Secular Cycles

From P2P Foundation
Jump to navigation Jump to search

* Book: Secular Cycles. By Peter Turchin and Sergey A. Nefedov. Princeton University Press, 2009



1. Mark Koyama:

"The leading modern day cyclical theorist is undoubtedly Peter Turchin. For my money Turchin’s best book is Secular Cycles (co-authored with Sergey A. Nefedov). Their innovation (building on an argument made by my GMU colleague Jack Goldstone in his 1991 book Revolution and Rebellion in the Early Modern World) is to take the Malthusian model of economic cycles and add to it a model of elite competition.

Tuchin and Nefedov show that periods of demographic expansion are often associated with the growth of elite incomes and inequality (as population growth causes rents to rise and wages to fall). More elites competing over the surplus, however, puts fiscal pressure on the surplus-extraction machine that we call the state. Elite overproduction thus brings about a political crisis. Secular Cycles applied this model to medieval and early modern England and France, Russia and ancient Rome. Turchin’s most recent book applies it to the United States." (

2. From the publisher:

"Many historical processes exhibit recurrent patterns of change. Century-long periods of population expansion come before long periods of stagnation and decline; the dynamics of prices mirror population oscillations; and states go through strong expansionist phases followed by periods of state failure, endemic sociopolitical instability, and territorial loss. Peter Turchin and Sergey Nefedov explore the dynamics and causal connections between such demographic, economic, and political variables in agrarian societies and offer detailed explanations for these long-term oscillations--what the authors call secular cycles.

Secular Cycles elaborates and expands upon the demographic-structural theory first advanced by Jack Goldstone, which provides an explanation of long-term oscillations. This book tests that theory's specific and quantitative predictions by tracing the dynamics of population numbers, prices and real wages, elite numbers and incomes, state finances, and sociopolitical instability. Turchin and Nefedov study societies in England, France, and Russia during the medieval and early modern periods, and look back at the Roman Republic and Empire. Incorporating theoretical and quantitative history, the authors examine a specific model of historical change and, more generally, investigate the utility of the dynamical systems approach in historical applications.

An indispensable and groundbreaking resource for a wide variety of social scientists, Secular Cycles will interest practitioners of economic history, historical sociology, complexity studies, and demography."


1. Donald J. Zeigler:

"The shadows of Thomas Malthus and David Ricardo loom large over Turchin and Nefedov's search for history's motive force. Neo-Malthusian demographers will feel vindicated by the theory of secular cycles. Classical economists will relish eight case studies demonsti-ating Ricardo's law of diminishing returns. Marxists, however, will avert their gaze from a theory that is predicated on a symbiotic relationship between elites and peasants, and a competitive relationship between elites and the state. Secular cycles seem to explain more variation over time in power relationships than does class struggle. Geographers and ecologists will recognize the important role of the biophysical environment in influencing the course of history. Malthus predicted catastrophe when the size of a population reached the carrying capacity of that population's territory. The theory of secular cycles incorporates Malthus's "positive checks" on population grouch by identifying them as ti-iggers of the disintegrative phase. When carrying capacity is exceeded, the stagflation phase of a secular cycle begins. Then, when a famine or epidemic stiikes, the disintegrative phase begins. This pattem is demonsti-ated consistently in England, France, Russia, and ancient Rome. To identify secular cycles in their regional contexts, Turchin and Nefedov mine deeply the annals of history. Out of their thorough literature review, they have pulled the quantitative data needed to test their theory. They have presented it, whenever possible, in the form of tables and graphs, of which there are well over a hundred. As one might expect, the availability of data is greatest for England and least for ancient Rome, yet in each case study the authors present a persuasive argument that the model of secular cycles holds for agrarian societies over time." (

2. Reviewed for EH.NET by Harry Kitsikopoulos, Department of Economics, New York University.

"This book is an audacious and ambitious attempt to promote the viewpoint that historical progression runs according to certain regular patterns. In its effort to prove this hypothesis it lays out a number of predictions testing them against empirical time series for four countries during different epochs: Rome (350BC-285AD), France (1150-1660), England (1150-1730) and Russia (1460-1922). The argument is constructed on the basis of measuring (sometimes speculating) four fundamental variables: actual population figures contrasted against ?ideal? population levels, i.e., upper limits defined by agricultural productivity; social structure measured by numbers and consumption levels of elites as well as annual budgets of typical peasant households; the power of the state measured by its fiscal health; and socio-political instability reflected in relevant events (e.g., uprisings, rebellions, civil wars) or, used as a proxy, in coin hoards.

According to Turchin (University of Connecticut) and Nefedov (Institute of History and Archaeology of the Russian Academy of Sciences, Ural Branch) each secular cycle lasts for several centuries and unfolds through the following phases: 1) An expansion phase characterized by relatively stable prices and modest wage declines (if any). 2) As population density tends to approach the limits imposed by the productive capacity of agriculture, we enter a stagnation or stagflation phase. A typical Malthusian scenario develops with an increase in the price of land and its products and a cheapening of labor. The ranks of the elite grow both due to biological reproduction and upward mobility and, simultaneously, its members get accustomed to higher levels of consumption. Gradually, however, we encounter a state of elite overproduction, so to speak, which leads to a relative (but not absolute, as in the case of commoners) decline of their living standards. Intensified oppression of peasants ensues as well as increased competition among members of the elite and between the elite and the state whose growth rate of revenues slows down. 3) A general crisis unfolds either abruptly or gradually which, in theory, can be addressed by raising productivity through technological innovations but more often than not it leads to military expansion into new territories and is resolved through the visitation of pandemics, extreme episodes of famine, or state collapse followed by intense civil war (or a combination of such events). The crisis lasts for a prolonged period. 4) It is followed by a depression phase during which resources per capita increase but fail to lead to population recovery due to the continuation of civil wars. This phase can be prolonged, particularly if the state continues to be dysfunctional, or it can lead to the beginning of a new cycle if the ranks of the elite are sufficiently pruned.

Secular cycles can acquire a periodic character but in societies with complex characteristics the dynamic may incorporate elements of sensitive dependence and non-linear feedback loops which lead to outcomes envisioned by chaos theory. The latter scenario is particularly plausible in the event of exogenous disturbances relating to geopolitical factors and the ecological environment as well as through the reaction of individual actors which can be portrayed ?as a stochastic process, a kind of Brownian motion that also results in erratic, unpredictable changes in the macrosocial trajectory? (p. 22).

The book relies on data drawn from the secondary literature but the authors? handling is not always as refined as it could be nor are they always accurate. For instance, in reconstructing the annual budgets of peasant households in pre-plague England, the authors neglect to include spending on the purchase of consumption goods not produced by the household or investment goods; the typical rent per acre was less than 1s and seigneurial dues at over half of the value of annual output exaggerate typical peasant obligations; gross yields of wheat were not at 10 and 8 bushels per acre before and after the Black Death but 10.5 and 11.9 bushels respectively (based on extensive demesnial records analyzed by Campbell), and when all grains are taken into account yields remained stationary between the two periods. In referring to seigneurial dues of French peasants during the same period, the authors take into account only the terrage and the tithe in reconstructing peasant budgets. But additional dues included the cens (a fixed cash payment), the taille (demanded both by lords and the state), and various payments stemming from the judicial and administrative authority of lords, conventionally known as seigneurie banale. Some of these payments may have been of nominal value, others irregularly imposed, but they amounted collectively to a significant draining of peasant resources; the authors are aware of their existence, since they mention them in other contexts, but for whatever reason they do not take them into account when reconstructing peasant budgets. Including such information is important in defining the acreage necessary to ensure subsistence and estimate the proportion of the population which fell below such thresholds.

Such issues may be deemed minor quibbles given the impressive breadth of evidence considered by the authors. But I found a little more problematic the lack of reference to publications presenting viewpoints inconsistent with the typology outlined in this book. Campbell (English Seigniorial Agriculture, 2000), for instance, argues that grain output per capita remained stable during the thirteenth century (corresponding to the stagflation phase in Turchin?s and Nefedov?s typology), hence challenging the notion of a growing immiseration of the peasantry." (


James Quilligan:

"The first chapter of Secular Cycles describes the general historical cycle of a civilization lifting itself out subsistence into greater social complexity and prosperity, then undergoing bitter class inequality, completely misunderstanding the need to manage economics and society by measuring resources according to population, and then undergoing a massive crisis when it is no longer able to support its complex support systems. In this cycle, society ascends through a long period where there appears to be endless resources in which population, growth and demand will increase forever; then descends through a long era when human demand crosses the sustainability threshold of its ecosystems, and energy and resource depletion drastically limit population growth. After the first chapter, the rest of the book provides historical case studies of civilizations tracing this rise and fall."

The book made me more aware of several things. No major civilization has EVER practiced carrying capacity as a basis for political and economic self-governance; carrying capacity has only succeeded in small communities. Of course, we know this from the modern Ostrom view of the commons; but Ostrom never put her finger on the pulse of carrying capacity as the *self-organizing principle between a species and its environment*. Nor has the commons movement recognized the importance of an *empirical way of measuring the metabolism of society* through the cooperative activities of people using resources to meet their biological needs. In other words, Ostrom and the commons movement have yet to define the dynamic equilibrium which they seek as the balance between two opposing forces - population and resources - which continually counteract each other. Instead, the commons movement is more focused on counteracting the Market and the State than on measuring the replenishment of renewable and non-renewable resources and managing them to sustain their yield. In short, the commons movement does not seem to be producing alternative indicators for the productive and provisioning which can be used to guide policy.

Secular Cycles made me realize that the commons, as Ostrom viewed it and as others are now envisioning it, is too informal and small-scale to work in a way that establishes empirical targets that will bring down exponential growth to arithmetic growth levels; and thus organizing society according to the dynamic equilibrium between population and the availability of food, water and energy. Instead, what we get in the commons movement is a general opposition to quantitative analysis because it reminds people too much of the metrics of unbridled capitalism. My point is that if we don't know how to develop evidence-based policy for a soft landing toward a reasonable level of subsistence -- and I've seen very little of this in the commons movement -- then I don't know how we expect to create a long-term system for meeting human needs through sustainable yields. I would hope that the commons movement begins to create the basis for a viable new society by actually focusing on the optimum rate at which a resource can be harvested or used without damaging its ability to replenish itself. That would be something." (


From: Chapter 1, Introduction: The Theoretical Background

From the reading highlights by Michel Bauwens.

1.1 Development of Ideas about Demographic Cycles

Peter Turchin et al.:

"The modern science of population dynamics begins with the publication in 1798 of An Essay on the Principleof Population by Thomas Robert Malthus. Malthus pointed out that when population increases beyond the means of subsistence, food prices increase, real wages decline, and per capita consumption, especially among the poorer strata, drops. Economic distress, often accompanied by famine, plague, and war, leads to lower reproduction and higher mortality rates, resulting in a slower population growth (or even decline) that, in turn, allows the subsistence means to “catch up.” The restraints on reproduction are loosened and population growth resumes, leading eventually to another subsistence crisis. Thus, the conflict between the population’s natural tendency to increase and the limitations imposed by the availability of food results in the tendency of population numbers to oscillate.

By the 1930s the empirical material had accumulated to the point where it became very clear that European prices had gone through a number of very slow swings between 1200 and 1900 (Simiand 1932, Griziotti-Kretschmann 1935, Abel 1980).

The most striking pattern to emerge was the wavelike movement of grain prices (expressed in terms of grams of silver). There were three waves or “secular trends” (Abel 1980:1):

1. An upward movement during the thirteenth century and early fourteenth century, followed by a decline in the late Middle Ages

2. Another upsurge in the sixteenth century, followed by a decline or apparent equilibrium (depending on the country) during the seventeenth century

3. A third increase during the eighteenth century, followed by irregular fluctuations during the nineteenth century, eventually converging to an early twentieth-century minimum.

By contrast, population moved more or less in the same direction as the food prices and in an inverse ratio to wages (Abel 1980:292–93). Abel concluded that the Malthusian-Ricardian theory provided a better explanation of the data than the monetarist theory.

Furthermore, the Malthusian-Ricardian theory predicted that an increasing population would result in a specific progression of effects. Rents would rise first, with grain prices lagging behind rents, the price of industrial goods lagging behind grain prices, and workers’ wages bringing up the rear. The evidence showed that this was precisely what happened (until the whole system was dramatically changed in the nineteenth century).

Speaking in 1973, he (LeRoy Ladurie) said, “it is in the economy, in social relations and, even more fundamentally, in biological facts, rather than in the class struggle, that we must seek the motive force of history” (quoted in Hilton 1985:4).

Such a radical Malthusian position could not but provoke a reaction from scholars working within the Marxist tradition.

In an influential book first published in 1946, Maurice Dobb argued that the cause of the crisis was the inefficiency of feudalism as a system of production, coupled with the growing needs of the ruling class for revenue (Dobb 1963:42–47).

One interesting contribution to the theory was Paul Sweezy’s proposition that the growing extravagance of the feudal ruling class was a result of the rapid expansion of trade from the eleventh century onward, which brought an ever-increasing variety of goods within its reach (Sweezy et al. 1976:38–39).

Brenner did not deny that the Malthusian model had a certain compelling logic (Brenner 1985a:14). However, its attempt to explain long-term trends in economic growth and income distribution was doomed from the start because it ignored (“abstracted away”) the social structure, the most important part of which was the surplus-extraction relationship between the direct producers and the ruling class (Brenner 1985a:10–11).

One deficiency of the Malthusian theory, according to Brenner, was the empirical observation that different societies within Europe, starting from similar demographic and economic conditions obtaining after the Black Death, subsequently followed divergent trajectories. For example, serfdom completely disappeared from certain Western European countries (England, France) while making a strong comeback in Central Europe (Poland, Prussia).

On the other hand, the extreme version of the Marxist thesis (perhaps found in the purest form in Sweezy), which assigns class relations the alldetermining role in the economic development of medieval and early modern Europe, would also fail to account for empirical facts.

What we need is a synthetic theory that encompasses both demographic mechanisms (with the associated economic consequences) and power relations (surplusextraction mechanisms). In the dynamical systems framework, it does not make sense to speak of one or the other as “the primary factor.” The two factors interact dynamically, each affecting and being affected by the other.

There is, however, a significant movement among historical sociologists “to bring the state back in” (Skocpol 1979). States are not simply created and manipulated by dominant classes; they are agents in their own right, and they compete with the elites in appropriating resources from the economy.

Historians have long recognized that there were recurrent waves of state breakdown and political crises in European history: the “calamitous” fourteenth century (Tuchman 1978), the “iron century” of 1550–1660 (Kamen 1971), and the “age of revolutions” of 1789–1849 (Hobsbawm 1962). Each of these periods was preceded by a period of sustained and substantial population growth. In a pathbreaking book, Jack Goldstone (1991) argued that there is a causal connection between population growth and state breakdown.

For this reason, Goldstone refers to his theory as demographic-structural: demographic because the underlying driving force is population growth, structural because it is not the demographic trend itself that directly causes the state crisis but its impact on economic, political, and social structures (Goldstone 1991:xxvi). We discuss this theory in more detail in the next section, but here we should mention that the verdict on Goldstone’s work among historical sociologists has been highly positive (see, e.g., Collins 1993, Wickham-Crowley 1997, Li 2002).

To summarize, it is becoming increasingly clear to specialists from very diverse fields—demographers and historical economists, social historians, and political scientists—that European societies were subjected to recurrent long-term oscillations during the second millennium CE (Braudel 1988, Cameron 1989, Fischer 1996). Furthermore, the concept of oscillations in economic, social, and political dynamics was not discovered by the Europeans. Plato, Aristotle, and Han Fei-Tzu connected overpopulation to land scarcity, insufficient food supply, poverty, starvation, and peasant rebellions (Parsons 2005).

In this book we examine the hypothesis that secular cycles—demographicsocial-political oscillations of very long period (centuries long)—are the rule rather than the exception in large agrarian states and empires." (

1.2 A Synthetic Theory of Secular Cycles

"It is clear that neither purely demographic nor purely class conflict explanations of secular cycles work very well when confronted with data. On the other hand, a synthetic theory that incorporates both of these (and some other) processes may provide us with a viable hypothesis that can be tested with data. The idea is that secular cycles can only be understood as a result of the interaction between several interlinked variables—economic (including demography), social structure (particularly, how the elites interact with the producing population and the state), and political (state stability or collapse).

The key variable is the population density in relation to the carrying capacity of the local region.

Carrying capacity is defined as the population density that the resources of the habitat can support in the long term (for an excellent discussion of human carrying capacity from an ecologist’s point of view, see Cohen 1995).

Carrying capacity thus is an upper ceiling on population growth. From the point of view of economics, this limit arises because labor inputs into production suffer from diminishing marginal returns. It is clear that the carrying capacity of a specific region is strongly affected by its physiographic features (the availability of land suitable for agriculture, water supply, soil characteristics, length of the growing season, and so on). It is also affected by year-to-year fluctuations in the temperature and the amount of rainfall, as well as by gradual changes in the climate. In other words, carrying capacity is a variable that changes in both space and time. Finally, and most important, carrying capacity is affected both by the existing level of agricultural technology and by how this technology is employed. As Ester Boserup (1966, 1981) famously argued, population growth can have a positive effect on economic innovation.

Thus, adverse effects of population growth on the standard of living can provide strong inducements for the adoption of new means of production.

A society that approaches the current limits of population growth can invest in clearing forests, draining swamps, irrigation, and flood control. All these measures will result in an increase in the carrying capacity. However, at some point there are no more forests to cut or swamps to drain.

As population density approaches the carrying capacity, a number of related changes affect the society. There are shortages of land and food, and an oversupply of labor.

Economic distress leads to lower reproduction and higher mortality rates, resulting in a slower population growth.

Other factors not taken into account by a purely demographic model would preclude the emergence of a stable equilibrium.

Population growth in excess of the productivity gains of the land has a fundamental effect on society’s structures. The typical changes accompanying population growth are high rents and land prices, increasing fragmentation of peasant holdings or high numbers of landless peasants.

However, as long as the elites are united and the state maintains control of the military, such popular uprisings have small chance of success.

This fundamental point was recently reiterated by Jack Goldstone:

- It is a profound and repeated finding that the mere facts of poverty and inequality or even increases in these conditions, do not lead to political or ethnic violence (Gurr 1980, Goldstone 1998, 2002b). In order for popular discontent or distress to create large-scale conflicts, there must be some elite leadership to mobilize popular groups and to create linkages between them. There must also be some vulnerability of the state in the form of internal divisions and economic or political reverses. Otherwise, popular discontent is unvoiced, and popular opposition is simply suppressed.

Surplus is the difference between the total production and what is needed for subsistence.

The amount of resources needed for subsistence increases linearly with population, while the total product grows slower than linearly as a result of the law of diminishing returns (figure 1.1a). As a result, at a certain critical population density, which we have defined as the carrying capacity, the two curves intersect.

Thus, when population increases from a low level, initially the amount of surplus increases (more peasants means more surplus). At some intermediate density, however, the surplus reaches a maximum: this is where the effects of diminishing returns on labor inputs into agriculture begin to be felt.

One important dynamic is that the elites are usually able to extract a larger amount of surplus during the late stages of population growth.

During the late stages of population growth, when commoners are already suffering from economic difficulties, the elites are enjoying a golden age. Both the reproduction of the existing elites and the recruitment of new elites from commoners will be fastest when the amount of extractable surplus is greatest.

The peak of elite numbers often lags behind that of the general population.

Such a happy state of events (for the elites) cannot continue for long. First, expansion of elite numbers means that the amount of resources per elite capita begins to decline. This process would occur even if the total amount of surplus stayed constant. But, second, as general population grows closer to the carrying capacity, surplus production gradually declines.

Modern studies of consumption level expectations suggest that people generally aim at matching (and if possible exceeding) the consumption levels of their parents (Easterlin 1980, 1996). Thus, what is important is not the absolute level of consumption but the level in relation to the previous generation.

Should their level of consumption decrease in relation to the pre­vious generation’s, the elites would be expected to react vehemently to this development.

The deteriorating economic conditions of the elites during the late stagflation phase of the secular cycle do not affect all aristocrats equally.

Poor aristocratic lineages tend to get poorer because they attempt to maintain their elite status on an inadequate economic basis.

A wealthier lineage, by contrast, can maintain the level of consumption necessary for preserving its elite status and have some resources left over to acquire land from its impoverished neighbors. As a result, the poor get poorer while the rich get richer. The same dynamic operates on peasants during the stagflation phase. During periods of economic hardship, poor peasants must sell land or starve. As a result, at the same time that the majority are sliding into absolute misery, a small percentage of thrifty, hardworking, or simply lucky peasants are able to concentrate increasing amounts of land in their hands.

During the stagflation phase, thus, economic inequality increases within each social stratum—peasants, minor and middle-rank nobility, and the magnates. Growing inequality creates pressure for social mobility, both upward and downward. Increased social mobility generates friction and destabilizes society. The growing gap between the poor and rich also creates a breeding ground for mass movements espousing radical ideologies of social justice and economic redistribution.

The declining incomes of the majority of aristocrats have two important consequences: intensifying oppression of the peasants by the elites and increasing intraelite competition for scarce resources.

If successful, elites may not only deprive the commoners of the surplus but may also cut into the subsistence resource, resulting in a negative growth rate for the commoner population.

It appears that this stage in the secular cycle may be what is known among dynamicists as a “bifurcation point,” a point at which the system may follow one of several alternative trajectories. A classic example of such divergent trajectories is the disappearance of serfdom in post-medieval England and France and, during the same period, the rise of new serfdom in Prussia and Poland.

This thesis is illustrated by the recent study of Stuart Borsch (2005), which compared the effects of the Black Death in England and Egypt. In post–Black Death England wages rose, rents and grain prices dropped, unemployment decreased, and per capita incomes grew.

The consequences of depopulation in Egypt were profoundly different. Wages dropped, land rents and grain prices rose, and unemployment levels increased. No economic recovery was anywhere in sight by 1500.

Borsch argues convincingly that the persistent stagnation of post–Black Death Egypt is explained by structural factors.

English peasants could resist elites by hiding in the hills and forests, of which there was an abundance in a depopulated England.

By contrast, Egypt’s narrow strip of arable land between uninhabitable desert left no room for evasive tactics. After the Black Death, Mamluks were able to use their tremendous coercive power to maintain the preplague level of resource extraction from a greatly diminished rural population. Extremely high levels of exploitation of individual peasants precluded any demographic revival. The system, thus, was caught in a “vicious equilibrium” .

The second consequence of plunging elite incomes is increased intraelite competition. The forms that this competition takes will depend (again) on the structural characteristics of the society. Probably the most important factor is the capability of the state to suppress overt violence. Here we consider the forms of intraelite competition in the presence of the state when internal order is maintained. The situation after the state collapses or is seriously weakened is considered later.

One recourse for elites facing declining incomes from agriculture was to seek employment with the state or church bureaucracy.

Impoverished elites could also improve their incomes by attaching themselves to the retinues of powerful magnates.

Limits on available land, civil and ecclesiastical offices, and royal patronage lead to increasingly polarized factional battles between patron-client groups for available spoils.

Because there are not enough resources for everybody, certain segments of elites, or groups aspiring to elite status, inevitably end up as the losers. We refer to them as the counterelites, or dissident elites. Usually, the counterelites do not constitute a true sociological group.

Social trends resulting from demographic growth—declining surplus production, popular immiseration, and intraelite competition—have a profound impact on the ability of the state to maintain internal order.

After a certain lag time, the negative effects of population expansion begin to affect the elites, who become riven by increasing rivalry and factionalism. Another consequence of rapid population growth is the expansion of youth cohorts. This segment of the population is particularly impacted by lack of employment opportunities.

Finally, growing economic inequality, elite competition, and popular discontent fuel ideological conflicts. For example, in early modern Europe, dissident elites and dissatisfied artisans were widely recruited into heterodox religious movements. As all these trends intensify, the end result is state bankruptcy and consequent loss of the military control.

On the ideological level, the feeling of social pessimism is pervasive and the legitimacy of the state authority is at its lowest point. The society approaches a condition that may appropriately be called “Hobbesian”.

In the previous sections we focused on the manifold effects of population growth on various structures of the society, including a bundle of variables that we call sociopolitical instability. Here we consider the feedback effect: how does instability affect population dynamics? We can envision two general (and, actually, interrelated) ways: by affecting demographic rates and by affecting the productive ability of the society.

International trade expands in the precrisis period (stagflation phase) and then gradually declines after the society has descended into anarchy. Thus, the rise of widespread epidemics—pandemics—is most probable during the late stagflation phase. In fact, the arrival of a pandemic is one of the most frequent triggers of the demographicstructural collapse.

Finally, political instability causes lower reproduction rates, because personal consumption plummets as a result of lowered production capacity.

In addition, during times of uncertainty people choose to marry later and to have fewer children.

The second and perhaps even more important effect of sociopolitical instability is on the productive capacity of the society (the carrying capacity). Vigorous states often invest in increasing the agricultural productivity by constructing irrigation canals and roads, implementing flood control measures, clearing land from forests, organizing the colonization of underpopulated regions, and so on. The end result of these measures is mainly an increase in cultivated area, although some measures also increase the productivity of land.

The other general mechanism is that the state offers protection. In a stateless society, people can live only in natural strongholds or in places that can be made defensible, such as walled cities.

After the collapse of Rome, settlements were moved to hilltops.

In other words, lack of effective suppression of internal violence by the state imposes a “landscape of fear,” in which a large proportion of agriculturally suitable lands is abandoned because they are too far from a place of security. By contrast, the strong state protects the productive population from external and internal (banditry, civil war) threats, and thus allows the whole cultivable area to be put into production.

Sociopolitical instability affects elite numbers in a fashion that is similar to its effect on commoners, although the relative importance of specific mechanisms can be quite different. Thus, the elites may be little affected by subsistence crises. They also tend to escape more lightly the effect of epidemics.

On the other hand, by virtue of their more active participation in politics, the elites ran a much higher risk of violent death. The death toll in some conflicts was extraordinary. For example, Dupaˆquier et al. (1988a:342) quote an estimate by Philippe Contamine that around 40 percent of the French elite may have been slaughtered in the Battle of Poitiers (1356), and the same proportion at Agincourt (1415). During the Wars of Religion in the late sixteenth century, 20,000 Huguenots were killed in just one day, the St. Bartholomew’s Day massacre (Kamen 1971:39).

Loss of life or elite status could also result from state purges. For example, the first Ming emperor purged 100,000 Chinese officials (Tignor et al.

2002:62). Sulla’s proscriptions eliminated a third of the Roman ruling class, senators, and another third was eliminated by proscriptions following Caesar’s death (see chapter 6).

A much less spectacular but perhaps ultimately more important process reducing the elite numbers is downward mobility. The plunge in elite incomes, which begins in the precrisis period and is greatly exacerbated by the general population decline, affects most strongly the status of the lowest noble stratum. A specific example is given by Christopher Dyer for late medieval England.

In summary, a number of social mechanisms exist by which elite surpluses can be reduced:

(1) deaths resulting from civil war,

(2) deliberate purges of elites by new rulers,

(3) limitations imposed on heir production (celibacy, primogeniture),

(4) downward social mobility, voluntary or forced by the state,

(5) increased material resources resulting from conquest or improvements in agricultural productivity, and

(6) the development of a new political order that directs a greater share of resources to the elites. Several such mechanisms are usually operating in combination; the specific mix depends on cultural peculiarities of societies and historical accidents.

Because the three main factors driving the rise of sociopolitical instability are general overpopulation, elite overproduction, and state insolvency, all these trends must be reversed before the disintegrative phase can end. S\

The last stages of the secular cycle are particularly rife with bifurcation points.

The problem of overpopulation is usually “dealt with” during the crisis phase. One of the most common proximate mechanisms of population collapse is disease, but not all population declines are accomplished by catastrophic epidemics. Prolonged periods of civil war can also cause drastic drops in population levels.

An alternative to population collapse is an increase in the carrying capacity.

The carrying capacity can increase as a result of technological progress. This is probably what happened in early modern England.

The carrying capacity may also increase as a result of the conquest of new underpopulated territories. An example is the conquest of the Kazan and Astrakhan khanates by Muscovy in the sixteenth century.

The manner in which elite overproduction is abated depends very much on the military strength of the aristocracy. A nonmilitarized ruling class can be expropriated en masse by warlords, such as rebel generals or even peasant bandits. A rapid and comprehensive elite turnover results in a relatively short period of sociopolitical instability that follows state collapse.

A rapid elite turnover can also result when there is a ready external source of potential elites, as was the case in the Maghreb described by Ibn Khaldun.

For a new secular cycle to get going, the pressures of the general population on resources and of the elites on commoners must be substantially reduced from their precrisis levels.

It is entirely possible for the civil warfare to gradually die out but a centralizing, integrative trend nevertheless failing to take hold. In this case, the area in question may persist indefinitely (or until it is conquered from the outside) in a fragmented state as a collection of small-scale polities.

Oscillatory dynamics do not go through truly discrete phases with clearly marked breakpoints, but for convenience in talking about each secular cycle, we need to divide it in phases.

Most broadly the cycle can be divided into two opposite trends. In the literature these are sometimes called the positive “A phase” and the nega­tive “B phase,” but we prefer the more descriptive terms integrative and disintegrative trends. Politically the integrative phase is characterized by a centralizing tendency, unified elites, and a strong state that maintains order and internal stability. Internal cohesion often results in the vigorous prosecution of external wars of conquest, which may result in the extension of the state territory (assuming there are weaker neighbors at whose expense the state can expand). The disintegrative phase, by contrast, is characterized by a decentralizing tendency, divided elites, a weak state, and internal instability and political disorder that periodically flare up in civil war.

More frequently it is the external enemies that profit from the internal weakness of the state and society, resulting in an increased frequency of raids, invasions, and loss of territory.

The population tends to increase during the integrative phase and decline or stagnate during the disintegrative one.

During the disintegrative phase, by contrast, population losses due to epidemics, famines, or wars are not made up by sustained population growth.

It is useful to further divide the broad integrative and disintegrative periods into subphases. Population growth is particularly vigorous during the first, expansion phase of the integrative trend. This is a time of relatively stable prices and modest real wage declines (if any). However, as the population density begins to approach the limits set by the carrying capacity, price increases or wage declines accelerate—this is the “stagnation” or “compression” or even more descriptively stagflation (stagnation plus inflation) phase. Although the majority of commoners experience increasing economic difficulties during the stagflation phase, the elites enjoy a golden age, and their numbers and appetites continue to expand. The stagflation phase (and the overall integrative trend) is succeeded by a general crisis. Whereas expansion grades smoothly into stagnation, the transition between stagflation and crisis is often (but not always) abrupt.

The crisis grades smoothly into a depression phase, characterized by endemic civil warfare. The population may grow during the intervals between intense civil wars, but such increases typically do not last and are followed by declines (although not as a catastrophic as those typical of the crisis phase). The depression phase ends when the ranks of elites are pruned by internal conflict to the point where the disintegrative trend can reverse itself, and a new secular cycle begins. Alternatively, if no functioning state can get going, then the depression phase grades smoothly into an intercycle of indeterminate length.

We wish to emphasize again that the classificatory scheme we propose is an ideal type.

1.3 Variations and Extensions

"The theory should, in principle, be applicable to any agrarian society.

one of the most important characteristics of oscillatory dynamics, (is) the average period of a cycle. Secular cycles are not periodic in the strictly mathematical sense, in which each succeeding cycle repeats exactly the preceding one.

Macrosocial dynamics of agrarian states cannot be strictly periodic. There are at least three reasons for that. First, nonlinear dynamic feedbacks can in theory generate not only strictly periodic (cyclic in the mathematical sense) dynamics but also aperiodic chaos.

Social systems are complex and feedback loops are nonlinear, so the possibility of chaos cannot be discounted (Turchin 2003b).

Second, the dynamics of agrarian states are affected not only by their internal workings but also by exogenous forces, such as changes in their geopolitical and ecological environment. Exogenous factors, unlike endogenous ones, are those that are not part of feedback loops.

Finally, individuals possess free will and can act in unpredictable ways. In principle, even the act of a single person, if it takes place in the right place at the right time, may be able to influence the trajectory of a whole society.

For all these reasons, we do not expect a strict periodicity in secular dynamics. Instead, dynamics should have an average period, a characteristic time scale, with a substantial degree of variation around this average.

Thus, the typical length of the expansion phase is primarily determined by

(1) the per capita rate of population increase and

(2) the population density in relation to carrying capacity at the beginning of the cycle.

For example, if population grows at the rate of 1 percent per year, it takes seventy years for it to double.

Abnormally long expansion phases result from successful territorial conquest, especially when it is accompanied by colonization of conquered territories, which serves to reduce population pressure in the metropole.

The length of the crisis phase is much less predictable, because while there is a definite biological limit on how fast a human population can grow, there is no comparable limit on how fast it can decline.

The characteristic lengths of the stagflation and depression phases depend more on the state and, particularly, on elite dynamics than on what the general population does.

Models tailored to the characteristics of Western European societies (largely monogamous elites enjoying a preponderance of military power over their internal and external enemies) suggest that the typical periods of secular cycles in these societies should lie in the range of two to three centuries.

Ibn Khaldun Cycles

The Maghrebin states differ from Western European states in two important respects:

(1) these Islamic societies permitted polygyny and

(2) there was a ready source of militarily powerful counterelites nearby.

Because aristocratic males could afford to support several wives and concubines, the rate of elite population growth in Islamic societies was (and is today) much greater than that for elites in Christian societies.

The fact remains that the biological reproduction rate of Islamic elites was several times higher than that of Christian elites.

The second factor is the location of Maghrebin societies in the rather thin strip of arable land squeezed between the Mediterranean Sea and the desert.

As soon as a Maghrebin society experienced state collapse, it became extremely vulnerable to conquest from the desert.

When demographic-structural models are modified to account for these two factors they exhibit very different dynamics (Turchin 2003b). High reproductive rate of the elites means that they increase much faster than the general population.

As a result, the integrative trend of the secular cycle is over much faster than in the standard model.

As a result of a shortened integrative trend and a missing depression phase, models predict a much faster secular cycle for Maghrebin-type societies, on the order of one century, rather than the two to three centuries for Western European states. This prediction is in agreement with the observation of Ibn Khaldun that the dynastic cycle in the Maghreb extends, on average, over four generations (a generation time in humans is typically twenty to thirty years).

Islamic societies that controlled the elite growth rates in one way or another are predicted to exhibit slower cycles, with periods similar to those observed in Western Europe. For example, in the Ottoman Empire the sultans had access to an essentially unlimited supply of wives and concubines. However, when the old ruler died, only one son was allowed to replace him; all others were killed. Furthermore, top levels of bureaucracy and army leadership were recruited not from native elites but by means of devshirme. In other words, the state, not biology, controlled the size of the high-ranking elite stratum. Only lower-rank landed elites were permitted to increase “biologically,” and, being not very wealthy, they could not afford too many wives.

The Mamluk polity in medieval Egypt: Its ruling class was recruited entirely from the slave markets. Children of Mamluks could not be Mamluks, and thus automatically dropped out of the ruling class.

The Fractal Nature of Historical Dynamics

Different social processes operate at a variety of temporal scales.

Different questions require approaching an analysis at different time scales.

Turning now to population dynamics, we observe that population changes also occur on a variety of scales: monthly (female menstrual periods), yearly (subsistence and epidemic cycles), generational (somewhere between two and three decades), and secular (one, two, or three centuries, according to the theory of secular cycles).

More information

  • intro the thematic: Korotayev, A., A. Malkov, and D. Khaltourina. Introduction to Social Macrodynamics:Secular Cycles and Millennial Trends. Moscow: KomKniga/URSS, 2006. P. 37