Long Waves and Historical Generations: A World System Approach Carl H.A. Dassbach 22 November 1993 Department of Social Sciences Michigan Technological University Houghton, MI 49931 Bitnet: DASSBACH@MTUS5 Internet: DASSBACH@MTUS5.CTS.MTU.EDU Phone: (906)487-2115 Fax: (906)487-2468 Part of the enduring attractiveness of the concept of generations is that it links collective character, life cycle, and social events. According to the premises of the Mannheimian tradition in generational research, the character of a generation is not simply a product of geography and the proximity of birth dates. Rather, the formation of historical generations must be understood in terms of what C. Wright Mills(1959) has called , the "intersection of biography and history." Concretely, this means that major political, economic and social events such as periods of prosperity, depressions and wars are central to the formation of the collective consciousness or entelechy of historical generations. Still, despite Mannheim's (1952) stress on the importance of both geography and "significant social events," their relationship to generations has not been analyzed in a systematic fashion. Generational researcher have largely seen geographical location as coincidental and "significant social events" have been viewed, with a single exception (Kearl and Hermes 1984), as discrete occurrence. Given this lacuna in generational theory, this paper will examine if there is any relationship between spatially and temporally patterned `events,' namely, long or Kondratieff waves viewed from a world system perspective, and historical generations. A world system approach to long waves is especially appropriate for this endeavor precisely because it locates `significant social events," i.e., long term period of expansion and contraction, in a temporally and spatially patterned framework. As such, a consideration of generations from this perspective may permit new insights into important questions for generational research such as: when do distinct generations emerge, why do they emerge and why do they emerge in different parts of the world at different times. Not only would generational research benefit from a consideration of long waves but long wave research might also benefit from a consideration of generations. But, with the exception of Screpanti (1989),[1] studies of social movements and long waves have focused on class struggle (Mandel 1980, Cronin 1980, Goldstein 1988, Screpanti 1984, 1987, Silver 1991, 1992) and ignored generational phenomena. While both long wave and generational research would benefit by considering one another, this paper will focus on the insights a world system theory of long waves can bring to the study of generations. The basic premises and concepts of world system theory will not be repeated: they are well known.[2] Instead, the first section of this paper will discuss major theories of long waves and generations. Each of the two distinct theoretical approaches to both long waves and generations will be reviewed in order to identify which are the most promising for considering their interrelation. In the next section, data on generational movements in the 19th and 20th in relation to phases of long waves and regions of the world economy will be examined. The concluding section will discuss the results and problems of this analysis. Long Waves and Historical Generations. Although the Russian economist N.D. Kondratieff was neither the first to observe nor discuss long term fluctuations in the economies of the industrialized nations, his work in the 1920's is generally seen as beginning the systematic and sustained discussion of long waves.[3] Over the last sixty years, two fairly distinct approaches to the study of long waves have emerged. One approach maintains that long waves are simply long term movements in price levels reflective of changes in the supply of gold and/or money.[4] While this approach has several adherents and has been important in the dating long waves, it is not useful for examining the relationship between long waves and generations because monetarists view long waves as nominal. In other words, monetarists see long waves as inflationary and deflationary `ripples' in the monetary life of society, without the deeper social, political and economic consequences which are decisive for the formation of historical generations. The second tradition in long wave research (Arrighi 1989, Schumpeter 1964, Mandel 1975, 1980, Wallerstein 1984, Goldstein 1988, and Gordon 1980) maintains that long waves are real. They represent long term economic, political and social `fluctuations' of the world economy as a whole.[5] Although all who work in this tradition assert that long waves are the result of recurring mechanisms, there is considerable debate over precisely which mechanisms `cause' long waves. For example, Schumpeter stresses innovation, Arrighi focuses on alternating phases in the competitive struggle between enterprises, Kondratieff points to the periodic replacement of capital goods, Mandel emphasizes fundamental changes in motive technology, and Wallerstein maintains that long waves are the result of periodic imbalances between world supply and world demand. Despite disagreements over causality, this tradition sees long waves as consisting of two distinct phases of approximately 20 to 30 years in duration: a period of expansion, known as an "upswing" or "A Phase," and a period of contraction, "downswing" or "B Phase."[6] Most authors also associate distinct processes and/or phenomena with each phase of the wave. Upswings, at least in core areas of the world economy, are associated with rising incomes, a growth in employment, a buoyant economy, the expansion of credit, increased investments, and the creation of new firms. Downswings are associated with the opposite conditions: contracting incomes, increased unemployment, a depressed economy, a `credit crunch,' decreased investments, and the elimination of firms. It should be noted however, that the terms "upswing" or A Phase and "downswing" or B Phase only refer to the overall characteristic of the period. Recessions also occur during A Phases but these are brief and, in general, "good times" predominate over bad during A Phases. Likewise, B Phases may be punctuated by periods of prosperity but these are brief so that the overall economic complexion of the period is stagnant or sluggish. This approach to long waves maintains that long waves represent the basic tempo of economic, political and social transformation. Schumpeter(1939), Gordon (1980) and Arrighi (1989) have suggested that each wave represents a distinct phase in the development of the world economy so that the society that "emerges" at the end of a wave is substantially different from the same society at the beginning of the wave. For this tradition, in other words, the ebb and flow of long waves represents "significant social events" in the broadest sense of the term.[7] Two fairly distinct approaches also characterize the study of generations. One, known as either the "pulse rate" model (Jaeger 1985), "interaction theory" (Huntington 1974) or "positivist" approach (Mannheim 1952), adopts a mechanistic perspective on historical generations, especially in explaining social and political change. Tracing itself back to the work of August Comte and his disciple J.S. Mill, this approach claims that the biological succession of generations produces both social continuity and social change. The succession of generations insures social continuity because the younger generation receives its values, attitudes and dispositions from the previous generation. At the same time, the younger generation does not simply adopt preexisting values, it also modifies them. Because the younger generation must wait for the death or passing from power of the previous generation before it implements its values, social change is based on the biological rhythm of life manifested in the succession of generations, and occurs with a fixed periodicity of ca. 20 to 30 years. The major weakness with this approach is that it equates and therefore conflates biological and historical generations. It assumes that because biological generations (based on a proximity of birth dates) succeed one another,[8] that historical generations also succeed one another. But historical generations are not identical to biological generations: what distinguishes the former from the latter is a generational consciousness or entelechy. Biological generations do not require a distinct consciousness to exist, historical generations can not exist independent of this consciousness. Hence, biological generations are successive: historical generations may not be successive. Instead, historical generations develop when a group of individuals share, at a certain point in their life course, a distinctive set of experiences resulting from significant social events. By denying the importance of significant social events in the formation of a generation's entelechy, the pulse rate models denies the historical character of generations. The second approach to the study of generations, known as the "imprint hypothesis" (Jaeger 1985), "experiential theory" (Huntington 1974), "entelechy" approach (Mannheim 1952) or "historical generational model"(Braungart 1984a) was first explicitly formulated by Mannheim (1936, 1952).[9] It has been subsequently refined and developed by several authors but the works of Braungart(1984a) and Braungart and Braungart (1984, 1986, 1989) are especially significant. This approach holds that generations are not simply coevals and that the biological duration of life is relatively unimportant in explaining social change. Instead, the crystallization of a generational consciousness, or entelechy, is the result of individuals, between the ages of 17 and 25, sharing the experience of significant historical events within a bounded geographical area. Because events of a sufficient magnitude to result in the crystallization of a generational entelechy do not regularly occur, there is no necessary succession of historical generations. According to Braungart (1984a, 115-116) During so-called routine periods, youth cohorts come and go.., processes of socialization and role allocation serves as linkages between the young and social options, [and] youth are incorporated in the adult structure without major incident. However, significant historical changes can threaten the fragile character of these linkages and disparities may emerge between youth and elders, as well as within their own youthful age group. For this perspective, the existence of generations and generational consciousness can not be taken for granted, they must be demonstrated through an analysis of formative events, coevals, and their activities. What this brief review demonstrates is that one can not simply and uncritically lump together any variant of long wave and generation theory. Because long waves are "nominal" for the monetarist tradition, this view of long waves can not provide any insights into the formation of historical generations. Likewise, the "pulse rate" model of generations, although amendable to long wave explanations because of its fixed periodicities (see Screpanti 1989), is unsatisfactory because it assumes precisely what should be demonstrated: the existence of historical generations. Historical generations can not be assumed: they must be discovered and explained. Discovery entails identifying a distinct consciousness or entelechy associated with the generation.[10] Explanation entails identifying both the geographical sites and the significant social events which have led to the formation of this consciousness and hence, of a historical generation. Most theorizing and discussion of generations in the entelechy tradition is quite adept with `discovery.' What is more problematic for this tradition is the problem of explanation, i.e., identifying the sites and significant social events that have produced the generational consciousness. Often this is simply assumed and not explained; and when it is explained, many of the explanations are ad hoc. A notable exception is Kearl and Hermes (1984). Like many others,[11] they note attitudinal and value differences between parents and children, as well as attitudinal and value similarities between grandparents and grandchildren. But unlike other approaches to the questions of these attitudinal differences and similarities, Kearl and Hermes attribute the differences between parents and children as well as the similarities between grandparents and grandchildren to phases of long waves. [T]he sociopolitical ethos associated with the time one comes to age ...produces [a]... historical imprint on the collective conscience of a generation. .. [T]he generational schema of those cohorts socialized within a particular phase of this Kondratieff wave will bear similar outlooks... [so that] ... the value similarities observed between skipped lineal generations can be explained by their lives common intersection with the fifty-year cycle of economic, demographic, social and political events. (Kearl and Hermes 1984, 262-263) Long Waves, Regions of the World Economy and Historical Generations. If long waves can provide some insight into attitudinal differences and similarities between generations, a world system perspective on long waves may be capable of providing some insights into the emergence of distinct historical generations at different locations in the world. Examining this proposition initially entails establishing a dating for long waves and identifying the existence and dates of distinct generational movements around the world. Despite some minor disagreements about the dates of the upper and lower turning points, most long wave analysts would agree with Goldstein's (1988, 67) dating for A and B phases of long waves during the last 200 years. Braungart (1984b, Table 2) has identified forty-one distinct generational movements in various parts of the world from the beginning of the 19th c. to 1980. Because Goldstein's dating is derived mainly from research into the core region of the world economy, only the relationship between long waves and core generational movements identified by Braungart will be initially examined.[12] Table 1 here (can't be converted ASCII) Table 1 indicates a relationship between core generational movements and phases of long waves. Specifically, 91% (10 of the 11) of the generational movements between 1800 and 1980 in core region of the world economy occur during B Phases. Although, as already noted, long wave datings are primarily based on the core, the results in Table 1 suggests that it may be fruitful to examine the relationship between non-core, i.e., peripheral and semi-peripheral, generational movements and long waves. Table 2 examines this relationship. Table 2 here (can't be converted into ASCII) Table 2 shows that the relationship between long waves and generational movements in non-core areas is the inverse of core areas. Where the majority of generational movements in the core occurred during B phases, the majority of generational movements in non-core regions, i.e., 26 of 30 or 87%, occurred during A Phases. The general correspondence between phases of long waves and generational movements in specific regions of the world economy suggest that further analysis is in order. In this context, it is useful to consider some of conclusions from the study of long waves and class struggle, especially regarding the timing of class struggle and phases of long waves (Mandel 1980, Cronin 1980, Goldstein 1988, Screpanti 1984, 1987, Silver 1991, 1992). These studies indicate that class struggle is not evenly distributed throughout a phase of the long wave. In core areas, class struggle tends to intensify during the latter portion of an A Phase and peaks either near the end of the phase or during the "T" or "transition phase" between two phases of the long wave. Some of these studies also conclude that class struggle is most intense in the "T" phase from A to B than either the preceding A Phase or the following B Phase and less intense in the T Phase from B to A than the preceding B or following A Phase.[13] Table 3 examines the data on historical generations based on these considerations in order to determine if they cluster during a specific portion of a phase or during certain T Phases. Table 3 here (can't be converted into ASCII) Table 3 shows that with the exception of the T Phase from 1968 to 1975, generational movements do not appear to cluster during either some portion of an A or B phase or during T Phases. Conclusions While Table 2 shows that generational movements can occur at any point in an A or B phase and do not cluster near the end of a phase or during a T Phase, the findings from Tables 1 and 2 are significant. These tables show a strong relationship between generations, long waves and regions of the world economy. Of the 41 generational movements in the world between 1800 and 1980 identified by Braungart, 91% of the core generational movements occurred during a B Phase and 87% of the non-core (peripheral and semi-peripheral regions) generational movements occurred during an A Phase. The prevalence of generational movements in the core during B phases is consistent with studies which demonstrate a connection between economic experiences and political attitudes such as Inglehart (1971, 1977, 1987) or Elder (1974) as well as older discontinuity explanations of social movements (Smelser 1963, Kornhauser 1959). Inglehart (1971, 1001), for example, maintains that "the degree of economic security that an individual felt during his formative years ... play[s] a key role in shaping his later political values." In the case of the core, it could be hypothesized that individuals who "come to age" under conditions of prosperity develop a distinctive "welt- anschauungen," largely independent of class position, and generational movements are a collective reaction to the stagnant and sluggish economy associated with a B phase. While these findings confirm observations about specific generations, such as the "children of the Depression," they also add to our understanding of generations by specifying one general type of shared "significant social event" which leads to the formation of historical generations, namely, the experience of an economy moving from expansion to stagnation. This was, in fact, implicitly suggested by Mannheim (1952, 303) when he argued that a generation can only become actualized if a "concrete bond" is created between individuals by "being exposed to the social and intellectual symptoms of a process of dynamic de-stabilization." The relative absence of generational movements in the core during A phases is also significant. It suggest that the experience of an economy moving from stagnation to expansion for the generations which comes to age during the B phase does not have the same impact as the movement from expansion to stagnation for the generations which comes to age during the A phase. Admittedly, the generations examined had different purposes and consisted of diverse generational units - as a result of geography, culture, and national character - but the fact remains that there is a strong relationship between the emergence of historical generations and periods of economic stagnation in the core. Explaining the relationshop between generational movements in the non-core and B phases is somewhat more problematic and three explanations are possible. One, generational movements in the non-core follow the same trajectory as class struggle in the core, i.e., intensifying during A Phases. Second, A and B phases have different, if not inverse effects, in the non-core and core. This has been suggested by several long wave researchers (Frank 1967, Rostow 1978, Research Working Group, 1979, Wallerstein 1984, Boswell 1989). In other words, during A Phases the core area undergoes expansion and the non-core undergoes stagnation and vice-versa.[14] If this were the case, it would explain why generational movements in non-core regions are associated with A phases. It would also suggest that the causal mechanisms for non-core generational movements during A phases of the world economy are similar to the casual mechanisms for core generational movements during B phases. A third explanation is that A and B phases have similar effects in core and non-core, namely economic growth, but the consequences of growth vary according to regions. In the core, economic expansion promotes economic and political stability and a redistribution of income while in the periphery and semi- periphery, rapid economic growth is "disequilibriating and full of outrageous corruption and dispossession of some would-be elites" which "results in angry generational movements."[15] Hence, while B phases in the non-core may also be periods of a stagnant economy, economic stagnation is less unsettling and thus less likely to produce generational movements than the conditions associated with economic expansion in an A phase. Although many of relationships between generational movements, long waves and regions of the world economy identified in this paper are consistent with the literature on both generations and social movements, the conclusion remain tentative, especially regarding the coincidence of non-core generational movements and A phases. This is mainly due to the uncertainty about the effects of A and B phases in non-core regions. Clearly, further research is needed on this topic and the results from this study may, in fact, be germane to this question. A second difficulty with these conclusion is the limited data on generational movements. While Braungart's compilation is wide ranging, not all generational movements in all counties of the world are included. A in-depth compilation of generational movements worldwide since 1800, similar to what the World Labor Research Group at the Fernand Braudel Center has compiled on capital-labor conflicts (Silver 1992) would be useful in order to further test whether there is a relationship between generations and regions of the world economy and long waves. Notes 1. Screpanti 1989, raises the issue of generations in relation to long waves but relies on a "pulse rate model" (discussed below) leading him to suggest that long waves may be related to the duration of an individual's "productive life." 2. Although Hopkins and Wallerstein have modified their position on certain issues, the best general statement of the World- System approach can be found in: Hopkins and Wallerstein, 1977. For a more recent summation of the World-System perspective, endorsed by Wallerstein, see Shannon 1989. 3. An excellent summary and discussion of the long wave literature can be found in Goldstein 1988. 4. For an annotated bibliography of the early work on long waves, especially the monetarist tradition which is not extensively discussed in Goldstein 1988, see Barr 1979. 5. Among those who accept the existence of long waves, there is a debate about whether they existed prior to 1790. Generally, Marxists deny their existence prior to 1790 because they associate long waves with industrial capitalism. For recent empirical works which demonstrates the existence of long waves prior to 1790, see Thompson 1992, Boswell and Misra forthcoming, Goldstein 1988, and Dassbach, Davuytan, Dong and Fay 1993. 6. Here it should be noted that some long wave theorists such as Schumpeter 1939, advocate a four phase model of long waves - expansion, recession, depression and recovery - but this can also be understood as two distinct phases consisting of recovery/ expansion and recession/depression. 7. One example of this view of long waves would be to compare the period from 1945 to ca. 1970 to the period after 1970. 8. Even the assumption that distinct biological generations succeed one another is problematic and one of the criticism frequently raised against this tradition is: How does one distinguish discrete generation from the continuous flow of human life? Ortega y Gasset's proposal, that Descartes' 23rd birthday marks the beginning of the first generation of modernity and subsequent generations can be dated in 15 years intervals after Descartes' 23rd birthday, does not solve this problem. See, Marias, 1970. 9. Most discussions of Mannheim acknowledge the importance of "The Problem of Generations" (1952) but I maintain that important insights into Mannheim's discussion of generations can be found in Ideology and Utopia (1936). I also maintain that Mannheim's understanding of generations is significantly different from both Ryder's concept of a cohort and Rintala's concept of a political generation with which it is often equated. I have discussed this in depth in Dassbach, 1992. 10. See, for example, Wohl 1979 or Spitzer 1987. 11. See Kearl and Hermes (1984) for a discussion of the relevant literature. 12. The assignment of different counties to core or non-core, i.e., periphery and semi-periphery, is based on Arrighi and Drangel 1986. 13. I have borrowed this formulation from Silver 1992, 284. With the exception of the T phase from 1848 to 1853, the datings of the other T phases are from Screpanti 1984, 521. 14. Why this should occur is less clear. One possible explanation can be extrapolated from Rostow's characterizations of A and B phases. According to Rostow 1978, during what is generally called an A Phase the terms of trade in the exchange between core and non-core favor manufactured goods over primary and agricultural goods while during a so-called B phase the terms of trade favor primary and agricultural goods. Hence, during an A phases imports of primary and agricultural products for the core are relatively "inexpensive" (for the core) which means that the non-core is getting relatively little value in return for its products. During B phases, on the other hand, primary and agricultural products are relatively "expensive" (for the core) which that the non-core is receiving more "value" for its products, hence the economy of non-core regions is strong and more buoyant. 15. I am indebted to my colleague Joe Heyman, for this third formulation of the possible effects of A and B phases in non-core areas. Works Cited Arrighi, G. 1989. "Custom and Innovation: Long Waves and Stages of Capitalist Development." in: DiMatteo, Goodwin and Vercelli, editors, Technological and Social Factors in Long Term Fluctuations. Berlin: Springer Verlag. 89-113. Arrighi G. and J. Drangel. 1986. "The Stratification of the World System." Review. 10, 1. Barr, K. 1979. "Long Waves: A Selective Annotated Bibliography," Review. II, 4, Spring, 675-718 Boswell, T. and J. Misra. forthcoming "Cycles and Trends in the Early Capitalist World-Economy: An Analysis of Leading Sector Commodity Trade, 1500-1600/50-1750." Review. Boswell, T. 1989. "Colonial Empires and the Capitalist World- Economy: A Time Series Analysis of Colonization, 1640-1960." American Sociological Review. 54, April, 180-196. Braungart, R. 1984b. "Historical Generations and Generation Units: A Global Pattern of Youth Movements." Journal of Political and Military Sociology. 12, 113-135. Braungart, R. 1984a. "Historical Generations and Youth Movements: A Theoretical Perspective." in Research in Social Movements. V.6. Greenwich, CT: JAI Press, 95-142. Braungart, R. and M, Braungart. 1986. "Life Course and Generational Politics." Annual Review of Sociology. 12, 205- 231. Braungart, R. and M, Braungart. 1989. "Political Generations." in: Research in Political Sociology. V.4, Greenwich, Ct.: JAI Press, Inc, 281-319. Braungart, R. and M, Braungart. 1984. "Generational Politics." Micropolitics. 3, 3, 349-415. Cronin, J.E. 1980. "Stages, Cycles and Insurgencies: The Economics of Unrest." in: Hopkins and Wallerstein. Editors, Processes of the World-System. Beverley Hills: Sage Publications. Dassbach, C.H.A. 1992. "Historical Generations and Political Generations: Problems of Conceptualization and Analysis." Draft. Dassbach, C.H.A., N. Davuytan, J. Dong and B. Fay. 1993. "Long Waves Prior to 1790: A Modest Contribution to the Study of Long Waves." Draft. Elder, G.H. Jr. 1974. Children of the Great Depression. Chicago: University of Chicago Press. Frank, A.G. 1967. Capitalism and Underdevelopment in Latin America. New York: Monthly Review Press. Goldstein, J. 1988, Long Cycles: Prosperity and War in the Modern Age. New Haven: Yale University Press. Gordon, D. 1980. "States of Accumulation and Long Economic Cycles." in: Hopkins and Wallerstein. Editors. Processes of the World-System. Beverley Hills: Sage Publications. Hopkins, T.K. and I. Wallerstein. 1977 "Patterns of Development of the Modern World-System." Review. I, 2, Fall, 111-145. Huntington, S.P. 1974. "Paradigms of American Politics: Beyond, the One, the Two and the Many." Political Science Quarterly. 89, 1. Inglehart, R. 1987. "Value Change in Industrial Society." American Political Science Review. 81, 4, 1289. Inglehart, R. 1977. The Silent Generation. Princeton: Princeton University Press. Inglehart, R. 1971. "The Silent Generation in Europe: Intergenerational Change in Post-Industrial Society" American Political Science Review. 65, 4, 991-1017 Jaeger, H. 1985. "Generations in History: Reflections on a Controversial Topic" History and Theory. 24, 3, 291. Kearl, M and M. Hermes. 1984. "Grandparents, Grandchildren, and the Kondratieff: Thoughts on "Period Effects" in Intergenerational Analysis." International Journal of Aging and Human Development. 19, 4, 257-265. Kondratieff, N.D. [1926] 1979. "The Long Waves in Economic Life", Review. II, 4, Spring, 519-562 Kornhauser, W. 1959. The Politics of Mass Society. Glencoe, IL: Free Press. Mandel, E. 1975. Late Capitalism. London: New Left Books. Mandel, E. 1980. Long Waves in Capitalist Development. Cambridge: Cambridge University Press Mannheim, K. (1928) 1952. "The Problem of Generations" in: Karl Mannheim. Essays in the Sociology of Knowledge. New York: Oxford University Press. Mannheim, K. (1929) 1936. Ideology and Utopia. New York: Harcourt, Brace and World, Inc. Mar!as, Jul!an. 1970. Generations. A Historical Method. Trans. by H.C. Riley. University: University of Alabama Press. Mills, C. Wright. 1959. The Sociological Imagination. London and New York: The Oxford University Press. Research Working Group, 1979. "Cyclical Rhythms and Secular Trends of the Capitalist World Economy: Some Premises, Hypotheses, and Questions." Review. II, 4, 483-500. Rostow, W.W. 1978. The World Economy: History and Prospects. Austin & London: University of Texas Press. Schumpeter, J. 1939. Business Cycles. 2 Vols. New York: McGraw- Hill Screpanti, E. 1987. "Long Cycles in Strike Activity: An Empirical Investigation." British Journal of Industrial Relations. 25, 1, 99-124. Screpanti, E. 1989. "Some Demographic and Social Processes and the Problem of Kondratieff Cycle Periodicity." in: DiMatteo, Goodwin and Vercelli, editors, Technological and Social Factors in Long Term Fluctuations. Berlin: Springer Verlag. 114-129. Screpanti, E. 1984. "Long Cycles and Recurring Proletarian Insurgencies." Review. 7,2, 509-548. Shannon, R.T. 1989. An Introduction to the World-System Perspective. Boulder: Westview Press. Silver, B. 1992. "Class Struggle and Kondratieff Waves, 1870 to the Present" in: Kleinknecht, Mandel and Wallerstein. Editors. New Findings in Long Wave Research. New York: St. Martin's Press. Smelser, N.J. 1963. Theory of Collective Behavior. New York: Free Press. Spitzer, A. B. 1987. The French Generation of 1820. Princeton: Princeton University Press. Thompson, W.R. 1992. "Systematic Leadership and Growth Waves in the Long Run," International Studies Quarterly. 36, 1, March, 25-48 Wallerstein, I. 1984. "Long Waves as Capitalist Process." Review. 7, 4, 559-575. Wohl, R. 1979. The Generation of 1914. Cambridge, MA: Harvard University Press.