¹ For Russett's views on the nature of “science” see, especially, the Preface and p. 217.

² For a general discussion of these problems consult Young, Oran R., Systems of Political Science (Englewood Cliffs, N.J. 1968)Google Scholar, chap. 7.

³ The problem of locating sound operational indicators of political phenomena is an especially difficult one in most branches of political science. Specifically, it seems to raise considerably more serious difficulties than the analogous problem in a field such as economics.

⁴ This judgment raises particularly serious problems. First, many relationships can be “demonstrated” if one is careful enough to select just the right points of comparison. Second, the notion of normality is a rather peculiar one if “normal” activities seldom occur in empirical terms. Third, taking Russett's statement at face value, we must assume that the economic autarky of the era of depression in the thirties was a “normal” development.

⁵ Especially 1–7.

⁶ For example, a typical dictionary definition of region is “any more or less extensive, continuous part of a surface or space.” The American College Dictionary, 1962, 1020.

⁷ The discussion of various definitional criteria used by other scholars in chap, i (see pp. 2–7) indicates concern with the dangers of leaving the definition of region fully open-ended. It does not, however, lead to the formulation of an alternative to existing lexical notions involving the necessity of physical contiguity.

⁸ It is true that chap. 1 contains a list of the factors to be explored empirically in the substantive sections of the book (see p. 11). This list of factors, however, is not offered as a formal definition of the concept “region.” Nor is it accompanied by a discussion of the conceptual implications involved in the selection of this list of factors rather than some other list.

⁹ Russett ultimately decides, on a de facto basis, to define regions in terms of socio-cultural homogeneity, United Nations voting, membership in intergovernmental organizations, trade, and geographical proximity. There would appear, however, to be no a priori reason to select this particular collection of variables from the infinite universe of possibilities.

¹⁰ For a classic statement of the distinction between essentialism and nominalism see Popper, Karl R., The Poverty of Historicistn (New York 1964), 26–34Google Scholar.

¹¹ The term “suitable” is here in quotation marks since the threshold of suitability must ultimately be designated arbitrarily.

¹² For a discussion of many of these problems couched in different terms see Levy, Marion J. Jr., “‘Does it Matter if He's Naked?’ Bawled the Child,” in Knorr, Klaus and Rosenau, James N., eds., Contending Approaches to International Politics (Princeton 1969)Google Scholar.

¹³ The discovery of fruitful empirical premises is one of the creative aspects of scientific analysis. Although relatively little is known about this form of creativity, it is clear that premises are often derived from many disparate sources. A process of correlating many variables with each other may, of course, produce interesting regularities about which to hypothesize. Because of the work involved and the vagaries of simple correlations, however, this often turns out to be a rather inefficient process of searching for potentially interesting links.

¹⁴ He evidently made this assumption because the “conventional wisdom of international relations” suggests such relationships.

¹⁵ Because of the striking absence of consensus in the discipline of political science on definitions of politics and the political, it is of considerable importance to specify the definitions to be employed in any given study.

¹⁶ For a sophisticated discussion of the uses of statistical procedures in such situations that, however, fails to demonstrate a greater role for such procedures than suggested here see Alker, Hayward Jr., Mathematics and Politics (New York 1965)Google Scholar, chap. 6.

¹⁷ In some situations, the identification of empirical regularities allows for useful projections (not predictions in the formal sense) of future developments. The universe-of-cases problem referred to here, however, is apt to make projections of this kind quite unreliable in the analysis of political phenomena.

¹⁸ For a brief but clear exposition of the philosophic arguments leading to the conclusion that puristic induction does not constitute an efficient procedure for the discovery of empirical premises see May Brodbeck, “Explanation, Prediction, and ‘Imperfect’ Knowledge,” 363–98 in Brodbeck, May ed., Readings in the Philosophy of the Social Sciences (New York 1968)Google Scholar. For a more general discussion of these aspects of the phenomenon of scientific creativity in terms of the notion of “pattern perception” consult Tullock, Gordon, The Organization of Inquiry (Durham, N.C. 1966)Google Scholar, chap. 5.

¹⁹ For explicit statements along these lines see, for example, pp. 191, 201, 210, and 212.

²⁰ All intrinsically empirical disciplines (unlike logic or mathematics) must engage in a process of matching their abstract formulations wit h empirical realities. Under the circumstances, the collection of data is necessary to the development of theory though it is not especially interesting as an end in itself.

²¹ See p. 191 and especially the statement that “we have produced no measures of integration that might be matched against the attributes which are supposed to affect it.”

²² For examples see Ibid., pp. vii and 94–99 as well as Chaps. 12 and 14.

²³ See especially the statement “it is a relationship or pattern of behavior, the ability to work out… conflict with a minimum of violence and without one party always making important concessions, that marks the condition of successful political integration.”

²⁴ One or two possible exceptions such as the link between regional ties and war are discussed in a later section of diis essay. In general, however, the statement in the text seems accurate.

²⁵ See especially the argument that “Responsiveness is a difficult phenomenon to measure reliably at the level of international politics….” An d then, “Accordingly, I have not attempted to do so for the macro-analysis of this volume.”

²⁶ For examples of this fallacy see pp. 200–201 and 224–226.

²⁷ Given the present state of the field of political science, highly speculative work concerning political phenomena is often justifiable. In a work that purports to be conceptually rigorous and empirically systematic, however, a drift toward loose and unsupported speculations constitutes a serious deficiency.

²⁸ For an extended examination of these problems see Young, Oran R., “A Systemic Approach to International Politics,” Research Monograph No. 33, Princeton Center of International Studies, 1968Google Scholar.

²⁹ For example, this possibility is not mentioned in the discussion of purposes in Russett's Preface.

⁸⁰ For a discussion emphasizing possibilities of this kind see Nye, Joseph S. Jr., “Comparative Regional Integration: Concept and Measurement,” International Organization, xxii (Autumn 1968), 855–80CrossRefGoogle Scholar.

⁸¹ For a discussion that essentially acknowledges this weakness see pp. 207–10.

⁸² This statement refers explicitly to the idea that similarity of political orientations is a prerequisite for the growth of formal organizational ties among states. The conclusion, however, applies equally well to other relationships.

³³ The hypothesis is never stated explicitly but this is the clear implication of the discussion (pp. 191–93).

³⁴ On the question of “new starts” vs. duration or intensity see the conceptualization (pp. 196–98). The same problem characterizes Singer's recent study of the relationships between war and alliances. See Singer, J. David and Small, Melvin, “Alliance Aggregation and the Onset of War, 1815–1945,” in Singer, J. David, ed., Quantitative International Politics (New York 1968)Google Scholar. For a telling critique of Singer (much of which is relevant to the present discussion) see Dina A. Zinnes, “An Analytic Study of the Balance of Power Theories,” Journal of Peace Research (No. 3, 1967), 270–88.

³⁵ The idea that overt violence (in contrast to coercion) has been inhibited by the growing impact of perceptions concerning the potential destructiveness of nuclear war has been discussed frequently in recent years. And, of course, the impact of such perceptions has been especially great in the European arena in contrast to other parts of the world. For one interesting example see Knorr, Klaus, On the Uses of Military Power in the Nuclear Age (Princeton 1966)Google Scholar.

³⁶ It would be interesting, for example, to make a similar study of the connections between regional ties and violent conflict in eighteendvcentury Europe. While that era's Europeanism and cosmopolitanism are frequently commented upon, it was certainly not a period noted for the relative absence of wars.

³⁷ This statement refers explicitly to the links between trade and violent conflict. With respect to some of the other variables in question it is not clear that we can even say this much.

³⁸ Russett himself suggests the ambiguity introduced by the distinction between opportunities and precipitating factors (p. 200). Ambiguities of this kind, of course, reflect the limitations of correlational procedures in the absence of theory.

³⁹ Russett himself does the job in this case. Consider the various explanations suggested (pp. 97–58).

⁴⁰ For Russett's own statement of this argument see pp. 92–93.

⁴¹ For Rosenau's own formulation see Rosenau, James N., “Pre-Theories and Theories of Foreign Policy,” in Farrell, R. Barry, ed., Approaches to Comparative and International Politics (Evanston, Ill. 1966)Google Scholar.

⁴² For examples of statements of this kind see pp. 227–33.

⁴³ Once again, this is because we have been given no indicators for political integration and, therefore, the links between regional ties and integration have not been established.

⁴⁴ For good illustrations of these criticisms consult the essays by Kenneth Waltz, 'The Stability of a Bipolar System,” Daedalus (Summer 1964), 881–909, and “Inter-lational Structure, National Force, and the balance of World Power,” Journal of international Affairs, xxi (No. 2, 1967), 215–31Google Scholar.

⁴⁶ It is important to stress this point because of Russett's pretensions about the ‘science’ of international politics that has been burgeoning in recent years. On this subject see the Preface, especially p. viii.

⁴⁶ For a clear statement of the disadvantages of this course consult Levy.

⁴⁷ For an exceptionally forceful and succinct statement of these rules see Ibid. Russett himself seems to have moved toward agreement with this argument in the period since the publication of International Regions and the International System. See, for example, his Introduction to his excellent collection of essays dealing with economic models that may prove useful in the analysis of international phenomena: Russett, Bruce M. ed., Economic Theories of International Politics (Chicago 1968)Google Scholar.

⁴⁸ For an extended treatment of the relevance of various epistemological problems see Young, Systems of Political Science, especially chap. 7.