, is not based on all of these assumptions and makes a number of allowances to the judgments expressed here.

²⁶ See Deutsch, Karl W., and others, France, Germany and the Western Alliance: A Study of Elite Attitudes on European Integration and World Politics (New York: Charles Scribner's Sons, 1967), especially p. 218Google Scholar.

²⁷ See the suggestion by Karl Kaiser that the distinction has relevance only to the attitudes of actors, not to the intrinsic character of the issue area or policy. Actors can shift back and forth as to the importance they attribute to any given issue area—with obvious implications for integration theory. The empirical problem for us is to be able to specify when and how such attitudinal shifts occur, an aspect of a theory of “learning” we shall discuss below. This is quite different from the eternally mystic qualities Hoffmann sees in these distinctions. Kaiser, , Journal of Common Market Studies, Vol. 5, No. 3, pp. 401–402Google Scholar.

²⁸ But we have learned, too. Scholars are less prone to circular definitions between independent and dependent variables than they were a decade ago when a “community” was held to require a “sense of community” as a precondition and an integrating region was thought to demand a preexisting propensity to integrate. Still there are more indeterminate variables now than desirable. We certainly should discourage the tendency to add “fudge variables” whenever the standard ones do not seem to do the explanatory trick. This means that we ought to stop talking about “functional equivalents,” “catalysts,” ”federalizers,” “compellingness,” “high politics,” and similar mythical animals.

²⁹ Possible exceptions are the (mostly unsuccessful) noncoercive unions of new states attempted in the last two decades, as analyzed in Franck and in Etzioni. At best, these experiences help to falsify certain theoretical postulates without explaining how a successful federal union can be achieved.

³⁰ For examples of such efforts see Nye, , International Organization, Vol. 22, No. 4Google Scholar; Jacob, Philip and Teune, Henry, “The Integrative Process: Guidelines for Analysis of the Basis of Political Community,” in Jacob, and Toscano, , pp. 1–45Google Scholar.

A similar approach is suggested in the quantification of twelve independent variables, which in the aggregate are expected to yield an outcome called “automatic politicization,” that could be considered a proxy variable for political community. See Barrera, Mario and Haas, Ernst B., “The Operationalization of Some Variables Related to Regional Integration: A Research Note,” International Organization, Winter 1969 (Vol. 23, No. 1), pp. 150—160Google Scholar; also the response by Schmitter, Philippe C., “Further Notes on Operationalizing Some Variables Related to Regional Integration,” in International Organization, Spring 1969 (Vol. 23, No. 2), pp. 327–336CrossRefGoogle Scholar. How politicization is sequentially linked to other developments during the integration process is set forth by Schmitter, , International Organization, Vol. 23, No. 1, pp. 161–166CrossRefGoogle Scholar. If interdependence along some highly salient dimension is considered a proxy variable for the outcome, the work of Reinton is most suggestive. See Reinton, Per Olav, “International Structure and International Integration,” Journal of Peace Research,1967 (No. 4), pp. 334–365CrossRefGoogle Scholar.

³¹ See Nye, , International Organization, Vol. 22, No. 4, pp. 864–875Google Scholar. I disagree with Nye's argument that the notion of integration ought to be “disaggregated” into economic, social, and political components without making any premature judgments as to how these separate ranges relate to each other causally. To avoid making causal assumptions and positing hypotheses merely sidesteps once more the difficulty of finding a dependent variable. Political integration, if that is what we are concerned about, is more important than economic and social trends; these are important because we think they are causally connected with political integration. Disaggregation helps in the attainment of operational accuracy but it hinders the removal of fuzzy dependent variables unless orders of causal priority are thought through. Humpty-Dumpty must be put together again, too, as it is in Nye's contribution to this volume.

³² Lindberg, Leon and Scheingold, Stuart, Europe's Would-Be Polity: Patterns of Change in the European Community (Englewood Cliffs, N.J: Prentice-Hall, 1970), chapter 4Google Scholar. For a most complete and perceptive list of possible measures and indicators of progress toward the achievement of any one of these three conditions (or of regress) see Lindberg's contribution to this volume. It should be stressed, however, that these specifications still do not add up to the definition of one or several dependent variables in the sense of my argument. They have the enormous merit of providing measures specifically related to a series of independent and intervening variables while being pinpointed in one causative direction. In short, these measures do not substitute disaggregation of a fuzzy dependent variable for more sophisticated reaggregation.

³³ These types were suggested by, and correspond to, certain constructs in organization theory, particularly by the literature on organizational complexity. Moreover, each type is derived from set theory and is capable of mathematical manipulation, thereby perhaps contributing to possibilities of computer simulation. Thus, regional state corresponds to a set-theory tree, regional commune to a full matrix, and asymmetrical overlap to a semilattice. I am greatly indebted for diese ideas to Todd LaPorte and John Ruggie.

³⁴ This way of seeing concepts and data involves a change in perspective on my part as to what constitutes a variable and a method of linking variables, as distinguished from a validated summary of a recurring process expressed by a conceptual label such as “spillover.” In short, I no longer accept as sufficiently well-explicated and “placed” the “variables” that appear in Haas, Ernst B. and Schmitter, Philippe C., “Economics and Differential Patterns of Political Integration: Projections about Unity in Latin America,” International Organization, Autumn 1964 (Vol. 18, No. 4), pp. 705–737CrossRefGoogle Scholar, and in Barrera, and Haas, , International Organization, Vol. 23, No. 1, pp. 150–160CrossRefGoogle Scholar.

A given action path is characterized by a certain kind of learning, a certain pattern of responsiveness, a certain degree of spillover, whether in scope or level. Michel Crozier has suggested a new term which seems appropriate for summing up the kind of process that we ought to use more as a device for linking variables: apprentissage institutionnel. It can be used to light up the interconnections between a pattern of transaction, disappointment, declining responsiveness, bargaining, and the eventual redefinition of common or converging interests.

³⁵ Etzioni, Nye, Haas, and Schmitter have all indulged in this practice. It is raised to a point of faith in Hansen, , World Politics, Vol. 21, No. 2, pp. 270–271Google Scholar. To those who argue that comparative studies of integration could still proceed if we had one model for industrialized and one for underdeveloped regions it must be pointed out that our findings suggest a different mix of variables at work in explaining the maintenance of regional economic arrangements in Central America, South America, East Africa, and Equatorial Africa, not to mention Eastern Europe.

³⁶ Puchala, Donald J., “The Pattern of Contemporary Regional Integration,” International Studies Quarterly, 03 1968 (Vol. 12, No. 1), pp. 38–64CrossRefGoogle Scholar. Among social psychologists there are exceedingly few whose assumptions and methods can be readily related to the study of social learning as mediated through experiences in bargaining, defining, and redefining interests because the bulk of the psychological literature is organized around the trust-love-friendship-peace versus distrust-hate-enmity-war dichotomy. For notable and important exceptions see Katz, Daniel, “Nationalism and Strategies of International Conflict Resolution,” in Kelman, Herbert C. (ed.), International Behavior: A Social-Psychological Analysis (New York: Holt, Rinehart and Winston [for the Society for the Psychological Study of Social Issues], 1965)Google Scholar, as well as Kelman's own work, as summarized by himself in ibid., chapter 16. For a discussion of stimulus-response theory in this context see Teune, Henry, “The Learning of Integrative Habits,” in Jacob, and Toscano, , pp. 247–282Google Scholar. The work of Chadwick Alger on United Nations diplomats illustrates the same set of assumptions. Iklé, Fred with Nathan Leites discusses the modification of actor utilities in “Political Negotiations as a Process of Modifying Utilities,” Journal of Conflict Resolution, 03 1962 (Vol. 6, No. 1), pp. 19–28CrossRefGoogle Scholar. In purely game-theory term the work of John Harsanyi is the most suggestive with respect to the exploration of this area.