How Chiefs Come to Power (1)

Timothy Earle’s How Chiefs Come to Power: the Political Economy in Prehistory is a study sufficiently well-grounded in archaeological and anthropological evidence that the well-presented facts rather escape from the theoretical framework Earle wants to fit them in.

I originally was alerted to Earle’s work via a harsh review archaeologist Michael E. Smith had posted on Yoram Barzel’s book on the theory of the state. I greatly admire Barzel’s book on property rights but had not read this later work, so I was intrigued by the response. Having read a selection of Prof. Smith’s publications and finding them informative, very accessibly written and analytically sensible, his strong endorsement of Earle’s book was encouraging. Particularly given he also admired a book I had read and thought excellent.

How Chiefs Come to Power Earle is attempting to answer a straightforward question:

What allows aspiring leaders to be successful in one situation but fail utterly in another? (p.2).

In other words, since aspiration to leadership and power are part of the human condition, why does rulership (in its first stage of chiefdom) occur in some places and not others?

Earle distinguishes between authority, “the right and responsibility to lead”, power, “measured by the mastery a leader exercises over others” and control, “the ability to restrict access to the resources which are the media from which power can be fashioned” (Pp3-4). As Earle accepts a Lockian-Marxian notion of property and value coming from labour, his thinking about control gets a bit confused: all property is based on control—an act of trade is an exchange of control. Things are clearer if it is grasped how basic to human society the notion of property—of legitimate control—is, something that grows up and extends (pdf) well before the development of authorising authorities. But that requires some notion of spontaneous order based on a strong notion of beneficial exchange. For it is not labour that distinguishes the productive power of homo sapiens from other primates—all animals engage in mere labour—it is the notion of property and thus the power of exchange: ideas having sex in Matt Ridley’s vivid metaphor. How our species' distinctive creation of niches through discovery and transmitted learning generates its productive power.

Earle identifies the sources of power as social relationships (particularly kinships), economic power (being able to buy compliance), military might (coercing compliance) and ideology (routines of compliance). He holds that kinship, military might and ideology are, on their own, all weak sources of power while economic power depends on military power to defend resources and ideology to institute rights of unequal access (Pp4ff). The discussion is punctuated with references to various Marxist positions on the broad issues. (An Israeli archaeologist once explained to me that archaeologists tend to adopt broadly Marxian positions as so much of what they study is the product of the generation of economic surpluses beyond subsistence.)

The problem with political power is that it:

… must be inherently problematic, as it is contingent on multiple factors that can be used against central authority as well as being used by it (p.10).

Yet the long-term trend of world history has been for central power to expand and the number of independent polities to shrink: from maybe 100,000 at the time of the Neolithic agricultural revolution to about 160 now, the largest covering over a billion people (p.10). Clearly, the integration of the sources of power has become more effective over time: a selection process in favour of more successful techniques.

Earle holds that:

… the different sources of power are fundamentally intertwined and interdependent, but that they grow from a material base (p13).

Very Marxian, but surely somewhat dubious: after all, how does one get that material base in the first place? That they have to be sustained by a material base is clear, but that is not quite the same.
Read More...
Earle takes the view that the process of creating and stabilising power is multi-linear: different techniques win the process of selection in different circumstances. The key variable to success or failure is the chief’s ability to control and extend access to the sources of power (Pp12ff).

A chiefdom being:

a regional polity with institutional governance and some social stratification organising a population of a few thousand to tens of thousand people (p.14).

Making them intermediates between village-based polities and large, bureaucratic states. Hierarchies are used for coordination, granting advantages resulting in social stratification: the use of monumental constructions and prestige goods being used by archaeologists to document their evolution. All of which makes them emergent polities. Earle holds they have fundamentally the same dynamics as states, so are a good basis for studying rulership more generally, including the basis of modern states (Pp14-5).

The book is based around three examples, each covering about a millennia: Neolithic and early Bronze Age Denmark (2300-1300BC); Hawai’i from settlement to incorporation in the modern world (800-1824); and high Andes Peru to the Inca conquest (500-1534). By interweaving the analysis of the three examples, and how specific (particularly economic) circumstances affect choices and outcomes, Earle seeks to reveal the dynamics of rulership (Pp15-6).

Having established the framing, Earle sets out the long-term dynamics of the chiefdoms of the three regions one after another (Pp17ff). The Andean case has some puzzling population surges: the first likely due to the introduction of new crops. The second occurred without any noticeable technological change and during a period of glacial advance: the production of extra warriors for defence during a period of increased warfare is Earle’s suggested explanation (Pp54-5).

Though the chosen examples display different pathways to complexity, Earle believes that the comparison is still analytically fruitful: such as the complex interaction with population pressures. Earle writes of the animal herding in the Denmark case study:

Animal herding lowers the carrying capacity of a region and makes no sense in the logic of a subsistence economy. Its increase most probably resulted from the use of animals as a source of moveable wealth within a prestige good exchange system (p.65).

Really? Power can consolidate itself by reducing the productiveness of its area? Whether looked at in terms of internal dynamics or competitive interaction with other cultures, this looks deeply implausible. Earle's claim about carrying capacity is true in some areas but highly misleading and does not sustain his conclusion. In some areas, such as river valleys and plains with particularly rich soils, farming is the more productive activity in the sense that it can support more people. In other areas, not so much. What herding typically does is produce more value for less effort. So animal herding can increase the resources available to a culture: otherwise it would not have evolved in so many different cultures and would not be persisted with. That herding societies – which certainly had chiefdoms, even states – were freer and richer than agrarian societies was a perennial observation of visitors from agrarian empires. It does require trade with farming cultures, and clearly access to prestige goods from trade can consolidate a ruler’s position, but it does so precisely by increasing the access to goods from choosing more productive activity. Earle could do with more confidence in human rationality and the implications of beneficial exchange.

What animal herding also provides is inheritable and expandable assets. There is the resource base for social differentiation – something herding societies tend to display. Earle then points to evidence that expanded agricultural activity (such as through irrigation works) in Hawai’i and Peru supported expanded surpluses rather than expanded populations, surpluses used for elite purposes (Pp64-6). So why wouldn’t animal herding also represent expanded capacity?

In examining the sources of economic power, Earle begins:

The development of political systems must be based on an ability to lead and to have others follow. Leaders who provide rewards are accepted, but those who fail to do so lose authority. Leadership is always problematic; it requires an individual, family, community, or nation to relinquish and transfer autonomy and transfer loyalties. No individual or group does this joyfully, for it means sacrificing personal or group interests to a different, larger, or distant body (p.67).

This seems to rather preclude that people may see political leadership as some sort of exchange beyond direct rewards. Unless there is some sense of “return” for accepting leadership, why would not revolt be a permanent (or at least a permanently incipient) thing? Do leaders not provide public goods, for example?

Earle says of economic control that:

Control over the economy is a direct and material control over the lives of people (p.67).

He then asks if such control is possible in a subsistence economy, exploring two lines of analysis: voluntarist, adaptionist theories seeing chiefdom as a form of solution to social problems or way of expanding social possibilities and coercive political theories arguing that autonomy is not give up except in the face of power. (Why the answer cannot be both escapes me: coercion as a way of delivering public goods.)

Earle argues that the model developed by Marx and Engels “may have general applicability to the emergency of social complexity” (p.70). Since it is based on a false theory of value, an incoherent theory of exploitation and has a truly terrible track record when operationalised as a political theory or economic management approach, I am more sceptical. Particularly as one starts off with two deep problems for analysis: first, not grasping that all exchange is about swapping control (i.e. property is prior) and, second, that exchange results in mutual improvement (i.e. gains from trade).

Earle differentiates staple finance (payments to rulership, either in kind, goods or labour service) and wealth finance (objects such as primitive valuables, prestige goods or money exchanged within the ruling elite). For the former, he sees land tenure as vital:

The critical factor appears to be how the development of technologically intensive farming provides the opportunity for control by ruling institutions through a land tenure system (p.71).

First, what about herding? Second, land is stationary, therefore farmers are more easily controlled whether they are direct owners of the land or not. (We come back to pastoralist societies being freer and wealthier than agrarian ones.)

That Earle may be going for rather too much analytical complexity here is indicated by his partly perceptive, partly way over-egged analysis that:

Technological improvements of the resource do two things. First, they radically differentiate land in terms of quality. … Second, improvements evidently delimit and mark the resource in ways that can be easily represented and recognized within the cultural landscape. Improvements such as walls, terraces, and ditches materialise the division of the landscape and form the basis of a cultural system of land ownership (Pp 71-2).

I get that academics often like talking about ideas (their expertise) and not the niceties of production and exchange (their distance from which often feeds into their sense of status), but really. Farming makes people much easier to control, the creation of a surplus-to-subsistence allows the support of an elite, expandable and inheritable productive assets promote social differentiation. Clearer and simpler, surely? What’s more, it is also usefully extendable, yet differentiable, to the herding case.

Trying to build irrigation into some generally important factor in the development of rulership – a continuing theme in Earle’s discussion: his original PhD thesis, a critique of the irrigation management thesis of Wittfogel and Service, analysis Hawai’i chiefdoms as using ownership of irrigation as a form of economic control and extraction – strikes me as dubious. Irrigation is surely much more likely to have its largest (and most common) effects in establishing patterns and structures of interactions between farmers. (One of Elinor Ostrom’s anecdotes in this presentation goes to that.) But, as previously noted, notions of spontaneous order do not appear to figure much in Earle’s mental landscape: he seems to lack the necessary building blocks of mutually beneficial cooperation as a pervasive feature of human economic interactions.

The sense of rather too much missing the basic point continues:

In chiefdoms, wealth is the means of symbolising relationships upon which social ranking rests. Since the social structure is the most important determinant of cultural, political and economic values, substantivist economists define the economy as “institute process” (Pp73-4).

Wealth is not a social ranker in itself? Nor a benefit in itself? Does not any signalling effect flow from that? And, if the social structure drives cultural, political and economic values, what creates and drives the social structure? Human action does not seem to be causally gradable in such a way.

So Earle’s contention that:

control over the ideology of social ranking rested on control over the system of wealth finance (p.74)

rests on the same dubious causal ranking. Building on his original dissertation, Earle argues that the use of control over irrigation by Hawai’ian chiefs fails to conform the “managerial” “solution to problems” hydraulic analysis of Wittfogel. The control by the chiefs was the dominant factor, with land use rights of commoners flowing from provision of corvee labour: so a widow could only retain land use if she remarried, transferring the rights to the new husband (Pp75ff).

The structure is not so different from medieval serfdom, where a warrior elite protected serfs in return for labour and inheritance generated an interest in long-term investment in making the lands more productive. Clearly, the elite control has to be accepted, but military power, protection and access to land provide a potent combination to ensure acceptance. I see no need to reify ideology in quite the way Earle does, but then I have no problem with property and exchange being endemic to human society.

If, however, you do not see property as endemic to human societies (particularly in farming societies), nor the control relationship as in any way protective, then the basic dynamics of power and property does become more a mystery to be explained. Hence Earle discusses the differences in dynamics between irrigation and dryland agriculture in Pacific islands, and particularly Hawai’i, making rather more of a meal of what seem to be the practical dynamics of control and extracting a surplus than seems necessary (Pp84ff). Particularly as intensive irrigation seems to have grown up in a period of intensive competition between chiefdoms – which is to say, one where the value of providing protection and extracting maximum value from the lands one does control was maximised, while the production possibilities of irrigation may have also encouraged conflict – more surpluses to seize.

That Earle rather doesn’t “get” the protective role is indicated by this following passage in his discussion of the emergence of chiefdoms in Denmark:

during this period permanent settlements were placed prominently on the highest ground. Why would these settlements have been built on these exposed hills, where the winds bite mercilessly? Perhaps their high visibility declared the territorial rights of the community over the lands upon which the settlements looked (p.100).

And perhaps tops of hills were easier to defend and made it easier to see enemies coming? That would be my vote, given it was the normal reason people set up their homes on tops of hills.

As herding developed, so did chiefly power. Earle does note that cattle are controllable assets that permit social differentiation. Though he seems not to understanding pastoralism very well:

Cattle … would have difficult to control unless their pasturelands were owned. The open pastures were naturally unbounded, and it must have been difficult to define their ownership and rights of use (p.101).

Pastoralist societies did not find it particularly so: with techniques such as branding, regulated use of common lands, the considerable memory capacity of pre-literate societies, they managed.

Earle then makes a meal of arguing that trading for metal goods would have consolidated chiefly power (Pp100ff): metal weapons would certainly have increased warrior power and the connection between cattle-owning chiefs and their warrior bands.

The ability to control economic resources and to extract a useable surplus does seem to have a great deal to do with the extent and permanence of chiefly power (p.104): but Earle makes reaching this unsurprising conclusion much more analytically inelegant than he needs to, not because the cases are complex, but because he too much wants to have social causality to be structured in a particular way and is too ready to overlook enduring basic patterns.


This review will be concluded in my next post.