Social Cycles Research Paper

1. Cycle As A Philosophical Concept

Since ancient times the cycle has been believed to be an ideal concept (Weininger 1920). It signifies complete-ness; all points on a cycle re equal. The wheel of time, eternally going round, is one of the powerful symbols of this kind of thought. Cycles seem to reflect nature quite elegantly: the cycle of life (in modern sociological terms ‘life cycles’), the cycle of four seasons (in moderate climates), and the cyclical movements of the sun and stars around the earth (in a Ptolemean view). It is thus not surprising that it took quite a long time for evidence-based science to overrule cyclical concepts so tightly connected to esthetics like Ptolemean worldview.

Christianity marks a counter-philosophy to a strict eternally cyclical worldview: Christ is a singularity, after him nothing will be as it was before. The spell of the eternal wheel of time hence is broken. Christian rebirth is, in contrast to other religions, not a cycle; it is but one step up to an end state. Some philosophers thus believe Christian noncyclical thinking to be among the elements of social and individual progress (Guardini 1946). However, Christian thinking does not exclude cyclical concepts. Christ could be thought of as being a singularity, separating one big cycle from the next.

Another strong symbol of societal and human dynamics is the straight, often directed, line. Linear, noncyclical dynamics of societal developments is one of the basic concepts of modernity and progress. Finally, a sequence of stages is also among the influential philosophical concepts, e.g., Kohlberg’s or Piaget’s stages in the ethical development of children, adapted by Habermas for societal development (Habermas 1989, Kohlberg 1981, Piaget 1970).

All three ideal concepts, cycle, linear trend, and stage, represent directed dynamics in contrast to stochastically, nondirected fluctuations. Thus they can be used to explain a given societal situation as being embedded in a specific movement towards an end or with a ‘known’ outcome (next phase going up, going down, coming back). They allow the sequence of historical events to be given an inherent and consistent meaning. Linear dynamics is inherent in socialist and capitalist ideologies, but not necessarily part of social realities (cf. Hader and Mohler 1995 on cyclical worldviews in the GDR). Modern cyclical philosophies, like those based on Friedrich Nietzsche or Oswald Spengler, hence can be understood as counter-attacks on modernism and its Christian roots by conservative intellectuals (A. Mohler 1972 Wendorff 1980).

2. Cycles In Sociological Theory

Sociology is, by definition, a modern science. Ideal societal cycles are seemingly concepts of premodern thinking. Thus it would be hard to find any modern sociologist who explicitly subscribes to cyclical movements. The term is almost absent in modern textbooks. However, reading through influential books on modernizing organizations and companies one finds quite often fourfold sequences of development, i.e., organizational dynamics (Senge et al. 1994). Whatever the labels used are, these fourfold tables resemble quite strongly Bales’ concept of problem-solving sequences in social groups (Bales 1953, Bales and Strodtbeck 1953) which in turn relates to Schumpeter’s classification of business cycles (1939). Based on empirical observation, Bales proposed a four-stage dynamic for small group problem solving: expressive, adaptive, instrumental, and integrative. Parsons adapted this scheme and transformed it into his famous AGIL scheme (Adaptive, Goal Attainment, Instrumental, Latency; Parsons 1971). Thus one of the most influential modern sociological theories takes advantage of a cyclical model in explaining societal developments. However, other than Bales for small groups, Parsons did not present strong empirical evidence for his assumption of similar dynamics on the societal level. While the AGIL scheme has been taught to many thousands of students during the latter half of the twentieth century, its assumed cyclical societal dynamics has had little influence on actual social re-search.

3. Cycles In Economic Theory

By contrast, in economics, cyclical dynamics has a long pedigree and was widely discussed (Forrester 1977, Gordon 1952, Gulap 1989, Kleinknecht et al. 1992, Metz 1984, Wallerstein 1980). However, one should differentiate between ‘cyclical movements’ and ‘cycles.’ The former refers to fluctuations but not to ideal, recurring cycles. It is this rather ambiguous concept which is mostly used in the literature and the media.

Among the most prominent economic cycles, in the sense of an ideal concept, are Kondratieff’s long waves in economic life (1926). These were understood by the Soviet government as a counterattack on the Marxian linear progress concept. Consequently they sent Kondratieff to the Gulag. Schumpeter (1939) in turn developed a complex system of interlaced and super- positioned cycles with Kitchin cycles as the shortest (average three years), followed by Juglar cycles of seven to 11 years, and Kondratieff cycles with 49 to 60 years wavelength.

Today ideal cycles in the Kondratieff and Schumpeter tradition are discussed in relatively small economic circles. Sources for information on cyclical socioeconomic thinking are: Long-Wave Theories of Development (http://faculty.washington.edu krumme/207/development/longwaves.htm) and the Longwave and Social Cycles Resource Centre (http://www.1-888.com/longwave/lwintro.html). However, the relative success of cyclical concepts in economics compared to sociology reflects observed ups and downs in economic development coupled with long empirical observations of macroeconomic characteristics like price indices, average growth rates, interest rates, etc. It is also arguable that the prevalence of cyclical thinking in economics is itself cyclical. In a phase of economic growth, linear concepts are prevalent (accepting few setbacks) without seriously investigating the next turning point. As soon as the economy slides into recession cyclical concepts are ‘re-invented’ by predicting the next turning point into an upward development.

4. Critical Issues In Identifying Cycles

Ideal concepts like cycles require sophisticated theoretical arguments to convince skeptics, and special techniques to extract cycles from the usual mass of data. Critical issues in defining cycles thus are: identifying periods with homogeneous dynamics (i.e., no singularity disrupting a cyclical movement), inter-lacing and/or superposition of cycles of different wavelength, combination of cycles and linear trends, identification of wavelengths and their respective measurement.

Identifying periods of identical wavelength is crucial for cyclical theories. Otherwise there would be eternal cycles, which are counter-evident even to everyday observations (cf. the change in the generational sequence due to higher life expectancy in the last century). Whether a certain historical event is a singularity or just a very prominent event within an otherwise homogeneous cycle is an important issue considering long social cycles. For instance, did the end of WWI and the beginning of WWII represent singularities separating distinct cycles? If so, there would be no time for a full Kondratieff cycle of at least 48 years length between the two events. How does one interpret the interlacing or superposition of different cycles? Are the cycles really a fine-tuned system of small and big cycles, as Schumpeter assumed? What happens if big cycles (long waves) are not multiples of small cycles? If uneven movements due to nonsynchronous multiple cycles in organizations, cohorts, institutions, business, the arts, education, science, welfare, etc. are at work, can one in the end identify the grand cycle? Or, will it be that due to the individual cyclical dynamics the overall outcome will result in unique cyclical superposition at each and every point in time (cf. the discussion between P. Ph. Mohler 1987, 1989, and Namenwirth and Weber 1989). Finally, as is the case with a straight line plotted through a cloud of data points, a cycle is an analytical construct. Stochastic fluctuations are filtered and smoothed to get as close as possible to the ideal concept of a pure sine wave (i.e., cycle, Namenwirth and Weber 1987, Allen 1997, Drucker and Wesche 1999). Because of the strong underlying, non-modern theoretical assumptions, models using cycles as a heuristic are much more controversial than the linear models used in everyday statistics.

5. An Example Of Empirical Analyses Of Cycles In Societal Development

Instead of the numerous and often less theoretically anchored economic analyses, a strict sociological approach is presented here as an example of identifying cyclical dynamics. Namenwirth and Weber showed in their seminal work on Dynamics of Culture, how to use sophisticated methods parsimoniously to establish an empirical model of the real world (1987).

It is obvious that a direct translation of small group dynamics into societal dynamics must fail due to the different time scales at work (Luhmann 1973, Wendorff 1980). Moreover, Bales’ work is based on laboratory experiments only. Such experimental settings make the standardizing of situations and filtering of noise much easier than is the case in observing real situations. Thus any empirical investigation into the validity of Parson’s assumption that the dynamics of society as a whole is similar to that of the small group has to overcome serious technical and methodological obstacles.

Any observation of a cycle requires at least a full cycle; better, a series of cycles. If societal dynamics works on a time schedule similar to that observed for economic cycles, then to observe one full long Kondratieff cycle requires time series of about 60 years. Considering that systematic sampling-based interviews only came into existence in the 1920s and assuming that enough comparable data is available, early twenty-first century sociologists have barely enough data to analyze one full Kondratieff cycle. However, Namenwirth and Weber found even longer cycles in their data (about 120 years), thus it will be the middle of the twenty-first century before their results can be verified with robust survey data. Moreover, for such long time series, the society observed should keep its basic identity over the period of observation, i.e., the time series should not be interrupted by a singularity.

As Namenwirth showed as early as 1973, there are indeed both time series and societies at hand, which allow for an empirical test of Parson’s assumption. However, as we will show, all observed dynamics are a superposition of both linear tends (modernism) and cyclical dynamics. Namenwirth’s solution to the data problem was to take textual indicators measured on a group level instead of individualistic interviews. In doing so, he freed long-term sociological investigation from the regression to interviews as its only source of information. He also avoided the individualistic fallacy of inferring from individual level data to societal level dynamics. As a data source he used party platforms of the two leading American parties since 1844. He found two cycles, one of about 152 years (actually longer than the observed period of 120 years), and one of about 48 years (Namenwirth and Weber 1987). Weber refined Namenwirth’s technique by separating linear trends and cyclical dynamics in his analysis of ‘speeches from the throne 1689–1972’ (statements of the government read by the monarch).

In additional to cycles similar to those found by Namenwirth in the United States, his data indicated a break singularity in the dynamics at about 1795— note the coincidence with the French Revolution. Thus he found two periods with different cyclical dynamics which he named ‘mercantilist’ and ‘capitalist.’ However, studies using equivalent measures and similar approaches did not reveal cyclical dynamics (Eisner 1991 for Switzerland, and P. Ph. Mohler 1987 for Germany). Finally in a reanalysis of Sorokin’s famous data on cultural changes, Klingemann et al. (1982) found strong evidence for long-term trends plus long waves of philosophical dominant thinking. In accordance with the above observation of the general absence of cyclical theories in the social sciences, the data showed an extremely strong (linear) trend for empirical philosophical theories since 1720, i.e., since then the predominant development in philosophical thinking has been the linear increase of a specific of philosophy over a long period of time.

The techniques developed by Namenwirth and Weber for identifying cyclical dynamics are non- standard in time series analysis. Other than the rather complex mathematical models used there, they use two straightforward strategies: first, they check for a linear trend in the time series of a single variable; second, they test for a short cyclical dynamic in the residuals of the de-trended data; and third, they search for a long trend in the residuals of the residuals of the de-trended data. Because the target is the wavelength, they do not attempt to estimate the magnitude of the cyclical dynamics; thus the amplitude is always set to ‘1’.

In essence, this approach is a special form of residual analysis. Translated into theoretical concepts: first, a linear dynamic is assumed, which is, second, under- pinned by one or two cyclical dynamics. These cycles can be interpreted within an AGIL-like pattern. Moreover, Weber’s observation of a break between mercantilist and capitalistic dynamics may be taken as an indicator of a singularity in a society (nothing will ever be like it was), which allows differentiation between epochs like Bales might differentiate between different social groups.

In doing so, this approach re-specifies non- directional theories (like Parsons,) into directed ones. It also bridges the gap between linear modernism and cyclical conservatism. Modernity is the linear (but not infinite) dynamic in societies, while the cyclical elements of the moving process point to basic laws of problem solving (assuming that problems have to be solved, if at all, in a finite period of time). Singularities, like the one between mercantilism and capitalism in 1795, make us aware that there is no infinite continuous movement in societies.

6. Conclusion

Cycles and straight lines are two idealizations of directed movement (dynamics). Linear trends, be they directed upwards or downwards, have predominated in sociological thinking to the end of the twentieth century. Cyclical theories seem to be absent in the grand social theories, but they are not. They are ever present in schemata like the famous Parsonian AGIL scheme. Cycles are, often without mentioning the concept, inherent in almost all contemporary approaches dealing with organizational and group changes. This indicates the robustness and validity of Bales’ early findings. Survey data will not be available for a long time as a base for analyzing societal dynamics, be they linear, curved, or even cycles. Thus one must rely on other data to model the dynamics of a given society.

In economics, long waves are a dated concept. The concept of business cycles, however, is widely used to identity regularities within the fluctuations of economic indictors.

Bibliography:

1. Allen D S 1997 Filtering permanent cycles with complex root units. Federal Reserve Bank Working Paper 97-0014A, St Louis, MO
2. Bales R F 1953 The equilibrium problem in small groups. In: Parsons T, Bales R R, Shils E (eds.) Working Papers in the Theory of Action. Free Press, New York
3. Bales, R F, Strodtbeck F L 1953 Phases in group problem solving. Journal of Abnormal and Social Psychology 46: 485–95
4. Drucker M Wesche K 1999 European business cycle: New indices and analysis of their synchronity. Federal Reserve Bank Working Paper 99-019A St Louis, MO
5. Eisner M 1991 Politische Sprache und sozialer Wandel. Eine quantitativ-semantische Analyse von Neujahrsleitartikeln in der Schweiz von 1840–1987. Dissertation, Zurich
6. Forrester J W 1977 Growth cycles. De Economist 125(4): 525–43
7. Guardini R 1946 Der Heilsbringer, in Mythos, Offenbarung und Geschichte. Thomas, Zurich, Switzerland
8. Gulap H 1989 The stages and long-cycles of capitalist development. Review of Radical Political Economics 21(4): 83–92
9. Habermas J 1989 Moral Consciousness and Communicative Action. Polity Press, Cambridge, UK
10. Hader M, Mohler P Ph 1995 Zukunftsvorstellungen der Menschen als Erklarungsvariable fur die Krise der DDR und die gegenwartige Situation in Ostdeutschland. Politik und Zeitgeschichte 27: 19–27
11. Kleinknecht A, Mandel E, Wallerstein I (eds.) 1992 New Findings in Long-Wave Research. Macmillan, London
12. Klingemann H D, Mohler P Ph, Weber R P 1982 Cultural indicators based on content analysis: A secondary analysis of Sorokin’s data on fluctuations of systems of truth. Quality and Quantity 16: 1–18
13. Kohlberg L 1981 The Philosophy of Moral Development: Essays on Moral Development. Harper and Row, San Francisco, Vol. 1
14. Kondratieff N D 1926 Die langen Wellen der Konjunktur. Archi fur Sozialwissenschaft und Sozialpolitik 56: 573–609 [first English partial translation 1935: The long waves in economic life. Review of Economic Statistics 17: 105–115. Full English translation 1979: The long waves in economic life. Review 2: 519–62]
15. Luhmann N 1973 Weltzeit und Systemgeschichte. In: Ludz P C (ed.) Soziologie und Sozialgeschichte. Kolner Zeitschrift fur Soziologie und Sozialpsychologie. Special Issue 16: 81–97
16. Mohler A 1972 Die konser ati e Re olution, 2nd edn. Wissen-schaftliche Buchgesellschaft, Darmstadt, Germany
17. Mohler P Ph 1987 Cycles of value change. European Journal of Political Research 15: 155–65
18. Mohler P Ph 1989 Cyclical ideology: A response to Namenwirth & Weber. European Journal of Political Research 17: 121–4
19. Namenwirth Z J 1973 The wheels of time and the inter-dependence of value change. Journal of Interdisciplinary History 3: 649–83
20. Namenwirth Z J, Weber R P 1987 Dynamics of Culture. Allen & Unwin, Boston
21. Namenwirth Z J, Weber R P 1989 Cycles of the fourth kind: A reply to Mohler’s cycles of values. European Journal of Political Research. 17: 118–20
22. Parsons T 1971 The Social System. Free Press, Glencoe, IL Piaget J 1970 Piaget’s theory. In: Mussen P H (ed.) Carmichael’s Manual of Psychology. Wiley, New York
23. Schumpeter J A 1939 Business Cycles. McGraw-Hill, New York
24. Senge P M, Kleiner A I, Roberts Ch, Ross, R B, Smith B J 1994 The Fifth Discipline Fieldbook. Strategies and Tools for Building a Learning Organization. Currency, New York
25. Weininger O 1920 Uber die letzten Dinge, 6th edn. Braumuller, Vienna
26. Wendorff R 1980 Zeit und Kultur. Westdeutscher Verlag, Opladen, Germany