Andre Gunder Frank

On Studying the Cycles in Social Movements

If cycles of protest are such watersheds of social and political change, then why is it that ... we have so few studies of such cycles?"

Sidney Tarrow (1991b:11)

The author of among the most outstanding of the "few" studies on cycles of social/protest movements (Tarrow 1983, 1991a,b) answers his own question: Because they are a moving target, they interweave with institutions, and there are problems with "the way they have been conceived and studied" (Tarrow 1991b:11). We can agree that there are such conceptual problems in the studies of "cycles" of social/protest movements, including those by Tarrow himself.

To help clarify these problems, we should distinguish and seek to answer the following different questions:

1. Are there and do we refer only to life cycles of birth-development-peak- decline-demise within social movements themselves?
2. Are there as well, and can we identify, a wave like pattern of social movements, whose recurrent rise and decline appears "cyclical"?
3. If so, does this wave like pattern meet the criterion of a true cycle in that both the upper and lower turning points are endogenously generated?
4. If so again, is this endogeneity within the cycle of social protest movements themelves, or is the endogeneity of the cycle at least to be found in the needs and opportunities, which are generated by the institutions with which the movements interweave? Or
5. can the cyclical pattern of social protest movements be traced to economic, demographic, generational or other factors that themselves display a recurent wave like or even cyclical pattern of growth and decline, which in turn generartes "cycles" of social protest movements? In other words, how do we explain and account for the "cycles" in and of social protest movements that we may observe?

The answers to these and other related questions are less than clear in the writings of Tarrow and other students of social/protest movement cycles. Tarrow himself hardly distinguishes between the first two questions and never poses the third, at least regarding the cycles of movements. He answers negatively to the 5th question about social movement cycles responding to economic or other "external" cycles. Instead, Tarrow's explanatory efforts are in terms of the 4th questions: He seeks - but we think partly fails - to explain cycles in and of social protest movements through the "political opportunity structure" [POS] generated by and in the mostly political institutions within wich the social movements rise and decline.

Sometimes, indeed mostly, Tarrow speaks to the first question; and he answers it affirmatively, that social/protest movements do have life cycles of their own. Here and there, he refers to the second question and/or refers us to others who have identified recurrent cycles of movements. Tarrow (1991a,b) suggests that social movement [SM] cycles are like business cycles. However, his analogy is NOT well taken. SM cycles are like a "life cycle" of upward, peak, downward, which of course most all SMs do have. So do business cycles. But that is NOT the important and interesting cyclical aspect of Sms or of business cycles. The important aspect of business cycles is that the upper and lower turning points are endogenous to/in the system in which they occur and/or that the up leads to the down and the down leads to the up. That does happen in Tarrow's treatment of BCs, and THAT is why they are cycles; but it does not happen in his treatment of SM "cycles."

We can briefly consider the question of the endogeneity or exogeneity of turning points in terms of a discussion of long economic "Kondratieff" [K] cycles, to which we will also return below. In regard to Kondratieffs, Ernest Mandel for instance argues that only the upper turning point is endogenous [that is the up leads to the subsequent down], but the lower turning point is not. If that is true, then the K cycle is not a true cycle. David Gordon and we among others think the lower turning point is probably also endogenous, that is the down also leads to the up of the K cycle, although in recent writings Gordon now seems to distance himself from this view (Frank, Gordon & Mandel 1992). In the Tarrow version of SM cycles, this whole problematique -and therefore the real cyclicalness of SMs - is almost absent. This problematique is present in our treatment of SM cycles below, but it remains not very satisfactorily resolved. The temporal and causative relation of cycles both in and of Sms to K cycles is in dispute and remains unclear. However, if it cannot be unambiguously established that the ups and downs of K waves generate the ups and downs of SM cycles, or vice versa; we must still identify the causative up and down dynamic behind SM waves and downs, or what makes the turning points endogenous and makes the "cycles" repetitive, be these causes in some other -eg. demographic or generational- cycle, or be they internal to SMs themselves, which would make them true cycles.

A number of students of social/protest movements have inquired into the 5th question above, especially with regard to the relations between social movements and long Kondratieff cycles. However, views differ widely. For instance, Frank and Fuentes (1986) and Fuentes and Frank (1988) suggest that social movements are "more numerous and stronger" in Kondratieff B downward phases. Friberg (1987:2) also sees a historical relation "to so called Kondratieff cycles...protest activity being more pronounced during the downturn" and citing 1815-48, 1873-96, 1914-45,"and the economic downturn after 1970." Moreover, Goldstone (1980) suggests that the incidence of social movements' success "seems to depend heavily on the incidence of broad political and/or economic crisis in the society at large" (cited in Tarrow 1986:46).

Huber (1987, 1988), on the other hand, argues that "the social movements gain strength at the top, upper turning point and decline (stagnation) of a long wave, and to defuse to wider popular circles with the further course of the decline, with which by and by they however also again loose strength. With the transition to a new long wave, they recede into the background insofar as they have not exhausted and undone themselves -- only to reappear again decades later with even greater force". For Huber, periods of dynamic economic expansion to 1815, bourgeois glitter-and-glory 1850-67/73, belle epoque 1890-1910, and economic wonder 1948/52-67/73 "forge reactive resistance and social and ecological problems," which then generate the cause and content of social movements. Nonetheless, Huber also says that economic "system development and social movement occur in mutual relations with each other, simultaneously or with a lag, but in part also independently of each other" (Huber 1988:431).

For Tarrow (1986), however, although "cycles of protest and their implications for change ... do not coincide with economic cycles in any way, protest movements appear to cluster in identifiable periods, and to be associated with substantial policy innovation during such periods." Similarly, Brand (1987) also finds that social movements come and go cyclically, but after comparing them with country-specific Kondratieff ups and downs concludes that "these movement waves coincide not with long-term economic cycles but with recurring waves of tendencies critical of modern civilization" (emphasis in original). Brand finds that in the past two centuries, the first wave of social movements he identifies coincides with the middle of the 1815-48 Kondratieff B downturn phase. The second one was at the turn of the century during the pronounced 1896-1913 Kondratieff upswing. An uncertain "cleft wave" of social movements in the 1920s and 1930s occurred during another Kondratieff B phase. Finally, the present wave of new social movements began at the 1960s upper turning point from the post-war Kondratieff upswing to the present Kondratieff B downswing. Thus, by Brand's reading, "mobilization waves are to be found in both down-swing and up- swing phases as well as at the turning-points of the K-cycles. There is clearly no systematic connection between the two." Van Roon (1988) also fails to find any systematic connection between social movements and either Kondratieff economic cycles or even industrial or other structural transformation. Finally and to complicate matters still more, the second author of this article now argues that we should not put all social movements into one bag, among other reasons, because some move with the A phase and others with the B phase of K cycles, as she will observe below.

Thus, the question of the relation between social movements and economic or other cycles remains in doubt pending further research. Trarrow is probably right when he says

we cannot identify cycles of protest by simply extrapolating to them from normal trends in economic activity. Nor can we predict mechanically the timing of a cycle or its magnitude from the frequencies of past occurrences. Protest cycles resemble politics in general in their uneven and irregular diffusion across time and space. What we can say about cycles of protest is that they are characterized by heightened conflict across the social system: not only in industrial relations, but in the streets; not only there but in the villages or in the schools (Tarrow 1991a: 45-6).

Nonetheless, we can begin to examine how social movements have (cyclically?) clustered and been related among themselves. In so doing, of course, we also begin to examine and (re)establish the historical presence of these "other" social movements.

A convenient, but by no means exhaustive, listing of "typical streams of social movements which ... largely overlap in real movements" is that of Huber (1988:427): labor movement, women's movement, youth movement, old people's movement, movements for reform of life and critique of civilization, ecological and environmental movement, peasant and rural movement, preservation of home and culture including regionalism and localism, peace movement, expansion of conciousness and sensitive self experience, and spiritual- religious movements. This variety, Huber suggests, signifies that not the sociocultural intentionality, but that the participants are the determinants of these movements. However that may be, the evidence summarized in Tables 1 and 2 seems to confirm Tarrow's (1991a:49) observation that "cycles of protest also seem to arise across systems and economic sectors during the same historical periods." Indeed, they do so not only among different but perhaps overlapping movements in particular countries, but also among various countries in the West and for peasant movements throughout very many around the world.

Social movements undoubtedly have a millenarian and global history. For present purposes however, we confine their review to the past two centuries, for which we also have a better historical record. Nonetheless, even this record is very concentrated in a few Western countries, for which it is very country-specific. We will try, however, also to expand our review to other areas of the world, even especially by reviewing records of peasant movements around the world.

Following the compilations by our principal sources (Brand 1987, 1988 and Huber 1987) for the past two centuries, we may distinguish and classify "other" (non- class or national) social movements in core countries, principally the United States, United Kingdom, Germany and France, as those by women, for peace, for ecology/against industry, for community, and for changes in consciousness. For other areas of the world, we may draw on the Encyclopedia of World History by Langer (1948, 1972). We shall also draw on Huizer (1972) for Latin America, Huizer (1980) for South East Asia, Mukherjee (1988) for India, and Wolf (1969) additionally for Russia, China, and Algeria to review peasant movements in these areas. The compilation of these "other" social movements is summarized in Tables 1 and 2, which offer a comparative overview of the incidence or timing by decades and sometimes years of occurrence and general location of these movements and their correlation or lack of it with Kondratieff up and down phases.

The first (impressionistic?) observation is that there seems to have been significant bunching or clustering of social movements. Not only did particular kinds of social movements, eg. womens, peace, or ecological take place over roughly the same historical periods in different countries; but all of these and other social movements in various countries also appear clustered during the same historical periods. Moreover in Table 1, we can distinguish three major and a couple of minor periods since 1800 during which these social movements apparently became stronger and more numerous than in the intervening times. Whether this constitutes evidence for the existence of a cycle of social movements themselves is another question. The last column of Table 1 summarizes the peasant movements detailed in Table 2 and suggests that they too rose and fell in wavelike form around the world, but that the timing of peasant movements hardly coincides with that of other social movements, except during the early 20th century.

The first upsurge of social movements (since 1800 though not necessarily the first if we look farther back) clusters in the twenties, thirties and forties of the 19th century. In 1811- 16, the British Luddites resisted the negative consequences of industrialization through a sort of ecological movement. Mennonites and Quakers founded peace societies after the Napoleonic Wars. Community movements in the US and UK and consciousness movements, such as romanticism, in Europe already begin earlier in the century, but continue towards mid century, when they appeared as "Young" Germany, France, Italy, Ireland, and similar movements. Womens, peace and ecological movements, particularly in the US and the UK, and the last also in Germany, predominate in the 1830s and 1840s, though womens movements in the UK and Germany also continue into the 1850s and 1860s.

Significantly, there were substantial links among these and other social movements. Thus the American womens movement, culminating in the 1848 Seneca Falls Convention and Declaration, had links to the contemporary temperance, other moral reform, and anti-slavery movements. Similarly, both American and, after the decline of the 1830s Chartists, the British (Mary Wollenstonecraft) womens movement had links or overlapped with the Owenite and Fourier utopian socialist alternative communitarian movements. In Germany, the brief upsurge of a womens movement was related to the 1848 revolution. In all these countries however, the following decades appear marked by a notable absence of recorded social movements, except for the continuation of the anti-slavery movement in the United States and the rise of peasant movements in many other parts of the world (see below).

The next marked upsurge, again of all of these movements and now also including peasant movements, is during the last decade of the 19th century and the first one of the 20th century. The first but also the second decade of this century witness new womens movements, now everywhere demanding the suffrage, in the US, UK and Germany, and also in Latin America. In various countries World War I is preceded by peace, anarchist and bohemian alternative, as well as ecological and community movements, like the American "wilderness cult," (conservationism, the National Park System, Sierra Club and Audubon Society), the British "back to the country" and "garden city" movements and the German "Heimat," and "blood and land" as well as "civilizational" consciousness movements. The 1920s and part of the 1930s witness a lesser renewed upsurge of social movements in core countries, again accompanied by peasant movements elsewhere. The latter reappear in some areas after World War II and in the 1960s. The next major cluster of bunched "new" social movements appears in the mid 1960s and continues today. Brand 1988 argues that social movements decline again in the 1980s. However, the peace, womens and ecological movements increased in core countries at least through the mid 1980s (and Brand's table sill displays them in the early 1980s), and all kinds of social movements have certainly grown in the 1980s in the then "Socialist East" and the Third World South.

What sense can we make of all or even any of this? How can we relate the ups and downs of these social movements to each other, to other circumstances or cycles of economic growth, hegemony or colonialism, and of course to the "classical" class and national movements?

First of all, the fact that other investigators not only identify but also compile and classify "other" social movements in the past is further evidence that they are not "new" but are instead a (partially hidden) part of our history. Secondly, the very fact that these social movements seem and tend to coincide in time from one country to another and also as between different movements suggests that their upsurge(s) and abatement(s) is/are not coincidental. Apparently, they respond largely simultaneously to changing historical circumstances, which seem to occur at least part system-wide.

They may be economic. But their correlation with, let alone possible determination by, Kondratieff cycles is less than clear. The first major wave of social movements coincides largely with a Kondratieff downturn (or begins, as Huber would read it, near the Kondratieff top). So does the current wave of social movements, which began in the late 1960s. However, the intervening second wave of social movements coincided largely with the 1896-1913/20 "Belle Epoque" Kondratieff upturn, and with some exceptions they weakened during the economic crisis of the late 1920s and 1930s.

With regard to peasant (social) movements however, we may be on firmer ground in looking for or attributing common world systemic changes in political economic opportunity/necessity structures. It might seem curious to expect or find that "local" peasant movements in very different parts of the world should also share temporal clusters. And yet, although some peasant movements also appear at some other times, many important and well known ones also seem to have occurred in bunched waves. The late 1850s to the early 1870s witnessed not only the famous Tai Ping (1850-65) and lesser known Nien (1852-68) rebellions in southern and northern China respectively, but also the well known Indian Mutiny of 1857 "which was undoubtedly the most widespread peasant revolt of the nineteenth century" (Mukherjee 1988:2115) and the 1859 Blue Mutiny or Indigo Revolt in India. However, the 1860s and 1870s also saw important peasant movements in Mexico at the time of Benito Juarez, the Brazilian Northeast, Colombia, and associated with liberal reforms in response to export agriculture elsewhere in Latin and Central America and the 1868 war in Cuba (Frank 1972), as well as in Algeria in 1871- 72, and India again in 1875 and 1879.

The turn of the century witnessed a new wave of peasant movements in China, including the Boxer rebellion, India, leading up to the Revolution in Mexico, Bolivia, again war in Cuba in 1898, in Zimbabwe and the Boer War in South Africa, and in 1902 and 1905 in Russia. The 1920s and early 1930s saw important peasant movements in Japan (1921-26 following earlier ones in 1916-18), China (1921,1925,1930s Long March), Philippines (1923 and 1926, 1931-35/38), Vietnam (1929), India (1922-24 and 1928), Mexico, Bolivia, Brazilian Northeast, and throughout Central America and the Caribbean (Sandino in Nicaragua, repression with 30,000 dead in El Salvador, Cuba, etc.). The decade following World War II had the Telengana Rebellion (1946-51) and Tebhaga movement (1946-47) as well as the movements related to partition in India, the Huk revolt in the Philippines, the 1952 peasant movements and revolution in Bolivia, Dien Bien Puh in Vietnam in 1954, and the beginning in 1954 - after 80 years of relative quiet - of peasant and urban based liberation movement in Algeria. The 1960s witnessed further notable peasant movements in India (Naxalite), Philippines (NPA), Brazilian Northeast (Ligas Camponesas), and elsewhere in Latin America.

These waves of peasant movements do, however, seem to coincide much more with Kondratieff upturn times in the 1850s and 60s and reaching into the 1870s downturn; the early 1900s and again the 1920s and early 30s; and the 1960s with some forerunners after wartime booms. Most students (eg. Wolf 1968) of these movements have interpreted them as peasant reactions to commercialization of agriculture in response to growing (often foreign) market opportunities for large landowners. As the latter respond to these market opportunities, they displace their tenant and neighboring independent peasants from subsistence production on the land and thereby threaten their livelihood and security (as Frank 1967 also observed). Moreover, these peasant movements are therefore often also associated with anti-colonial liberation movements. Therefore, we should not be surprised at such temporal correlations of Third World peasant and liberation movements first with world economic Kondratieff upturns, which generate the conditions for them, and then with the even sharper pain of subsequent crashes, which in turn constrain and threaten commercial agriculture and landless agricultural laborers, as after 1873 and 1930.

There may be relations between the other social movements and hegemony or peasant movements and nationalist anti- colonialism. However, the fact that social movements coincide in time across countries with different and rising and falling hegemonical status also leaves their possible relations less than clear. On the other hand, the de facto relation and even alliance between some peasant and some national(ist) anti- colonial and also anti-imperialist movements may be easier to establish.

To relate the "other" social movements to the "classical" labor/class and national ones, we may also begin by looking at their respective timing. The timing of strike waves, measured by adding up all their available data, has recently been surveyed by Gattei (1989) for five core countries using Screpanti's and other data and by Silver (1989) counting (New York Times and London Times) newspaper mentions of strikes throughout the world. Both authors found marked upsurges and peaks of strikes in the late 1840s and around 1870 (but by inference from the historical record prior to the beginning of their data series) and in their own data after 1890, around 1920 (after World War I), the late 1940s (after World War II), and Gattei but not Silver for the late 1960s. Both authors try to relate their strike peaks to Kondratieffs, and Gattei remarks that his peaks coincide with both upper and lower Kondratieff turning points. His argument that they reflect increased turning point tensions is less convincing. We must consider that some strike peaks come after wars (although Goldstein 1987 argues that these wars in turn come at Kondratieff peaks). Also strikes occur mostly locally and sectorally (even if Gattei adds them up internationally). Yet his Kondratieff dating refers to the world or at least core economy and is not necessarily matched by all local, sectoral or national peaks and troughs, which might be reflected by strikes.

Nonetheless, we can see some temporal overlaps between their strike peaks and our "other" social movements, which we have plotted in a more rough and ready fashion by decades in our Table 1. Our first upsurge of "other" social movements, especially in the 1830s and 1840s, certainly coincides with the class (and also national) movements of these same decades, culminating in the revolutionary and reform movements of 1830-34 and 1847-52, centering on 1848. For the first ones for instance, Goldstone (1991: 285-6) mentions revolutions or rebellions in England, France, Belgium, Poland and Ireland. During the second period he lists "revolutions or serious revolutionary crises" in France, Germany, Austria, Hungary, Bohemia, Switzerland and Romania. Excepting in Prussia, "these nineteenth-century crises were full-fledged cases of state breakdown" in which "social protest was absolutely central," certainly in France, but apparently also elsewhere.

The renewed upsurge of social movements at the turn of the century also begins with, but continues after, the strike peak of the early 1890s. In our own century, social movements again coincided in the time, but less so in strength and extension, with the strikes after World War I, and again (if there was a strike peak) in and after the late 1960s. However, we did not find a marked upsurge of social movements other than peasant movements around 1870. Now, a century later, the labor movement is weakening (and nationalist movements are growing) during the present period of social movement upsurge. Thus, the evidence seems to support Tarrow when he suggests -- and refers to detailed but more nationally confined studies of his own on Italy and by Tilly on France --that labor and "other" social movements rise and decline together.

It would be desirable to make such a comparison with possible waves of national movements, but we lack a similar plotting for them. However, we can observe roughly that national movements also increased in the 1840s, around 1870, around and of course during the world war periods and again now. So national movements seems to have a rough "coincidence" with the strike and other social movements. Moreover, the peasant movements plotted in Table 2 probably contain components of both national and agricultural labor movements in the Third World with some relation to both Kondratieffs (as noted above) and other movements elsewhere.

On the other hand as mentioned above, the second author of this article now suggests some differences and distinctions among social movements and their behavior within the K and other economic cycles. Labor movements grow in Kondratieff A phases, when economic expansion strengthens the number and bargaining power of workers [This correlation is confirmed by the findings of Boll (1985), Screpanti (1987) and others, although for the latter the social movements seem to drive the economic cycle. Moscoso (1991 and nd) reviews and largely affirms these findings but seeks to qualify them somewhat]. Feminist [but not all womens] movements also grow during or after the A phase sustains more education and initially more employment of [primarily middle class] women, who then promote feminist demands, albeit to some extent also when employment opportunities commensurate with their education dry up. Peace movements respond to growing war clouds, which tend to come near the end of the A phase of K cycles (Goldstein 1989).

The last part of the A phase for peasants, and the impoverishment at the beginning of the B phase for them and for the urban poor, generates movements to defend economic survival, which include many women but are not particularly feminist. Indeed, B phases generate anti-feminist backlashes, especially during periods of unemployment, which are accompanied by ideological ploys about "saving the family" and sending women "back home, where they belong." [This was particularly flagrant under every fascist regime in the 1930s, and it is becoming so once again under the "post-Communist" regimes in Central and Eastern Europe today. However, various forms of anti-feminist backlash and even "movements" are also in evidence today in various countries of the industrial West]. The depth of economic crisis in turn generates nationalist, racist, religious and other redemptionist movements, which offer spiritual solace to the victims of this crisis, not incidentally also at the expense of both women and feminism. [This generates some anti-racist and feminist defense movements in response]. However, a recent survey was not able to establish any significant correlation between church attendance and short cyclical recessions in the United States (International Herald Tribune xx, 1992).

Thus, the evidence does seem to confirm Tarrow's (1991:49) above cited observation that "cycles of protest also seem to arise across systems and economic sectors during the same historical periods." Indeed, they do so even more than Tarrow probably expected, since he confines his focus to social movements in industrial countries. Perhaps, however, we should distinguish even more, using recent regional or sectoral experience as a guide.

Of late in the West, peace and womens movements have certainly abated, and the labor movement has been notably weakened. As we write, the peace movements mostly shine by their absence regarding the fighting in the former Jugoslavia and Soviet Union - not to mention Somalia and other parts of the Third World - as they also mostly did during the 1990-91 crisis and war in the Gulf. Womens and feminist movements, if anything, have become rather defensive against the abovementioned anti- feminist backlashes. The labor movement seems altogether defenseless in the new recession, which began in 1989. Environmental movements still survive more, although they seems not to mobilize people very much.

In the East, social protest movements blossomed at the end of the 1980s in response to growing economic crisis; and they were instrumental in promoting changes in political regime in various countries in 1989. However, then the economic crisis turned notably worse for the population with growing unemployment and inflation [due in part to the simultaneous recession in the West and the "marketization" and "privatization" in the East itself] -- yet the earlier "human rights" oriented social movements disappeared entirely and/or they were institutionalized in and by the new "democratic" party politics. The previous official peace "movements" disappeared with their regimes, of course. The non-official peace movements have hardly survived either, however, despite the growing threats and actuality of civil - and perhaps soon foreign - war. Some ecological movement has survived and grown in some regions but abated in others, as the people's concern for their economic, political and physical survival has become paramount. Instead, social protest is channeled into a variety of nationalist, ethnic, and racist "movements," which balkanize the Balkans and the former Soviet Union more than ever before and violate -- indeed deny - all civil and civic rights and often that to life itself to "the enemy." Under the circumstances, womens movements are hard to find.

In the Third World South, democratization also advanced in the late 1980s and early 1990s. However, so did the economic crisis; which continued to impoverish the people. Their defensive movements of protest and for survival have also continued unabated and in rural areas also take the form of ecological/environmental defense movement. The participation and leadership of women in these defensive movements continues or still increases. At the same time, there has been a marked growth of defensive and even offensive movements among indigenous minorities. Similar movements also grew on previous occasions at the same time as, or even in relation to, earlier peasant movements.

Apart from these "sectoral" movements however, the previously progressive political content or direction of social movements seems to be turning rightward in many countries. More liberal democratic institutions - or institutional democracy - is taking the wind out of some movements' political protest. In Latin America right wing evangelical fundamentalism is replacing more progressive community organization around the theology of liberation and other popular currents in the Catholic Church. In South Asia, right wing Hindu and Buddhist communalism and populism is capturing increasing popular allegiance. In the Muslim world right wing fundamentalism is on the rise. At the same time, the economic crisis continues and worsens and the liberal democratic and other regimes prove powerless and/or incompetent even at minimal crisis management. Thus, in several regions and many countries round the third world -- and now in the thirdworldized former "second" world as well -- military takeovers threaten soon to replace the democratic regimes and thereby also to alter the "political opportunity structure" for social protest movements again.

It may be too early to say what these variated sectoral and regional manifestations of social movement mean for this cycle, or how they fit into the historical pattern of social movements cycles.

In any case, we still have not accounted for these cycles in and of social movements, if indeed they are real cycles. Outstanding among the attempts to do so are, again, the writings of Sidney Tarrow (1983, 1991, nd.), who pursues our 4th question above and seeks to explain the cycles in and of protest movements themselves within their institutional contexts. As time and his work progresses, so does his ability to explain and persuade--but not yet quite satisfactorily. Moreover, his work concentrates rather exclusively on the industrial West.

Tarrow, and following him also Brand (1987), tries to account for the mobilization and especially the successes of social movements at some times and not at others on the basis of changing "political opportunity structure(s) [POS]." Tarrow (1983) analyzes and summarizes the latter in terms of changing openness and closure of social movements' political access to power, the stability or instability of political alignments within which social movements can operate, and their greater or lesser ability to find allies and mobilize support groups beyond themselves. Tarrow (1991) adds emphasis on a fourth "main component" of POS, political conflicts within and among elites, which also strengthen the other three. His emphasis is more on life cycles within movements [our first question] than cycles of movements [our second and third questions]. Although Tarrow concentrates on the industrial countries, similar POS, and especially the opportunity for popular protest movements to find allies among divided elites, have also been diagnosed by for instance the authors of Power and Popular Protest. Latin American Social Movements (Eckstein, Ed. 1989).

Following Mancour Olson (1965) it has been argued that at some times the socio-political cost/benefit ratio of action, or at least its perception, pulls some people off the fence and into movement. This kind of analysis also helps meet the objections to the "volcanic eruption" explanations by partisans of rational choice theory like Aya (1990), who argue that the analyst also has to account for when and why individual decision makers chose to participate in eruptions of social movements or to stay at home. Tarrow's "deconstruction" of why and when POS tilts the cost/ benefit ratio for individual decision makers and mobilizes people into social movements. He helps explain when, how and why movements of very small minorities grow into bigger minorities and some perhaps even into majorities, by which time they become institutionalized and cease to be movements. Thus, Tarrow also helps explain or render more plausible why different social movements or potential ones -- eg. in our Table 1 -- should experience the same increase of opportunities at the same time in the same society.

However, it is less explicative of why the life cycles in the resulting social movements should coincide all the way up and down again. Moreover, POS does not explain why there are recurrent cycles of social movements [our second and third questions]. It is not clear why the permissive if not causative POS underlying these movements itself increases and decreases in recurrent waves, not to mention in cycles that are self generating or at least have otherwise endogenously generated turning points. That was our fourth question, on which Tarrow concentrates his attention.

Extending this question internationally, POS fails even more to account for the simultaneity of movement growth, not to mention decline, among several different countries, which we also observe in Table 1. For Tarrow traces these four POS components most essentially back to the ebb and flow of a particular country's political institutions, within which he sees the generation of both the need and opportunity for protest movement. Tarrow and Brand hardly consider, and even less answer, why and how this institutional process, and therefore the SM cycle, might be the same from one country to another. The common participation of different countries in a common Kondratieff cycle and its influence on political institutions and policy across political boundaries could be one such explanation in asnwer to our fifth question. However, Tarrow and Brand reject that, and our evidence disconfirms it at least for the turn of the century A phase. That is, unless we can accept Huber's argument that social movements increase in all Kondratieff A phases, albeit in some with some delay, and in no B phases, which our data - in part derived from him - also disconfirm.

What else then [returning to our fifth question], might account for simultaneous social movements internationally? Jack Goldstone (1991) offered an explanation for Rebellions and Revolutions in the Early Modern World. Goldstone examines state breakdowns and associated social movements, which were simultaneously bunched at various periods in the 17th and 18th centuries in various countries of Europe, West Asia and East Asia. He also makes some comparative excursions into 19th century crises in Qing China, Tokugawa/Mejii Restoration Japan, and again the Ottoman Empire. "Any claim that such trends were produced solely by unique local conditions is thoroughly undermined by the evidence" (p.462). His conclusion is "that the periodic state breakdowns in Europe, China and the Middle East from 1500 to 1800 were the result of a single basic process.... The main trend was that population growth, in the context of relatively inflexible economic and social structures, led to changes in prices, shifts in resources, and increasing social demands with which the agrarian-bureaucratic states could not successfully cope" (p. 459). Population growth was "exogenously" determined by rising and falling death [not birth] rates; and it impinged on state finances, and generated greater inter-elite conflicts and social protest movements, "thus producing worldwide waves of state breakdown." In contrast, when population did not grow world wide, this process did not occur. Goldstone (187 ff) notes that, The previously mentioned social movements, rebellions and state breakdowns of the early 1830s and late 1840s occurred predominantly in the more "traditional" regions of least industrial growth, where population growth had impinged on the carrying capacity of the land. This socio-political unrest occurred less in regions of greater industrial growth, which offered more possibilities to absorb population growth. Per contra the Marxist thesis that stresses industrial capitalist generated inter-class struggle, this regional pattern of social movements and intra-class inter-elite struggle conforms more to demographic/structural crises.

Indeed, Goldstone also demonstrates that in each of the earlier cases he analyzes, the important conflicts and struggles were among the existing and emerging elites, and not between the "people" and them. "Factional conflict within the elites, over access to office, patronage, and state policy, rather than conflict across classes, led to state paralysis and state breakdown" (p. 461). Grass roots social movements from below were supplementary in that they helped further destabilize an already unstable state, if only by obliging it to spend already scarce resources to defend itself; and that the popular movements favored the interests of some elite factions against others. "I know of no popular rebellion that succeeded by itself without associated elite revolts or elite leadership in creating institutional change" (p. 11).

Goldstone's discussion of social movements is also welcome for other reasons: He shows [1] that they come and go in cycles of their own, and he relates them to wider systemic/structural cycles; [2] that they display much variety and changeability, but they share individual mobilization through a sense of morality and [in]justice and for survival and identity; and [3] that none of this is new. By implication neither are our contemporary "new" social movements. These observations correspond in reverse order to the first three of the "Ten Theses on Social Movements" of Fuentes and Frank (1989) and Frank and Fuentes (1990). Thus, we also welcome Goldstone's guidance for the study of bottom up social movements, which have always been important but often neglected actors in history, even if - or perhaps because - they often do not lead to state breakdown.

However, Goldstone explicitly exempts more recent times from this demographic/structural process and explanation. So, even if Goldstone's analysis of the international process is correct for the early modern world -- and perhaps for the medieval and ancient world (Frank 1992, Frank and Gills 1992), we are still and again left without an explanation of social movement cycles in recent times.

Why then is it that in different countries and apparently circumstances these structures of political opportunity increase(d) almost simultaneously in the second quarter of the 19th century, at the turn of the century, and apparently again in our own time, and why they decrease(d) in between? Indeed also, as Tarrow (nd:52) asks, "but why does the cycle end? We know much more [but still not enough (AGF)] about the factors that lead to social movement mobilization than those that produce their demobilization."

So the main questions still remains without an answer, on economic and/or political opportunity grounds, or otherwise. Tarrow only suggests as "a plausible hypothesis" that the external opportunity structure becomes more important for movement success towards the peak of the cycle. Here, however, he refers to the (peak of the) movement cycle itself, and not the (external) economic or political cycle, even though a couple of pages later he quotes Goldstone on the influence of economic and political crisis on movement success. Finally, Tarrow also observes that a favorable political opportunity structure is not sufficient for movement success, which is notoriously difficult and controversial to define to begin with (Tarrow 1991 chapter VI, Gamson 1975, Goldstone 1980).

This historical review of social movements leaves unresolved - -indeed unconsidered -- the question of whether their more or less synchronized ups and downs constitute or are the result of a social movement cycle in and of itself [our third and fourth questions]. It has been argued that there are independent cycles of ideology (Sorokin, Sarkar), American politics (Schlesinger Senior and Junior), and other aspects of social life. Brand (1988 and personal correspondence) argues that social movements reflect "discontinuous social change" in response to "cultural crises when the cultural paradigm is eroding," which is specific to and differs from one socio- cultural-political unit to another. However, these "cycles'" supposed generational and other mechanisms of phase changes, recurrence, and self-perpetuation are far from satisfying the criterion of sine wave like autogeneration of a true cycle. Moreover, while these supposed ideational cycles may overlap here and there or now and then with waves of social protest movements, it would be hard to demonstrate their identity over history. Thus it would be hard to demonstrate that the ups and downs of social movements coincide with, much less have their source in, an underlying ideational cycle.

On the other hand, Andrew Jamieson argues that

Social movements have been the source of many important social innovations in the development of science and technology, new ways to organize both the production, as well as the dissemination of knowledge. Even more important perhaps, social movements have altered the boundaries of the officially sanctioned institutions for knowledge production. By bringing new concerns into the arena of public debate, social movements have provided much of the basis for re-organization of the social institutions of knowledge production....Could they perhaps even be a crucial ingredient in the eruption of Thomas Kuhn's famous - or infamous - "scientific revolutions"?....Social movements can be said to have a cosmological function, acting as "social carriers" for new world-views or conceptions of man and nature (Jamieson 1988: 72,74).

Thus, Jamieson also examines some of the abovementioned bunched conceptual and ideological developments, such as utopian socialism in the second quarter and environmentalism in the last quarter of the 19th century, as manifestations of their apparently cyclically arising social (movement) carriers. As to the possible existence of some independent cyclical mechanism of auto generating phase change among social movements themselves, we are not aware of any serious attempt to demonstrate any and certainly cannot attempt any here.

* There are two tables that go with this manuscript which, as of 9/92, are not available electronically. Gunder Frank hopes to be able to provide snail mail hard copies to persons who would like the tables. Please send requests to agfrank@neu.edu (editor)