THE COLLAPSE OF COMPLEX SOCIETIES PDF

adminComment(0)

Printed in Great Britain at the University Press, Cambridge. British Library cataloguing in publication data. Tainter, Joseph A. The collapse of complex societies. "While the theoretical part of the book is quite remarkable and based on exceptional erudition, I also found the accumulation of the supporting data to be. The sublime mystery of collapsed civilizations and dark portents (“the possibility that a . “In a complex society that has collapsed, it would thus appear, the.


The Collapse Of Complex Societies Pdf

Author:ARLEEN JARRARD
Language:English, French, Portuguese
Country:Marshall Islands
Genre:Children & Youth
Pages:391
Published (Last):16.06.2016
ISBN:801-3-70003-657-7
ePub File Size:27.88 MB
PDF File Size:8.80 MB
Distribution:Free* [*Sign up for free]
Downloads:36453
Uploaded by: MARIELA

joseph tainter the collapse pdf. Societal collapse is the fall of a complex human society. Such a disintegration may be relatively abrupt, as in the. PDF | Review of book by Joseph Tainter that provides a systematic analysis of the collapse of ancient civilizations around the world. Joseph Anthony Tainter (born December 8, ) is an American anthropologist and historian. According to Tainter's Collapse of Complex Societies, societies become more complex as they try to solve problems. the original (PDF) on 15 December , retrieved 11 October ; ^ "In the collapse of states Tainter has.

So the change becomes not really an improvement, but instead a permanent commitment to additional infrastructure, additional complexity. Why then don't I live a life of ease with the US average 10kW at my disposal?

The answer is that we use it up. Also I suppose I DO live a life of ease and luxury in comparison to a 19th century Russian peasant with only one actual horse in his stable. Anyway, Tainter does talk about this obliquely at one point. Re-reading my review a year later, both Tainter's analysis and mine leave me uncomfortable, unsatisfied. Trying to be succinct, I don't think it's about diminishing returns, but about slack and momentum. Consumption of all the slack technology provides leads to an irrevocable commitment to complexity and interdependency.

Commitment means you can't retrench to simpler technology in a bad year because the greater efficiency has become the new minimum baseline. Now you're in danger of actual collapse.

One more last thought how many "last thoughts" is one allowed? The farmer downloading the fast tractor doesn't see the oil cost in it's impact on fertilizer cost, or food prices in poor countries, or something similar, so he makes what is for him the right Local decision, but Globally, it is the wrong one.

Of course bringing oil into it pulls Jared Diamond to mind. Only soil, Diamond's favorite limited resource, could be more apt. Complexity encompasses both social stratification the proliferation of classes or castes or estates and also social differentiation the multiplication of social roles, professions, occupations, specialisms and so on.

See a Problem?

Our own world is marked by an extreme social division of labour, by significant inequality and very low levels of self-sufficiency. In essence this is what makes a society complex.

Shifts in complexity Systems theory offers not only a way of distinguishing traditional from more complex societies but a means of describing transitions from one to another. It has been observed that larger social entities rarely emerge gradually but more often undergo sudden shifts as new levels of complexity are added.

Urbanization, for example, often occurs as a step change even when it follows a long period of demographic growth. States often form through rolling up a series of more local communities. Empires in particular are often created in very rapid periods of consolidation: this is true of the Qin unification of China at the end of the Warring States period, or of the Roman takeover of the Mediterranean.

Joseph Tainter

Sudden increases in scale pose new challenges of coordination. Many phase transitions fail at this stage but those that stabilize have successfully added another level of complexity. Systems Collapse. Conversely, rapid decline can be imagined as the loss of one of more level of complexity. We are aware that the end of the Roman Empire in the west was accompanied by the reduction of differentials between the richest and the poorest, by a reduction in the division of labour, by reduced long distance communication and a narrowing of social and political horizons.

This is aptly described as a reduction in complexity. The easiest way to imagine this is in terms of human experience. A complex social world was replaced by one in which a greater proportion of the population were peasant farmers and many professions have ceased to exist.

There are other indices of diminished complexity too: settlement patterns offer a good proxy indicator of social complexity, and so too do the range of artefacts produced and used. So far complexity theory offers a more precise way of envisaging decline. But it also offers ways of understanding it. One of the earliest attempts to formulate a theory of systems collapse applicable to ancient civilizations was that of Renfrew who theorized that the earliest complex systems — he was thinking of the Aegean Bronze Age, of Mesopotamia after the Uruk phenomenon, and of the Indus Valley civilization - were typically precarious hierarchies.

Bronze Age civilizations were created by coordinating the activities of scattered villages, by drawing off a greater surplus and using it to sustain non-producers soldiers, priests, kings, gold workers and so on. The margin of surplus was so low, and the means of organization so difficult, that these systems were vulnerable to the slightest disruption. A change in crop yields, a new military stress, the collapse of a trading partnership might be enough to destabilize the whole at which point it collapsed back to the next level of complexity.

Villages became self governing again, tribute to the centre stopped and so on. This is very close to what has been argued for the paramount chiefdoms of early Iron Age Europe or the immediate sub-Roman populations of western Europe after the fifth century.

Those smaller, simpler worlds might be reconnected again in time, or they might not. Meanwhile a reduction in complexity simply meant a reversion to village life. There are potential objections raised to this mode of analysis, some more cogent than others. One common response is to challenge decline on empirical grounds: the Dark Ages were not that dark. Another is to claim that this mode of analysis includes a build in preference for complexity and fails to recognize simpler organization as simply different.

Bryan Ward-Perkins has made a robust rebuttal of those approaches in so far as the Fall of Rome is concerned. For prehistory this is certainly true. One thing complex societies do that simpler ones do not is generate records and monuments.

Joseph Tainter

Dark Ages are called dark for a good reason. This does not make the thesis of collapse less plausible of course: it just makes it more difficult to support. Yet the idea that some complex societies are more resilient than others has some mileage in it. Renfrew was correct that the collapses of Bronze Age civilizations were more comprehensive than many of those that followed.

It took only a few centuries for Mesopotamian urbanization to recover following the collapse of the Uruk civilization, only half a millennium for states and cities to reappear in the Aegean after the fall of the palaces, and even in the British Isles the gap between the collapse of Roman systems of government and manufacture and early mediaeval growth was not much longer. A systems collapse is never absolute, and the fragments and memories of previous complex systems make it easier to rebuild.

Perhaps too those complex societies that work more thorough transformations of the smaller scale societies out of which they were assembled, make their own resurrection more likely. The palatial civilization of Minoan Crete might have been a more superficial overlap on early Bronze Age societies, than was the Roman Empire in the west which on its collapse left behind cities, a common language and much communications infrastructure far superior to anything the societies it conquered had replaced.

As for the triggers of collapse, debate continues. A collection edited by Cowgill and Yoffee attempted through case studies to consider competing causes for the collapse of complex societies. These studies, and others, consider the impact of migrations of population, climate change, warfare and so on, but few general conclusions emerged. As societies become more complex, he argues, the marginal returns for increasing complexity increase.

Collapse, he suggests, is often a result of the diminishing value of complexity. This is naturally formulated at a quite abstract level but the proposition is worth considering, especially when we consider that the cost of the tasks complex societies perform varies over time. So too do the profits of complexity. Consideration of the case of Rome shows that this approach does not provide us with a silver bullet, a magic answer to the question Why did Rome Fall?

Complexity theory does however allow us to think about survival in a more precise way. The Roman Empire could sustain or grow its complexity so long as it was able to carry out its functions within a budget set by its success at raising revenue. Failure might come from rising costs whether military threats, or the cost of a more complex administration or both or diminishing profits whether caused by demographic decline, the end of the Roman Warm Period or increased inability to get access to surpluses produced within the empire, whether because of corruption, a diminution of loyalty or other factors.

Of course costs may have risen at the same time as profits fell. Whatever the reasons we see the empire eventually resorting to reducing its complexity, by reducing its governmental ambitions for the territory it controlled and ultimately by reducing the scope of the territory it tried to control.

The post-Justinianic empire collapsed back ultimately into something more like the Athenian Empire in scale, an Aegean hegemony run by a single city-state with grandiose ambitions beyond the area it could tax and police. None of this was planned of course, which brings us to the final criticism of a systems theory approach to decline, that it has little place in it for individual human beings.

Much of the attack on processual archaeology in the nineties revolved around its lack of engagement with the social sciences and social theory in particular, and with traditional humanistic approaches to the past. It is true that there is little space for human agency in this sort of narrative, and some historians do find this repellant. I disagree. A better understanding of the nature of decline — and growth — enables us to see better the constraints within which individuals did exercise their agency.

How far it ever made a major difference to the unrolling of events on this scale must be doubtful. A characteristically thoughtful discussion is that of Bowersock, On late antique migrations Heather, , with comment by Kulikowski, Diamond, represents an attempt to generalize but its focus on ecological strain, in some ways Neo-Malthusian, derives from a rather small set of case studies.

Migration in Archeology. The Baby and the bathwater. American Anthropologist, 92 4 , Finally, it is genuinely difficult to disentangle causes of change from data of this kind. No wonder migrations seemed such a convenient explanatory device. For later European prehistory, the migration hypothesis seemed further supported by classical testimony about Gallic invasions of Italy, the Celtic attack on Delphi and invasion of Anatolia, the movements of the Cimbri and Teutones and late antique movements of Goths, Alamanni and Huns, even if it remains enormously difficult to link any of these historically attested population movements to wholesale changes in material culture.

Even V. Gordon Childe, whose grand view of historical change was based on a generally optimistic historical materialism, wrote of the whole of human history as a series of peaks and troughs.

Archaeological notions of decline in a grand narrative of this kind were limited to describing one phase of a familiar sequence, the downward motion through which once culture passed rapidly away before being replaced by another. Declines of this sort were the essential concomitants of periods of growth. Systems Theory and Decline Processual archaeology, which developed in opposition to culture-historical approaches, was much more concerned with social evolution than with taxonomy, with explaining sequence of change rather than characterizing cultures.

Societies were now described as systems of variable complexity, systems through which the distribution of goods and information were managed. Some historians and archaeologists of less comparativist or anthropological inclination were tempted to treat all this as jargon, as a scientizing rhetoric that disguised what in effect was putting old wine in new bottles. Yet complexity theory has several advantages over the more confused humanistic formulations with which I began this paper.

First, it is very clear about what it is that changes, viz: Second, it offers a more detailed account of the social effects of a shift in complexity. Third it provides the basis for some empirically based observations of the kind of circumstances in which shifts in complexity are most likely to occur. Levels of complexity. Complexity encompasses both social stratification the proliferation of classes or castes or estates and also social differentiation the multiplication of social roles, professions, occupations, specialisms and so on.

Our own world is marked by an extreme social division of labour, by significant inequality and very low levels of self-sufficiency. In essence this is what makes a society complex. Shifts in complexity Systems theory offers not only a way of distinguishing traditional from more complex societies but a means of describing transitions from one to another.

It has been observed that larger social entities rarely emerge gradually but more often undergo sudden shifts as new levels of complexity are added. Urbanization, for example, often occurs as a step change even when it follows a long period of demographic growth.

States often form through rolling up a series of more local communities.

Empires in particular are often created in very rapid periods of consolidation: Sudden increases in scale pose new challenges of coordination. Many phase transitions fail at this stage but those that stabilize have successfully added another level of complexity.

You might also like: THE EMPERORS HANDBOOK PDF

Systems Collapse. Conversely, rapid decline can be imagined as the loss of one of more level of complexity. We are aware that the end of the Roman Empire in the west was accompanied by the reduction of differentials between the richest and the poorest, by a reduction in the division of labour, by reduced long distance communication and a narrowing of social and political horizons.

This is aptly described as a reduction in complexity.

The easiest way to imagine this is in terms of human experience. A complex social world was replaced by one in which a greater proportion of the population were peasant farmers and many professions have ceased to exist.

There are other indices of diminished complexity too: So far complexity theory offers a more precise way of envisaging decline.

Collapse of Complex Societies

But it also offers ways of understanding it. One of the earliest attempts to formulate a theory of systems collapse applicable to ancient civilizations was that of Renfrew who theorized that the earliest complex systems — he was thinking of the Aegean Bronze Age, of Mesopotamia after the Uruk phenomenon, and of the Indus Valley civilization - were typically precarious hierarchies.

The margin of surplus was so low, and the means of organization so difficult, that these systems were vulnerable to the slightest disruption. A change in crop yields, a new military stress, the collapse of a trading partnership might be enough to destabilize the whole at which point it collapsed back to the next level of complexity. Villages became self governing again, tribute to the centre stopped and so on. This is very close to what has been argued for the paramount chiefdoms of early Iron Age Europe or the immediate sub-Roman populations of western Europe after the fifth century.

Those smaller, simpler worlds might be reconnected again in time, or they might not. Meanwhile a reduction in complexity simply meant a reversion to village life. There are potential objections raised to this mode of analysis, some more cogent than others. One common response is to challenge decline on empirical grounds: Another is to claim that this mode of analysis includes a build in preference for complexity and fails to recognize simpler organization as simply different.

Bryan Ward-Perkins has made a robust rebuttal of those approaches in so far as the Fall of Rome is concerned.

For prehistory this is certainly true. One thing complex societies do that simpler ones do not is generate records and monuments. Dark Ages are called dark for a good reason. This does not make the thesis of collapse less plausible of course: Yet the idea that some complex societies are more resilient than others has some mileage in it.

Renfrew was correct that the collapses of Bronze Age civilizations were more comprehensive than many of those that followed. It took only a few centuries for Mesopotamian urbanization to recover following the collapse of the Uruk civilization, only half a millennium for states and cities to reappear in the Aegean after the fall of the palaces, and even in the British Isles the gap between the collapse of Roman systems of government and manufacture and early mediaeval growth was not much longer.

A systems collapse is never absolute, and the fragments and memories of previous complex systems make it easier to rebuild. The palatial civilization of Minoan Crete might have been a more superficial overlap on early Bronze Age societies, than was the Roman Empire in the west which on its collapse left behind cities, a common language and much communications infrastructure far superior to anything the societies it conquered had replaced.

As for the triggers of collapse, debate continues. A collection edited by Cowgill and Yoffee attempted through case studies to consider competing causes for the collapse of complex societies. These studies, and others, consider the impact of migrations of population, climate change, warfare and so on, but few general conclusions emerged. As societies become more complex, he argues, the marginal returns for increasing complexity increase. Collapse, he suggests, is often a result of the diminishing value of complexity.

This is naturally formulated at a quite abstract level but the proposition is worth considering, especially when we consider that the cost of the tasks complex societies perform varies over time. So too do the profits of complexity.

Consideration of the case of Rome shows that this approach does not provide us with a silver bullet, a magic answer to the question Why did Rome Fall? Complexity theory does however allow us to think about survival in a more precise way. The Roman Empire could sustain or grow its complexity so long as it was able to carry out its functions within a budget set by its success at raising revenue.

Failure might come from rising costs whether military threats, or the cost of a more complex administration or both or diminishing profits whether caused by demographic decline, the end of the Roman Warm Period or increased inability to get access to surpluses produced within the empire, whether because of corruption, a diminution of loyalty or other factors.

Of course costs may have risen at the same time as profits fell. Whatever the reasons we see the empire eventually resorting to reducing its complexity, by reducing its governmental ambitions for the territory it controlled and ultimately by reducing the scope of the territory it tried to control. None of this was planned of course, which brings us to the final criticism of a systems theory approach to decline, that it has little place in it for individual human beings.

Much of the attack on processual archaeology in the nineties revolved around its lack of engagement with the social sciences and social theory in particular, and with traditional humanistic approaches to the past.

It is true that there is little space for human agency in this sort of narrative, and some historians do find this repellant.

I disagree. A better understanding of the nature of decline — and growth — enables us to see better the constraints within which individuals did exercise their agency. How far it ever made a major difference to the unrolling of events on this scale must be doubtful.I believe it!

New Brunswick: Rutgers University Press. Town Planning Review, 80 1 , The Urban Revolution.

Navigation menu

The institutions of complex societies often conflict with our social instincts. Two seem especially significant for the present question. Dynamic cycles of Mesoamerican states. Goths and Romans Archaeological Perspectives. Trying to be succinct, I don't think it's about diminishing returns, but about slack and momentum.

DEVIN from Kansas City
I enjoy reading novels often . Browse my other posts. I have only one hobby: pitch and putt.
>