Population, Coercion and State Formation. Some Comments on Robert Carneiro's Reformulation of Circumscription Theory

Population, Coercion and State Formation. Some Comments on Robert Carneiro's Reformulation of Circumscription Theory
Author: Sneath, David
Journal: Social Evolution & History. Volume 11, Number 2 / September 2012

In refining his classic theory, Robert Carneiro moves from circumscription to something we might think of as a concentration model of state formation. In this narrative populations become confined by a concentration of resources that leads to intensified conflicts and an increased likelihood of the conditions needed for the formation of the state – in particular warfare and the creation of a strata of unfree subjects.

Carneiro's central insight, to my mind, remains highly convincing – warfare and coercion were central to the formation of early polities. He rejects the notion of a purely ideological process that could generate a state form. This is strong theoretical ground to occupy, and Clastres, for example would agree on the primacy of coercion – since the economic relations of class require ideology. ‘The political relation of power precedes and founds the economic relation of exploitation … the emergence of the State determines the advent of classes’ (Clastres 1977: 167–168). This remains persuasive; Engels's refutation of Dühring on this point was never very convincing, and more recently postcolonial studies have emphasized the foundational nature of violence to political orders (e.g., Bhabha 1995; Mbembe 2003) and fashionable social theory has emphasized the coercive aspect of knowledge systems, so that a sort of convergence of coercive and ideological visions becomes plausible – the ‘religion with teeth’ concept, as Carneiro describes it.

Carneiro's notion of the pendragon leader is certainly an improvement on Lewis Henry Morgan's early equivalent – the Ancient Greek basileus (‘king’). In Morgan's scheme the basileus was not really a monarch but a general military commander chosen for collective defense. But Morgan's claims were supported only by the logic of his own speculative model of kinship society in which the essentially egalitarian corporate groups he termed gens (clans) must have behaved in a certain way (Morgan 1964[1877]: 210).1 Carneiro's pendragon, however, stands only for the possibility of elective military leadership, and is much more robust since his wider scheme does not require a speculative kinship structure; he takes the basic unit to be the ‘village’ residential group rather than some imagined kinship structure. The point is simply that both the projects of conquest and defense can generate king-like power. Having been naturalized, for such a long span of human history, hereditary power became so thoroughly denaturalized in the age of national populist politics that the social sciences required a convincing mechanism to explain its presence. Carneiro's theory provides one, but refreshingly, it does not depend upon some variant of the social contract theory logic that Trigger (1993) identifies as the implicit rationale behind so many narratives of state formation. The hierarchy of a conquest polity may be an end in itself for the rulers; its structure need not reflect any sort of adaptive advantage or general social interest.

The move to stress the attractive power of resources rather than the ‘caging’ potential of environmental circumscription is another improvement, as is Carneiro's broader point that ‘population pressure’ (i.e. competition over resources) can be said to exist in regions with very low average population densities. Colonial expansion frequently involved settlers annexing ‘terra nullius’ land that they claimed to be virtually empty or under-used, but nevertheless this frequently led to competition, violence and displacement. Indeed, the old idea that populations would need to lack mobility for the state to dominate them never seemed very convincing. Armies are generally more mobile than general populations (as Moses found before the timely miracle of the Parting of the Red Sea), and for those studying steppe empires it is clear that the rulers can be just as mobile as the ruled. Although many of the 13th century neighbours of the Mongols were ‘nomads’, for example, they found that ‘avoidance’ was no solution to the threat of conquest and incorporation. Populations do not need to be immobile to be conquered.

Most of the work on state formation to date has been concerned with sedentary states, and the associated terminology reflects this bias. However, a number of large and powerful ‘nomadic’ polities (i.e. polities including large numbers of mobile pastoralists) certainly did exist, some very early. Discussion as to the applicability of the term ‘state’ to these societies still goes on (Bondarenko, Korotayev, and Kradin 2003) but in my view, the evidence suggests that many steppe polities were every bit as state-like as most medieval European kingdoms (Sneath 2007).

In any case, Carneiro's theory follows the dominant trend in that it is based upon sedentary examples, and in this respect, his use of the analogy of the pressure-cooker is an interesting choice. Boxed-in by geographical or other factors, he argues, rising population pressure will tend to make the state form faster. However, increasing the pressure actually raises the boiling-point of water, rather than speeding it up. It is the higher temperature that cooks the food more quickly, and since we are interested in the product we think of the process in terms of accelerated boiling. Similarly, my guess is that early state-like political relations existed in many regions and the speed of their development is less important to us than what they produced in terms of material remains that we can study. Resource-rich regions such as Egypt and Mesopotamia had the means of creating large urban complexes and monumental architecture very early, and have been taken as the archetypal state forms as a result. But similarly large and complex political forms that existed in steppe regions would have left far less evidence of their existence. Had the Scythians not left us the kurgan mortuary sites, and been sufficiently connected to the Hellenistic world to appear in Greek texts, there would be very little sign now of their once-powerful empire. But as the archaeology of the steppe improves our appreciation of the scale and complexity of early steppe polities will, I think, continue to build (Di Cosmo 2002; Brosseder and Miller 2011).

Personally, I find Carneiro's perspective is weakened by his inclusion of the concept of ‘chiefdom’. My own inclination is not to try and second-guess Aristotle but to accept the term polis (city-state) for the polity formed by the integration of villages, and not designate some of them ‘chiefdoms’ on the grounds of scale or putative ‘complexity’. Indeed, thinking in terms of evolutionary ‘stages’, rather than historical ages, seems a somewhat zoological approach. The organic analogy has had its day now, and since holistic functionalism has been largely abandoned for the analysis of contemporary societies I think we are bound to be wary of its use to illuminate the past. But this is not really a problem for Carneiro's broader approach since it does not posit any systemic social logic beyond that created by concrete political processes such as conquest. Indeed, his critical perspective on what he terms voluntaristic theories is relatively compatible with recent trends in political anthropology of the state, which we might term ‘power-centric’ approaches (Aretxaga 2003; Hansen and Stepputat 2001).

Whether his latest fine-tuning has emphasized concentration over circumscription or not, to my mind the strength of Carneiro's model continues to be what Earle (1994) described as his ‘hegemonic integrationist’ style of theory.Carneiro reminds us of the centrality of coercion and stratification for the many different projects of rulership we seek to study under the subject of ‘state formation’. This is surely an enduring insight.


1 So, for example, Morgan (1964[1877]: 209) writes: ‘Monarchy is incompatible with gentilism, for the reason that gentile institutions are essentially democratical’.


Aretxaga, B.

2003. Maddening States, Annual Review of Anthropology 32: 393–410.

Bondarenko, D., Korotayev A., and Kradin, N.

2003. Introduction: Social Evolution, Alternatives, and Nomadism. in Kradin, N., Bondarenko, D., and Barfield, T. (eds.), Nomadic Pathways in Social Evolution (pp. 1–10). Moscow: Center for Civilizational and Regional Studies of the Russian Academy of Sciences.

Bhabha, H.

1995. In a Spirit of Calm Violence. In Prakash, G. (ed.), After Colonialism: Imperial Histories and Postcolonial Displacements (pp. 326–343). Princeton, NJ: Princeton University Press.

Brosseder, U., and Miller, B. K.

2011. State of Research and Future Directions of Xiongnu Studies. In Brosseder, U., Miller, B. K. (eds.), Xiongnu Archaeology. Multidisciplinary Perspectives of the First Steppe Empire in Inner Asia (pp. 19–33). Bonn Contributions to Asian Archaeology. Bonn: Bonn University Press.

Clastres, P.

1977. Society Against the State. Oxford: Blackwell.

Di Cosmo, N.

2002. Ancient China and Its Enemies: The Rise of Nomadic Power in East Asia. Cambridge: Cambridge University Press.

Earle, T.

1994. Political Domination and Social Evolution. in Ingold, T. (ed.), Companion Encyclopedia of Anthropology (pp. 940–961). London: Routledge.

Engels, F.

1954[1878]. Anti-Dühring: Herr Eugen Dühring's Revolution in Science. Moscow: Foreign Languages Publishing House.

Hansen, T., and Stepputat, F.

2001. Introduction: States of Imagination. in Hansen, T. B., and Stepputat, F. (eds.), States of Imagination: Ethnographic Explorations of the Postcolonial State (pp. 1–38). Durham: Duke University Press.

Mbembe, A.

2003. Necropoloitics. Public Culture 15(1): 11–40.

Morgan, L. H.

1964[1877]. Ancient Society: Researches in the Lines of Human Progress from Savagery through Barbarism to Civilization. New York: Holt.

Sneath, D.

2007. The Headless State: Aristocratic Orders, Kinship Society, and Misrepresentations of Nomadic Inner Asia. New York: Columbia University Press.

Trigger, B.

1993. The State-Church Reconsidered. InHenderson, J., and Netherly, P. (eds.), Configurations of Power (pp. 74–111). Ithaca: Cornell University Press.