¹ System and Process in International Politics(New York, 1957)Google Scholar, ch. 2. For a more general presentation, see also his “Balance of Power, Bipolarity, and Other Models of International Systems,” this Review, Vol. 51 (September, 1957), pp. 684–95Google Scholar.

² System and Process, pp. 22, 50.

³ Ibid., pp. 45–49.

⁴ Ibid., pp. 36–44.

⁵ Ibid., p. 78. A note on the use of the word “bloc” is in order. This term will be used to describe a group of nation-states to the extent that they act as a unit in international relations. Although at the outset it will be assumed that each bloc acts as a unit, the word “bloc” itself implies neither particular institutional structures nor any measure of unity of action. Indeed, one might consider alliances asthe generic term for relationships between states; in this case a “bloc” would be regarded as a species of alliance—i.e., an alliance having a “functional organization” as an ongoing institution. For thepurposes of theoretical exposition, however, the reverse usage will be adopted here. A “bloc” will be considered as a genus or general category, describing all types of relationships between a group of nation-states. This usage is advantageous because, although the term “alliance” usually connotes a relationship in which states are the constituent units, it is possible to conceive of alliances between blocs as well as between states. Hence, considering the entire international system as the frame of reference, an alliance is but one way in which several “actors” (nation-states or blocs) may coordinate their political, military, and economic actions. From a general theoretical perspective, the varying degrees of coordination between nation-states can best be expressed in terms of a continuum; for a typology expressing the range of possibilities in this continuum, see below, Table I.

⁶ U. S. War Aims (Boston, 1944), p. 187 Google Scholar. Cf. Carr, Edward Hallett, Nationalism and After (New York, 1945), pp. 53–54, 60–62 Google Scholar.

⁷ Organisations Internationales et Pouvoirs Politiques des États, Cahiers de la Fondation Nationale des Sciences Politiques, No. 52 (Paris: Librairie Armand Colin, 1954), pp. 416–17Google Scholar. (My translation.)

⁸ “Regionalism, Functionalism, and Universal International Organization,” World Politics, VIII (January, 1956), 263 Google Scholar.

⁹ Herz, John H., International Relations in the Atomic Age (New York: Columbia University Press, 1959), pp. 34–35 Google Scholar.

¹⁰ Ibid., p. 182.

¹¹ Kaplan himself has implied that a multiplicity of blocs might arise if the use of nuclear weapons were prohibited by joint agreement. (System and Process, p. 52). This suggests that, for Kaplan, the “unit veto” model describes a state of an international system rather than a determinant structure. But certain structures are more likely to produce self-limiting agreements than others, as will be argued below. Kaplan's “unit veto” model, by stressing the implications of nuclear capabilities, ignores the power context within which such military weapons are placed.

¹² International Politics, chs. 7, 8. See esp. p. 111.

¹³ Wolfers, Arnold, “Stresses and Strainsin ‘Going It with Others’,” in Wolfers, Arnold, ed., Alliance Policy in the Cold War (Baltimore: JohnsHopkins Press, 1959), p. 7 Google Scholar.

¹⁴ The definition of “integration,” the extent to which it may vary, and the effects of different degrees of integration must all be of crucial importance for the operation of the system. Although it is convenient, in order to establish the general configuration of the multi-bloc model, to assume that all blocs are equally integrated, this assumption will be examined, in Section V.

¹⁵ Cf. Mattingly, Garrett, Renaissance Diplomacy (London: Jonathan Cape, 1955), Pt. IIGoogle Scholar.

¹⁶ The bibliography on the “balance of power” system is, of course, quite large. For a survey of the operations of the system after 1848, see Taylor, A. J. P., The Struggle for the Mastery of Europe (Oxford, 1954)Google Scholar.

¹⁷ Kaplan, , System and Process, p. 23 Google Scholar.

¹⁸ Lippmann's reply to this argument with reference toWorld War II is not irrelevant: “It will be feared that great constellations like the Atlantic Community, the Russian Orbit, the Chinese Community will become engaged in a titanic conflict. The members of these communities are now engaged in a titanic conflict, and in order to survive they have had to organize impromptu and tardily, the strategical combinations which, in my view, they ought to maintain and perfect. It cannot be said that this titanic conflict was causedby regional combinations. It can be said that it was not prevented and has very nearly been lost because they did not exist.” U. S. War Aims, p. 191.

¹⁹ Deutsch, Karl W. et al., Political Community and the North Atlantic Area (Princeton: Princeton University Press, 1957), p. 25 Google Scholar.

²⁰ “The smaller the society, the fewer probably will be the distinct parties and interests composing it; the fewer the distinct parties and interests, the more frequently will a majority be found of the same party; and the smaller the number of individuals composing a majority, and the smaller the compass within which they are placed, the more easily will they concert and execute their plans of oppression.” The Federalist, Earle, Edward Mead, ed. (New York: Modern Library, n.d.), pp. 60–61 Google Scholar. While Madison here refers to internal oppression, the same argument could be made with reference to foreign policy.

²¹ Kaplan, , System and Process, pp. 104–105 Google Scholar.

²² Cf. Organisations Internationales, pp. 417–18. Hoffmann even implies that the unique virtue of regional associations is that they couldn't engage in “power politics,” at least if properly regulated.

²³ Kaplan, , System and Process, p. 23 Google Scholar.

²⁴ Ibid., pp. 6–7. See below, Section III.

²⁵ The possibility of the opening of space to conquest, while not absurd, is ignored for these purposes (if only because its consequences seem hard to predict, although the parallel with the exploration and control of the non-European world after the Sixteenth Century obviously suggests itself).

²⁶ This restriction will, as has been suggested, be removedin Section V.

²⁷ Kaplan, , System and Process, p. 23 Google Scholar.

²⁸ It may not be irrelevant in this respect that the Soviet Union rejected suggestions, after World War II, that its Eastern European satellites join the U.S.S.R. as integral member-republics. If the maintenance of the form of nation-states, even with reference to minor ones, was compatible with Soviet ideology and political domination, self-restraint with major rivals might also be expected.

²⁹ Ibid., p. 43.

³⁰ Cf. Good, Robert C., “The U. S. and the Colonial Debate,” in Wolfers, , Alliance Policy, esp. pp. 269–70Google Scholar.

³¹ In Kaplan's models, these nonaligned states might be: (a) “neutral” balances within an international actor (“loose bipolar system”), (b) powerlessly integrated into two opposed blocs (“tight bipolar”), (c) relatively equal members of a world government (“universal system”) or largely dissolved into a world government (“hierarchical system”), or finally, (d) all either provided with an invulnerable nuclear capability or destroyed (“unit veto system”). The first of these roughly corresponds with the present situation, and does not indicate how the rising power of these areas will affect the “balancing” procedure; the second and third are relatively unlikely. The last (“unit veto”) would be not only highly unstable, but, in any prudent sense, “undesirable”.

³² Kaplan, , System and Process, p. 23 Google Scholar.

³³ The religious wars of the Sixteenth and Seventeenth Centuries are a frequently cited illustration of this; the “liberal”—“monarchist” split after the Congress of Vienna is a more recent example.

³⁴ An even more extreme proof may be found in the tacit opposition of the Papacy to Hapsburg expansion (most especially when directed toward Italy) and the somewhat less than messianic zeal of the Pope's oppositionto the Protestant princes on some occasions.

³⁵ Herz, , International Relations, pp. 153–56Google Scholar. Cf. Walter Lippmann's definition of a “solvent” foreign policy, U. S. Foreign Policy (Boston, 1943), pp. 9–10 Google Scholar.

³⁶ Kaplan, , System and Process, p. 23 Google Scholar.

³⁷ U. S. War Aims, pp. 189–90.

³⁸ It would seem that this is the weakest point in Herz' analysis, for he appears to leap from the era of allegiance to nation-states (now “obsolete” because of nuclear weapons) to “universalism” (International Relations, ch. 12). In point of fact, men's loyalties have been shifted from narrow to broader communities, but rarely has the scope of the community involved such a revolutionary change. To commit himself to world government, the average American citizen would be forced to “identify” himself with Laos, South Africa, and Yemen as well as South Carolina.Herz is forced to argue that the threat of nuclear annihilation must induce thischange, but popular apathy would seem to belie the actuality of this process. Of course, a nuclear war or an attack from Mars might do the trick …. In the meantime, worldwide organization, if it assumes direct powers over all regions and states, is likely to do so under the “guidance” of a hegemonic power center. Cf.: “A world organization may be a necessary convenience as well as a valuable symbol. But the intermediate unit is more likely to be the operative factor in the transition from nationalism to internationalism.” Carr, , Nationalism and After, pp. 46–47 Google Scholar.

³⁹ This is not the place to enter into an extensive discussion of the current role of the UN or its possibilities. See Claude, Inis L. Jr., Swords Into Plowshares (New York, 1959)Google Scholar for a balanced judgment. Cf. Haas, Ernst B., “Regional Integration and National Policy,” International Conciliation, No. 513 (May, 1957), esp. pp. 438–42Google Scholar.

⁴⁰ See his International Equilibrium (Cambridge: Harvard University Press, 1957)Google Scholar, esp. chs. 5, 6, 7. Because of the general unwillingness of statesmen to surrender power to an external organization beyond their control, it has seemed wiser to take Kaplan's “balance of power” model rather than Liska's “multiple equilibrium” as a theoretical starting point. Nonetheless the effects of the possession of nuclear weapons by a small number of actors may well create a congruence of the two seemingly diverse schemes. See below, Section IV.

⁴¹ Cf. Haas, “Regionalism, Functionalism, and Universal International Organization.”

⁴² This part of the rule is implied in Rule 3. However, since the changes in the identity of actors are not instantaneous, this rule also implies that the number of members of the system may change. This part of the rule will not be analyzed in any detail, save for passing references to the influence of numbers in Section III.

⁴³ Kaplan, , System and Process, p. 23 Google Scholar.

⁴⁴ Cf. Haas, , “Regional Integrationand National Policy,” pp. 381–83, 433–37Google Scholar; Miller, William Lee, “The American Methods and the Alliance System,” in Alliance Policy, pp. 33–34 Google Scholar; and William Welch, “Soviet Commitments to Collective Action,” Ibid., pp. 280–81. This remark ignores, obviously, the varying integration in different blocs; moreover, current regional agreements are integrated ata low level, whereas the “pure” model assumes uniformly and highly integrated units.

⁴⁵ “Rigidity” here does not mean that two blocs with a common member state could not conceivably be aligned against each other, but only that the presence of the common member makes them less likely to chooseallies according to the needs of the moment. E.g., the dynastic union of the states of Austria and Spain under the Hapsburg family (and later of France and Spain under the Bourbons) limited the ability of the “branches of the family” to oppose one another. It might be added that if a single nation-state is a member of two blocs, these two blocs might be encouraged to oppose one another should each bloc seek sole control over the common member; in this case the “rigidity” introduced by overlapping membership is an increase in the probability of opposition, rather than an increase in the probability of alliance.

⁴⁶ E.g., it is often argued that the rigidity of the Austro-German alliance brought about World War I as a “total” ratherthan a “limited” war, because it forced conflict with any European state to become conflict with all. Cf. Kaplan, , System and Process, p. 29 Google Scholar.

⁴⁷ Hence an Indian author has opposed a regional organization in the Indian Ocean having “punitive” collective security powers,largely on the ground that English participation in the bloc would enable Great Britain to maintain a kind of quasi-colonial control. Pannikar, K. M., “Regionalism and World Security,” in Pannikar, et al., Regionalism and Security (New Delhi: Indian Council of World Affairs, 1948), pp. 3–6 Google Scholar.

⁴⁸ This type of rigidity might be comparable to the Englishpreference for the Austrian alliance prior to the “Diplomatic Revolution” of 1756.

⁴⁹ Kaplan, , System and Process, p. 6 Google Scholar.

⁵⁰ In the terminology of system-theory, a highly flexible system is “ultra-stable” since, in responding to stimuli, it can “search” for the level or general type of response. In contrast, a system which can only adapt to its environment on a given level or “field” of behavior is technically called “stable.” See Ashby, W. Ross, Design for a Brain (New York, 1952), p. 99 Google Scholar; Kaplan, , System and Process, pp. 6–8 Google Scholar. An ultra-stable system can be illustrated by a system offive powers in which a large number of alliances is possible; within any set of alliances, the system would be capable of adapting to changes and maintaining itself. In contrast, a bipolar system may be “stable” (since it can maintain itself in the face of internal and external changes), but no different arrangement of power is possible without altering the very structure of the system.

⁵¹ Ibid., p. 52.

⁵² Kaplan recognized that the “balance of power” model could be stable only if the rate of change of alliances is within limits (Ibid., p. 35), but his rules themselves contain no provision for this “realistic” limitation.

⁵³ Cf. Kaplan's image of the precarious nature of the “balancing” process, Ibid., p. 32.

⁵⁴ See especially Fellner, William, Competition Among the Few (New York, 1949), pp. 15–16, 24–50, 120–136 Google Scholar and passim.

⁵⁵ Elasticity of demand can be ignored, at least for those “price wars” in which the selling price is reduced below total average cost. Since such severe price wars are not incomparable with large scale modern warfare in their effects, the parallel seems justified.

⁵⁶ “Power Politics and the Growing Nuclear Club,” Policy Memorandum Number 20 (Princeton University, Center of International Studies, 1959), p. 18 Google Scholar and passim. The following analysis will assume the unlikelihood of technological changes which make an invulnerable second-strike capability impossible; suchan innovation would probably lead to hegemony for its possessor, total war, or both.

⁵⁷ Burns distinguishes between “passive triggering “ (in which the expenditure of missiles by an attacking independent power would remove the invulnerability of the attacker's second-strike capability vis-à-vis a third, independent nuclear power) and “active triggering “ (in which the triggering power, by using its weapons in a firststrike on an independent nuclear power, renders the victim vulnerable to a succeeding first strike by another nuclear power). Ibid., pp. 12–13. Itshould be noted that while the two types of triggering are different, especiallyunder the technological assumptions which Burns makes, they could well reinforceone another.

⁵⁸ This would follow when the more powerful nuclear actor wished to avoid both the initiation of nuclear attack (necessary to neutralize the small, vulnerable stockpile of its neighbor) and the loss of one or two cities(which would result from any non-nuclear attack). See the example of Sweden deterring Russia, Ibid., p. 17, and Rathjens, George W. Jr., “NATO Strategy: Total War,” in Knorr, Klaus, ed., NATO and American Security (Princeton: Princeton University Press, 1959), ch. 4, pp. 88–89 Google Scholar.

⁵⁹ Note that although the “unit veto” model assumes that each of a large number of countries has an independent deterrent, nuclear symmetry is far less likely to be realized on the level of nation-states, where economic and technical prerequisites are often limited by sheer size, than on the level of regional blocs.

⁶⁰ Op. cit., p. 16.

⁶¹ Ibid., (original italics).

⁶² Cf. Fellner, op. cit., pp. 120 ff.

⁶³ It is, however, uncertain that nuclear war would be as rapid as is generally assumed in the West. The Russians apparently plan for a more protracted battle. See Garthoff, Raymond L., Soviet Strategy in the Nuclear Age (New York, 1958), pp. 87–89 Google Scholar.

⁶⁴ The experience of those who tried to compete with John D. Rockefeller's Standard Oil Company attests to this similarity. Cf. Tarbell, Ida, The History of the Standard Oil Company (NewYork, 1904), II Google Scholar, passim.

⁶⁵ Compare the behavior of the United States, England, and the USSR on attaining a nuclear capability with their strategic attitudes (as well as that of France) at the present time.

⁶⁶ It is, however, true that the emergence of a “tacit agreement” requires the kind of “learning” and “testing” which is associated with the “stochastic” model in game theory. See Kaplan, , System and Process, pp. 225 ff, 247 Google Scholar.

⁶⁷ International Politics, pp. 285–86.

⁶⁸ Current discussions of disarmament, with their emphasis on controls, usually fail to recognize that the possession of nuclear weapons, as distinct from their manufacture, is now probably beyond any feasible system of international inspection. Agreements to disarm should be restricted to those armaments whose presence can be detected with a relatively high probability— i.e., especially conventional forces. Extending agreements, especially in writing, beyond this point merely invites the least scrupulous power to violate the agreement, since a cheater has some grounds for believing his rivals will not cheat. It is one of the great advantages of quasi-agreements in general that the certainty of being the only cheater is denied the unscrupulous; hence when a quasi-agreement covers an action which can be detected by suspiciousrivals (e.g., weapons testing or all-out nuclear assault), violations arerelatively unlikely, though by no means impossible.

⁶⁹ See Kissinger, Henry A., Nuclear Weapons and Foreign Policy (New York, 1957)Google Scholar, ch. 6.

⁷⁰ See Kaplan, Morton A., “The Strategy of Limited Retaliation,” Policy Memorandum Number 19 (Princeton University, Center of International Studies, 1959)Google Scholar.

⁷¹ Cf. Kaplan's, discussion of limited retaliation in the context of NATO policy: “Problems of Coalition and Deterrence,” in Knorr, , NATO and American Security,ch. 6, pp. 142–48Google Scholar.

⁷² Cf. Kissinger, , Nuclear Weapons, pp. 178–82Google Scholar.

⁷³ In contrast, “leadership” should not be expected to prevent other blocs or powers acquiring nuclear weapons, because on thisissue each actor may feel that without nuclear weapons he is not really a participant in the system. See Burns, , “Power Politics,” pp. 6–9 Google Scholar.

⁷⁴ This type of “controlled diffusion” is discussed by Burns, Arthur Lee, “NATO and Nuclear Sharing,” in Knorr, , NATO and American Security Google Scholar, ch. 7.

⁷⁵ E.g., Kaplan has suggested a distinction between “hierarchical,” “mixed-hierarchical,” and “non-hierarchical” blocs (System and Process, pp. 74–83); Deutsch et al. have proposed a simpler dichotomy between “amalgamated” and “pluralistic” “security communities.” (Political Community, pp. 6–7). In neither case, however, do the authors propose any generalrelationships between the types of integration and the kind of behavior which islikely.

⁷⁶ Cf. Easton, David, “Political Anthropology,” in Siegal, Bernard J., ed., Biennial Review of Anthropology, 1959 (Stanford: Stanford University Press, 1959), pp. 239–41Google Scholar. It should be noted thatthe theoretical problems concerning the political structures of so-called “primitive societies” are closely parallel to those concerning the international system. Incidentally, this parallel shows the weakness of Kaplan's definition of the “political system” (System and Process, pp. 12–14)—or rather, of his interpretation that the international system hasa null political sub-system. Here Kaplan equates the political system with political structures, thereby risking an error made in some early studies of what have been called “stateless societies.” Cf. the debate on this point in the literature of anthropology (largely cited in Easton's article, just noted).

⁷⁷ Easton's continuum, applied to primitive societies, is based on three interrelated variables: the differentiation of political rolesfrom other roles, the differentiation among political roles (i.e., their number), and the specialization of roles by kind or by level. These variables are implicit in Table I.

⁷⁸ Kaplan, , System and Process, p. 16 Google Scholar.

⁷⁹ Ibid., p. 18.

⁸⁰ Cf. Ibid., pp. 56–58.

⁸¹ E.g., note the foreign policy of the French revolutionaries of 1848, which did not differ from the policy of the July Monarchy inits effect, despite verbal differences. Taylor, , Struggle for the Mastery of Europe, pp. 8–11, 15–16 Google Scholar. This does not mean that “directive” blocs may not be more “aggressive” than “non-directive” ones, but it suggests that the difference lies in the goals motivating decision-makers (and not in structure perse). Cf. the Second Empire and the Second Republic on the one hand, and present daySpain with Imperialist England on the other.

⁸² E.g., the United States would of course be capable of devoting more resources to its NATO commitments if it did not have commitments to SEATO, etc.

⁸³ The other major war of the century (the American civil war) might well have produced an international cataclysm but for the logistic andtechnological isolation of the American continent at the time and the relative strength of the contending parties.

⁸⁴ In fact, one can visualize the international political system as already having a structure similar to what has been described, in the context of African tribes, as a “segmentary state.” Southall, Aiden W., Alur Society (Cambridge: W. Heffer and Sons, n.d. 1956?)Google Scholar, esp. ch. 9. Cf. Middleton, John and Tait, David, eds., Tribes Without Rulers (London: Routledge and Kegan Paul, 1958)Google Scholar, esp. Introduction; and Smith, M. G., “On Segmentary Lineage Systems,” Journal of the Royal Anthropological Institute, Vol. 86 (July–December, 1956), pp. 39–80 Google Scholar. In this respect, the emergence of bloc organs, functionally responsible for nuclear weapons, represents merely a further trend in contemporary developments—and a development which has already been suggested with respect to NATO. See Hilsman, Roger, “On NATO Strategy,” in Alliance Policy Google Scholar, ch. 6; Kaplan, Morton A., “Problems of Coalition and Deterrence,” in Knorr, , NATO and American Security, ch. 6, esp. pp. 139–42Google Scholar.

⁸⁵ Cf. the argument of Nitze, “Coalition Policy and the Concept of World Order.”

⁸⁶ Cf. Ibid., p. 29.

⁸⁷ Cf. the general pattern of mid-Eighteenth Centurydiplomacy, in which the Anglo-Austrian alliance opposed France and Prussia priorto the “Diplomatic Revolution.”

⁸⁸ Such a “bi-axial” system is rendered all themore likely (at least in the short run) by current weapons technology, in which the United States and the USSR seem likely to maintain a considerable superiority both in quantitative and qualitative terms. Nonetheless, the bi-axial alignment need not be taken as the only possible form of a multi-bloc system; the present model has been conceived in highly abstract, long run terms (i.e., as apossibility for the next 100 years).