¹ Durkheim, , The Rules of Sociological Method, ed. by Catlin, George E.G. (New York: Free Press, 1964), 116, 103, xlvii, 112.Google Scholar

² Cf. Durkheim, , The Division of Labor in Society, trans, by Simpson, George (New York: Free Press, 1964)Google Scholar, wherein this model was first developed. It should be noted that for Durkheim the designation “social fact” does not refer to all phenomena that take place within society, but only to those that exist exterior to individuals, are not subject to modification by a simple effort of will on the part of individuals, and function as a constraint on individual behavior (fn. 1, chap. 1).

³ Lévi-Strauss, , Structural Anthropology (Garden City, N.Y.: Doubleday Anchor, 1967)Google Scholar, esp. chaps. 2 and 15.

⁴ First published in Comparative Studies in Society and History, XVI (September 1974); reprinted in Wallerstein, , The Capitalist World Economy (New York: Cambridge University Press, 1979)Google Scholar; references here are to the latter source.

⁵ Ibid., 3.

⁶ Ibid., 5.

⁷ See the penetrating critique along these lines by Brenner, Robert, “The Origins of Capitalist Development: A Critique of Neo-Smithian Marxism,” New Left Review, No. 104 (July-August 1977).Google Scholar

⁸ “Capitalism has been able to flourish precisely because the world economy has had within its bounds not one but a multiplicity of political systems” (Wallerstein, , The Modern World System, I [New York: Academic Press, 1974], 348)Google Scholar; this structure in turn is maintained by the functional needs of capitalism, specifically the high economic costs of political imperium (fn. 4, p. 32), and the tendency of capitalists to resort to the instrumentalities of their respective states so as to enhance their international competitive position (fn. 4, pp. 19–20).

⁹ Waltz's critical review of the literature has generated a sizable secondary literature of rejoinders and counteroffensives, of which the most offensive no doubt is by Kaplan, Morton A., “The Genteel Art of Criticism, or How to Boggle Minds and Confooz a Discipline,” in Kaplan, , ed., Towards Professionalism in International Theory (New York: Free Press, 1979).Google Scholar More generous readings may be found in Hoffmann, Stanley, Primacy or World Order (New York: McGraw Hill, 1978), 146–47Google Scholar, and Rosecrance, Richard, “International Theory Revisited,” International Organization, XXXV (Autumn 1981).Google Scholar

¹⁰ Waltz acknowledges that Wallerstein has also developed a systemic theory, but rejects Wallerstein's claims for its logical priority (p. 38). In principle, Waltz allows for the possibility of co-equality, but in deed he argues for the priority of the international polity, as we shall see below.

¹¹ Zolberg, Aristide R., “Origins of the Modern World System: A Missing Link,” World Politics, XXXIII (January 1981).Google Scholar

¹² In addition to the references cited above (fn. 9), see also the review by Fox, William T.R., in American Political Science Review, Vol. 74 (June 1980).Google Scholar

¹³ Durkheim is referenced four times in the index of Waltz's book; in a footnote (p. 115), Waltz promises to elaborate on Durkheim's typology of social ordering principles in a future work.

¹⁴ Durkheim (fn. 1), xlvii, 103.

¹⁵ There has been inordinate confusion about these distinctions, stemming largely from the way in which the so-called levels-of-analysis problem is usually interpreted. As originally defined, it simply says that the international system and national states constitute two different levels of analysis in the study of international relations. (David Singer, J., “The Level-of-Analysis Problem in International Relations,” in Knorr, Klaus and Verba, Sidney, eds., The International System: Theoretical Essays [Princeton: Princeton University Press, 1961].)Google Scholar But that isn't the whole of it. The two terms, international and system, are frequently conjoined, and the assumption is made that any model expressing international factors is automatically a systemic model. However, as Waltz shows (chaps. 3–4), the norm—even when systems language is employed—is to explain international phenomena in terms of units and their interactions, not in terms of systems as ontologically distinct totalities.

¹⁶ The distinction here is between generative and descriptive structures. Descriptive structures are simply abstract summaries of patterned interactions within a system. For example, national capabilities are measured, and hierarchies of state power are depicted. Trade and capital flows are measured, and hierarchies of economic power are adduced. Most uses of the concept of structure in contemporary international relations theory employ this meaning', the structural theories of Stephen Krasner and Johan Galtung offer a representative sampling. In the realm of generative structures, the concern is “with principles, not things” (Leach, Edmund, Rethinking Anthropology [London: Athlone Press, 1961], 7).Google Scholar The object is to discover the underlying principles that govern the patterning of interactions, to infer their syntax. Saussurean linguistics probably was the first self-conscious expression of generative structuralism in the social sciences, which has transformed the study of linguistics and cultural anthropology. For useful surveys, see Glucksman, Miriam, Structuralist Analysis in Contemporary Social Thought (London: Routledge & Kegan Paul, 1974)Google Scholar, and Kurzweil, Edith, The Age of Structuralism (New York: Columbia University Press, 1980).Google Scholar

¹⁷ Waltz thus rejects the conventional view that a balance-of-power system requires a minimum number of effective actors larger than two—preferably five, so that one can act as balancer. This, he points out, “is more a historical generalization than a theoretical concept” (p. 164). In fact, balancing takes place in a bipolar world no less than in a multipolar world, except that the methods of balancing are largely internal rather than external.

¹⁸ The relationship between number and stability is not perfectly continuous, since, ceteris paribus, a world of three great powers is thought to be less stable than a world of four, though it may be so unstable that it inevitably resolves into bipolarity in any case (p. 163). Note also that the emergence of two opposing alliances in a multipolar world does not transform it into bipolarity; by the same token, the loss of an ally in a bipolar system does not transform it into multipolarity. Polarity is a structural attribute of systems, measured by the number of great powers, whereas alliances are process-level phenomena that serve as one of the means by which states pursue their interests (pp. 169–70).

¹⁹ One vexing problem does require special mention, however. The absence of system-wide wars is not the only definition of stability employed by Waltz. He also uses the term in the economists' sense—of the system returning to a prior or corresponding point of equilibrium after a disturbance. Confusion ensues because either bipolarity or multipolarity comes out being more stable, depending upon the definition of stability, and Waltz is inconsistent and often unclear in his usage. As I understand him, multipolarity is more stable in the dynamic equilibrium sense (see p. 162, on the relative durability of the multipolar era in the modern state system), and bipolarity is more stable in the sense of the absence of system-wide wars (pp. 170–76, and “The Stability of a Bipolar World,” Daedalus, No. 93 [Summer 1964]). But it remains to be seen whether the current bipolarity will do as well at averting system-wide wars as the 19th-century multipolarity did after 1815.

²⁰ These notions closely parallel Durkheim's distinction between organic solidarity, linking highly differentiated units in a complex society, and mechanical solidarity, linking like units in a segmental society. Organic solidarity represents a qualitatively higher form and quantitatively greater extent of interdependence. Durkheim (fn. 2).

²¹ “The Myth of National Interdependence,” in Kindleberger, Charles P., ed., The International Corporation (Cambridge: MIT Press, 1970), 206.Google Scholar Waltz means international system-level interdependence (see below) relative to domestic system-level interdependence.

²² Waltz is quite explicit in denouncing what he calls unit-level measures of interdependence, but much less clear in defining what he means by systems level interdependence. I infer from the Durkheimian inspiration and from the kind of evidence that Waltz presents that he defines it in terms of two factors: (1) the relative size of the external sector, and (2) the degree of national specialization reflected in international transactions. Both of these factors are expected to co-vary with the number of great powers. Matters are muddled further, however, because Waltz insists that in measuring systemic interdependence we take into account only “the relatively high or low level of dependence of the great powers” (p. 145). But to do so is to employ the same indicator for both independent and dependent variables! The number of great powers is a structural attribute used to predict systemic outcomes; surely, in order to describe those outcomes, we need some aggregate measure that will include, but not be limited to, the economic activities accounted for by the great powers.

²³ Waltz's original argument was with Cooper, Richard N., The Economics of Interdependence (New York: McGraw Hill, 1968).Google Scholar Cooper shows, among other things, that the price sensitivity of factors is much higher today than in the pre-World War I period. That may be economically a more interesting form of interdependence, Waltz maintains, but it is politically less important. The quick re-allocation of factors of production in response to relatively small margins of advantage demonstrates that those ties do not need to be maintained, that they do not reflect mutual dependence stemming from functional differentiation (pp. 141–42). The debate concerning these two positions is ably conceptualized and summarized by Keohane, Robert O. and Nye, Joseph S., Power and Interdependence (Boston: Little, Brown, 1977), chap. 1.Google Scholar

²⁴ Economists would point out that intrasectoral trade, which accounts for an ever-increasing share of total world trade, also reflects an international specialization of labor. Waltz's response would be that this increases interdependence at the level of the firm, while it decreases it for the state compared to what it would be given an equivalent level of intersectoral trade.

²⁵ Waltz might be inclined to discuss the origins of World War I in this fashion, for instance, but then we would also need to have an explanation for the preceding “Hundred Years' Peace.”

²⁶ Cf. Dehio, Ludwig, The Precarious Balance (New York: Random House, 1962)Google Scholar, to whom Waltz, curiously, makes no reference.

²⁷ Waltz, Kenneth N., Man, The State and War (New York: Columbia University Press, 1959).Google Scholar

²⁸ Anarchy, recall, is defined as the absence of central rule. On the concept of “feudal anarchy,” see Poggi, Gianfranco, The Development of the Modern State (Stanford, Calif.: Stanford University Press, 1978)Google Scholar: “It arose from the fact that the system of rule relied, both for order-keeping and for the enforcement of rights and the redress of wrongs, on self-activated coercion exercised by a small, privileged class of warriors and rentiers in their own interest” (p. 31). Moreover, any standard text will document that neither the papacy nor the empire constituted agents of centralized political authority; see, for example, Strayer, Joseph R. and Munro, Dana C., The Middle Ages, 4th ed. (New York: Appleton-Century-Crofts, 1959).Google ScholarStrayer, , On the Medieval Origins of the Modern State (Princeton: Princeton University Press, 1970)Google Scholar, demonstrates nicely the balancing consequences triggered by threats of supranationality from the papacy, most profoundly in this instance: “the Gregorian concept of the Church almost demanded the invention of the concept of the state” (p. 22).

²⁹ Such an attribution would be historically inaccurate because there is a good deal of continuity in the “core units,” if these are identified retrospectively as the units that would become the major nation-states. But the exercise is nonsensical because, as Hedley Bull has pointed out, contemporaries found it impossible to enunciate a “fundamental constitutive principle or criterion of membership” in the international system. The major units were known as civitates, principes, regni, gentes and respublicae, the common element among them, the idea of statehood, not yet having taken hold (Bull, , The Anarchical Society [New York: Columbia University Press, 1977], 29).CrossRefGoogle Scholar To these must be added cities, associations of trades, commercial leagues, and even universities, not to mention the papacy and empire—all of which, for some purposes, were considered to be legitimate political actors, though of course they varied in scope and importance. For example, the right of embassy could be granted or denied to any of them, depending upon the social status of the parties involved and the business at hand; see Mattingly, Garrett, Renaissance Diplomacy (Baltimore: Penguin Books, 1964).Google Scholar

³⁰ Meinecke, Friedrich, Machiavellism, trans, by Scott, Douglas (New Haven: Yale University Press, 1957)Google Scholar; Meinecke spoke of the “heteronomous shackles” of the Middle Ages, referring to the lattice-like network of authority relations.

³¹ Poggi refers to a protracted dispute over whether this designation is appropriate (fn. 28, p. 26, n. 11). The end of the feudal period does not end the cause of the dispute: see Chabod, Federico, “Was there a Renaissance State?” in Lubasz, H., ed., The Development of the Modern State (New York: Macmillan, 1964).Google Scholar

³² Strayer and Munro (fn. 28), 115; Strayer (fn. 28), throughout.

³³ Anderson, Perry, Lineages of the Absolutist State (London: New Left Books, 1974), 37–34Google Scholar

³⁴ For instance, the lines between France, England, and Spain did not harden until the early 13th century. “It was at this period that not only were the boundary lines decided but, even more important, it was decided that there would be boundary lines. This is what Edouard Perroy calls the [fundamental change] in the political structure of Europe” (Wallerstein, fn. 8, p. 32). But the story does not end there. As late as 1547, when Francis I reformed the apparatus of the French state, he fixed the number of secrétaires d'É;tat at four; but the conception of “internal” and “external” was still so blurred that, rather than separating their duties according to it, each of the four supervised the affairs of one quadrant of France and the relations with contiguous and outlying states (Mattingly, fn. 29, p. 195).

³⁵ Strayer (fn. 32, p. 83) relates the hypothetical example of a king of France, who “might send letters on the same day to the count of Flanders, who was definitely his vassal, but a very independent and unruly one, to the count of Luxembourg, who was a prince of the Empire but who held a money-fief (a regular, annual pension) of the king of France, and to the king of Sicily, who was certainly a ruler of a sovereign state but also a prince of the French royal house.”

³⁶ “Angevin lineages could rule indifferently in Hungary, England or Naples; Norman in Antioch, Sicily or England; Burgundian in Portugal or Zeeland; Luxemburger in the Rhineland or Bohemia; Flemish in Artois or Byzantium; Hapsburg in Austria, the Neth erlands or Spain” (Anderson, fn. 33, p. 32).

³⁷ Mattingly (fn. 29), 41 and throughout.

³⁸ Anderson (fn. 33), 428.

³⁹ It is quite common, particularly in liberal writings on interdependence, to read of “the relative irrelevance of sovereignty” in the contemporary world wherein all states “are subject to diverse internal and external conditioning factors that induce and constrain their be havior,” and in which some states apparently are “more 'sovereign' than others.” The cited snippets are from Richard W. Mansbach and others, The Web of World Politics (Englewood Cliffs, N.J.: Prentice-Hall, 1976), 20–22. Waltz's definition of sovereignty is not helpful either: “To say that states are sovereign is not to say that they can do as they please…. To say that a state is sovereign means that it decides for itself how it will cope with its internal and external problems …” (p. 96). If sovereignty meant no more than this, then I would agree with Ernst Haas, who once declared categorically: “I do not use the concept at all and see no need to.” “Letter to the Editor,” Journal of Common Market Studies, VIII (September 1969), 70.

⁴⁰ More precisely, the internal side of sovereignty had to do with sovereignty as a legiti mation for central state authority vis-à-vis competing domestic claimants. That was Bodin's concern. My discussion below addresses only the external side, which dealt with sovereignty as a legitimation for the interstate order.

⁴¹ “[Private,] to put it another way, refers not so much to the nature of the entity that owns, but to the fact that it is an entity, a unit whose ownership of nature … signifies the exclusion of others from this ownership.” R.N. Berki, “On Marxian Thought and the Problem of International Relations,” World Politics, XXIV (October 1971), 99; emphasis added.

⁴² Neo-Thomists like Vitoria and Suarez sought to adapt both inclusive property rights and natural law to the new circumstances, without abandoning either. Filmer and Hobbes abandoned both, arguing—on Adamite and utilitarian grounds, respectively—for the necessity of absolutist arrangements internally and, in the case of Hobbes, for the inevitability of the state of war externally. Grotius and Pufendorf developed mixed solutions that pointed the way toward the future. Both accepted the idea of exclusive property rights. Grotius allowed for some natural rights in things while Pufendorf argued that these rights must be conventional. But, critically, both defined the only remaining natural rights basis for sociableness or community negatively, in terms of the duty to abstain from that which belongs to another. Liberal theories of social order followed directly from this premise. A good summary, on which this characterization has drawn, may be found in Tully, James, A Discourse on Property: John Locke and His Adversaries (New York: Cambridge University Press, 1980)CrossRefGoogle Scholar, chaps. 3–5. For a brief and useful overview of the international side, see Gross, Leo, “The Peace of Westphalia, 1648–1948,” in Falk, Richard A. and Hanrieder, Wolfram H., eds., International Law and Organization (Philadelphia: J.B. Lippincott, 1968).Google Scholar

⁴³ Cited in Tully (fn. 42, p. 95); the quotation is from the chapter in the Second Treatise of Goverment entitled “On Property,” section 27; emphasis in original. Tully tries to debunk the notion that Locke was an apologist for absolute private property and emergent capitalist relations of production, as argued most forcefully by Macpherson, C. B., The Political Theory of Possessive Individualism (New York: Oxford University Press, 1962)Google Scholar; but he seems to me to go too far in the opposite direction.

⁴⁴ Cited in Tully (fn. 42), 129, from Second Treatise, section 27.

⁴⁵ Cited in Tully (fn. 42), 150–51, from ibid., section 136.

⁴⁶ Cited in Tully (fn. 42), 163, from ibid., section 131; italics omitted. Note, however, that Locke defined property very broadly here, to include that in which individuals have rights, including life, liberty, and possessions.

⁴⁷ Macpherson (fn. 43), 80. Macpherson develops this point in his discussion of Hobbes, but subsequently applies it to Locke as well.

⁴⁸ Gross (fn. 42), 65. Vattel, Emeric, The Law of Nations, trans, by Fenwick, Charles G., in Scott, James Brown, ed., The Classics of International Law, IV (Oxford: Clarendon Press, 1916).Google Scholar

⁴⁹ Hinsley, F. H., “The Concept of Sovereignty and the Relations between States,” Journal of International Affairs, XXI (No. 2, 1967), 242–52, at 245: “It was a condition of the discovery of the international version of sovereignty that the notion of Christendom be replaced by a different understanding of international society—one that was compatible, as the medieval understanding was not, with belief in the sovereignty of the state…. [T]here could be no successful international application of the theory until the notion of the sovereign power of the individual state had been reconciled with the ethical principles and the political needs of an international community consisting of independent states.”Google Scholar

⁵⁰ Macpherson (fn. 43), chap. 6. Autonomy, then, which is so often confused with the very term sovereignty, characterizes the ontological basis of the legitimation expressed by sovereignty.

⁵¹ Mattingly (fn. 29), 244; see also Bozeman, Adda B., Politics and Culture in International History (Princeton: Princeton University Press, 1960)Google Scholar, chap. 13, where the origin, generalization, and acceptance of this “necessity” is traced.

⁵² Barnard, , “On Planning for World Goverment,” in Organization and Management: Selected Papers (Cambridge: Harvard University Press, 1948), 151.CrossRefGoogle Scholar

⁵³ Mattingly (fn. 29), throughout.

⁵⁴ Polanyi, Karl, “Aristotle Discovers the Economy,” in Polanyi, and others, eds., Trade and Markets in the Early Empires (Glencoe, lll.: Free Press, 1957).Google Scholar

⁵⁵ At least, that is what I have attempted to show in my paper “International Regimes, Transactions and Change: Embedded Liberalism in the Post-War Economic Order,” International Organization, XXXVI (Spring 1982).

⁵⁶ The term is Zolberg's (fn. 11).

⁵⁷ Durkheim (fn. 1), 115; (fn. 2), Book II, chap. 2.

⁵⁸ E.g., Furubotn, Eirik G. and Pejovich, Svetozar, eds., The Economics of Property Rights (Cambridge, Mass.: Ballinger, 1974).Google Scholar

⁵⁹ North, Douglass C. and Thomas, Robert Paul, The Rise of the Western World: A New Economic History (New York: Cambridge University Press, 1973).CrossRefGoogle Scholar My summary perforce is a highly stylized rendering of what is already fairly stylized historical work.

⁶⁰ For a preliminary and still largely descriptive effort in this direction, see Ruggie, , “On the Problem of [The Global Problematique,]” Alternatives, v (January 1980).Google Scholar My tentative conclusion in that paper is that greater global dynamic density has produced change in the international framework of states' “private property rights,” but that to date this change continues to reflect an underlying determining logic that has not itself changed. Thus far, therefore, it represents an adaptive redeployment of this structural level, not a fundamental rupture in it.

⁶¹ This mode of reasoning permeates the last three chapters of Waltz's book; but see especially pp. 146–60.

⁶² See, most recently, Rosecrance (fn. 9).

⁶³ Constructing and then adhering to generative structural models are extremely difficult intellectual exercises. Perhaps it is some consolation to know that, according to Lévi-Strauss, Durkheim failed too, as a result of which “he oscillates between a dull empiricism and a prioristic frenzy.” Lévi-Strauss, Claude, “French Sociology,” in Moore, Wilbert and Gurvitch, Georges, eds., Twentieth Century Sociology (New York: Philosophical Library, 1946), 528.Google Scholar Waltz's empiricism is never dull.

⁶⁴ Waltz imputes this unidirectional causality to the structural mode of explanation: “Structural thought conceives of actions simultaneously taking place within a matrix. Change the matrix-the structure of the system-and expected actions and outcomes are altered.” Waltz, “What Causes What? Systemic and Unit-Level Explanations of Change,” Institute of International Studies, University of California, Berkeley, draft, January 1982, 35. In point of fact, structural explanations in the social sciences are far more complex, and sometimes even dialectical, as the surveys in Glucksman (fn. 16) and Kurzweil (fn. 16) testify.

⁶⁵ Giddens, Anthony, Émile Durkheim (New York: Penguin Books, 1978), 125.Google Scholar To avoid any possible misunderstanding, let me add that Waltz does not argue that unit-level phenomena are important for nothing, but that they have no place in systemic theory. In international relations, according to Waltz, they belong to the realm of foreign policy. (See his exchange with Rosecrance, Richard, in International Organization, xxxvi [Summer 1982], 679–85.)Google Scholar

⁶⁶ For a structural model of international systemic continuity/transformation which stresses the concatenation of “synchronie articulations” and “diachronic processes,” and which I find more satisfactory than either Waltz's model or the prevailing alternatives, see Anderson (fn. 33), 419–31.

⁶⁷ For Durkheim, the notion of dynamic density at one and the same time reflected structural effects and aggregated unit-level processes into a systemic variable that in turn affected structure.