The mismeasure of human history?

David Wengrow responds to Sam Bowles' and Amy Bogaard's presentations at the Stone Centre Symposium on The evolution of wealth inequality.

‍

‍Gap between rich and poor began 7,000 years ago - Capital in the 21st century (BC, that is) - are headlines generated by our colleagues’ research, which applies the tools of Neo-classical economics to the archaeological record, to understand the evolution of inequality.

Turning to their written papers on what Sam and Amy call the ‘farming-inequality nexus’ (Bogaard et al., 2019; Bowles and Fochesato, 2024), I find a project that is more cautious and modest in scope. It deals with wealth inequality in a highly abstract sense, rather than the lived experience of social inequality. The question of how wealth in material goods is translated into power over others is not broached, except in relation to arable land. Questions of class, race, gender, age, caste, or patron-client relations are largely left aside. Nor are questions of value addressed, or the social forms taken by labour, except for a highly generalised definition of slavery. In other words, although it is concerned with material things, this is not a historical materialist approach, in any sense that I am familiar with.

Instead, a new method is proposed, based on the application of econometric tools to archaeological evidence. These provide measurements of a strictly limited kind: the relative sizes of house foundations in ancient settlements, and the number of goods placed with the dead in ancient burial grounds. Or more accurately, those which have survived for archaeologists to inspect, because they were produced in durable materials. Based on such evidence, we are presented with a broad account of the transition from unstable to enduring systems of wealth inequality, which I would paraphrase as follows:

‍

In the beginning, economic inequality was based on labour. In labour-limited economies, growth was limited. So life was relatively egalitarian. Such inequalities as did exist were also limited, and unstable. Then, in several specific locations, durable wealth inequalities came into being. What triggered this was a shift from labour-limited to land-limited economies. This happened in three ways. Firstly, the introduction of the plough increased the productivity of arable land and the scale of farming. Second, this created competition for land, and increased its value. Third, a permanent gap emerged between hereditary landowners and an underclass of ‘renters, sharecroppers, clients, employees or unfree labour,’ now alienated from the prime means of food production.

‍

In outline, this story will be familiar to economists. In fact, it looks back to foundational theories of political economy in the seventeenth and eighteenth centuries, as advanced by men such as John Locke, Anne-Robert Turgot, and Adam Smith, at a time when archaeology barely existed. It is deeply embedded in European philosophy, the social sciences, and the historical experience of land enclosure and privatisation, industrialisation, and empire. For all these reasons, I think we must pay close attention to the methods employed in making such arguments today, bearing in mind that Sam and Amy characterise this work as one of ‘plausible conjecture, rather than tested hypothesis.’

In a paper, co-authored with Mattia Fochesato, the authors have done their best to control for problems with archaeological sample size, and the relative invisibility of those “without wealth” in the archaeological record. However, I think some fundamental questions of method have yet to be addressed, before we can move with confidence from measurements to models. With that in mind, I want to raise several questions around issues of plausibility, and then return to some broader observations on the proposed reconstruction. I have thirteen points, which I will try to keep brief.

1. ‍Grave goods: can the measurement of goods placed with the dead serve as a reliable proxy for hereditary wealth inequality in ancient societies, without attention to contextual factors? To offer a counter-example: in relatively recent times, Indigenous populations of north-western California buried a chief’s most treasured possessions alongside him. They did this as a strategy of social levelling. The purpose was to prevent his descendants from inheriting them, ensuring they would have to make their own fortunes, on a level playing field. If a similar society, say 5,000 years ago, buried its dead this way for generations, we might be tempted to place it near the top of our inequality index, based on its “rich” burials. But we’d be mistaken to do so, because the burials are a social mechanism for suppressing hereditary wealth inequality among the living.‍
2. Architecture (I): can house size really serve as a proxy for wealth inequality, without attention to contextual factors, or what Amy refers to as “governance”? Another counter-example: historically, Iroquoian and Algonquian societies had impressive long houses, with copious storage space. So too, did societies on the Pacific Northwest Coast. If we rely on disparities of house-size, we might be tempted to rank these societies in similar positions on an inequality index. But what would we really learn from this about the evolution of inequality, given that household wealth in Iroquoian societies - including agricultural surplus - was managed by women’s collectives, while on the Northwest Coast, households were internally divided into ranks of male nobles, commoners, and hereditary slaves?‍
3. Slavery: there is no more enduring form of inequality among human beings than hereditary slavery. In the history of the Pacific Northwest Coast, we find such institutions highly developed across non-agricultural societies—hunters, fishers, and gatherers—covering the entire region from Alaska down to Washington State. A detailed statistical analysis was provided by the historian Leland Donald, in his (1997) book Aboriginal Slavery on the Northwest Coast. He concluded that, in economic terms, these truly were slave-based economies, in which perhaps a quarter of the indigenous population lived in bondage – which is roughly equivalent to proportions found in the Roman Empire, or classical Athens, or indeed the cotton plantations of the American South. Which brings me on to:  ‍
4. Labour-limited versus land-limited forms of economy: in their wider reconstruction, Sam, Amy, and colleagues acknowledge the presence of slavery in some hunter-gatherer populations, but they characterise it as ‘rare and episodic.’ Elsewhere, they appear to suggest that slavery intensifies as part of the transition from labour-limited to land-limited forms of economy, associated with the shift to mechanised farming. But on the Pacific Northwest Coast, archaeologists such as the late Ken Ames have plausibly traced a long and stable history for institutions of warfare and slavery, extending back almost four thousand years, in the absence of agriculture. “Growth” in these societies often took the form of raiding one’s neighbours for people, and especially women. Anthropologists have generally agreed that this was not because of any overall shortage of people, but because nobles and commoners refused to perform manual labour, including the seasonal processing of fish harvests, on which their economy depended. Uneven access to prime fishing sites no doubt created conditions comparable to a land-limited economy, but the ultimate cause of slavery—according to Donald and others—was a socially induced shortage of labour. Fernando Santos-Granero has devoted a detailed study to economies of this kind, which also existed in other parts of the Americas, including among non-agricultural populations like the Calusa of Florida Keys and the Guaycuru of the Paraguay palm-savannah. It’s called: Vital Enemies: Slavery, Predation, and the Amerindian Political Economy of Life. He calls these ‘capturing societies,’ and shows how in some cases, such arrangements led to stable, full-blown kingdoms, with enduring distinctions of class and caste. In any general account of the evolution of inequality, clearly they should feature, but it is difficult to see how they could ever be accommodated to the ‘farming-inequality’ nexus.‍
5. Mobile wealth: the assumption that, before the inception of land-limited economies, economic differences were based solely on human labour seems problematic for the history of large parts of the world, including much of Africa. In Nilotic and Bantu populations, cattle provided a form of stable, intergenerational wealth inequality—including the economic basis of large kingdoms or even empires—long before land became scarce or intrinsically valuable, and long before cattle were used for traction. This is also true for some parts of Eurasia, and there is an echo of it in the Latin root of the word ‘capital’: caput, which refers both to chattels and to a ‘head of cattle’. In such societies, wealth was traditionally counted, not by the size of one’s house, but by the size of one’s herd.‍
6. Archaeology and the Gini co-efficient: as originally applied to modern economies in the early twentieth century, the Gini co-efficient assumes comparability between the units compared. These were nation-states, because Corrado Gini (an advocate of Italian Facism and eugenics, who ran the Central Institute of Statistics under Mussolini; see e.g. Gini 1927) considered ‘society’ and ‘the State’ to be largely coterminous, forming a superordinate unit of economic comparison as opposed to, say, international class relations. In his response to Walter Scheidel’s paper, Suresh Naidu referred to this as “methodological nationalism.” Without some such baseline of comparability across institutional forms, the comparative exercise quickly loses meaning. What, then, is the institutional baseline that allows us to compare Gini co-efficients from, say, Neolithic Turkey to the early Roman Republic to modern Britain? This is really a question about:‍
7. The archaeological record: in the pre-circulated paper (Bowles and Fochesato, 2024, fig. 1) we are asked to compare Gini co-efficients ascribed to archaeological sites such as Tell Brak in Syria, Tepe Gawra in northern Iraq, and Eridu in southern Iraq. Each site appears as a data point on a graph, captioned ‘Inequality in material wealth in western Eurasia from the end of the Palaeolithic to the Roman Empire.’ However, as Fochesato and colleagues themselves note, archaeological sites are not equivalent to past populations, or societies, let alone miniature versions of “nations” or “states.” Nor are the individual stratigraphic phases within them. Which raises further questions about:‍
8. Scale: Tepe Gawra is a tiny mound covering just two hectares, and containing architectural remains that span roughly three millennia. Tell Brak, in the fourth millennium BC, spanned an area of 130-hectares: a different order of magnitude. Tepe Gawra was excavated across much of its surface area, while only a very small fraction of Tell Brak has been investigated by archaeologists. Yet both appear on the graph as equally weighted data points, with an assigned Gini co-efficient. Sam and colleagues themselves note that ‘Two societies with equal wealth inequality by our measures may differ substantially in social complexity, political hierarchy, disparities in consumption, economic injustice or other dimensions associated with the term ‘wealth inequality’ (Fochesato et al. 2019). I agree, and admire their candour, but again, it makes me wonder what use we can make of the findings, for example in thinking about the roots of contemporary inequalities.‍
9. Forms of property (I): similar problems arise in relation to a study that has been widely referred to, by Kohler at al., which argues that overall levels of inequality were much higher in Bronze Age Eurasia than in pre-Columbian North America and Mesoamerica. Again, on closer inspection, the findings turn out to be more modest. They are based on a single proxy measure of wealth inequality, in this case, the relative sizes of preserved house-foundations in the archaeological record. The results are still interesting, because of their scope, but perhaps there may be better ways of interpreting them. Which brings me on to: ‍
10. Forms of property (II): in many parts of pre-colonial North America, social influence was not primarily grounded in the stockpiling of wealth or defence of landed estates. Instead, influence was vested in control over forms of intellectual property that comprised specialised knowledge and ritual prerogatives, pertaining to things such as the design of calendrical earthworks, the right to perform a certain dance, administer a certain kind of medicine, use a particular type of image, etc. Ownership of such goods - which may be compared to our patents or copyrights - unlocked rights of usufruct over land and resources (Graeber and Wengrow, 2021, passim). Land and tools might be freely shared, but the initiatory powers to grow maize or pursue game were individually held, inherited, bought, and sold. When arguing that overall differences of wealth are rooted in factors of agrarian production, it seems important to recall that dispossession of First Nations’ lands—not just in the Americas, but in Africa and Oceania—was based on a denial that such alternative forms of property counted as legal title, which European scholars and settlers insisted must be based on territorial sovereignty, and extensive farming. Today, such claims are contested in courts of law by Indigenous communities in America, Canada, Australia, and elsewhere.‍
11. Linear and taxonomic inequality: the work under consideration here treats inequality along a single dimension, which is linear, allowing us to rank societies along a scale. However, inequality in human societies always has at least two basic dimensions, the linear (or “how much”) dimension, and what we might call the taxonomic (or “what kind”) dimension. The latter is closer to the classificatory hierarchies used by biologists. Levels become taxonomically higher as they become more encompassing, extend upwards to a value which is always singular (like ‘life’ in evolutionary biology, or ‘god’ in those medieval ‘great chains of being’). An amusing example of how people play on the distinction is provided by the fourteenth century explorer, Ibn Battuta (after Graeber 2011: 107-8). He tells of a King of Sind, who delighted in receiving gifts from visitors, and then presenting the bearer with something many times their value. A business developed, where local bankers would lend money to visitors to finance gifts, knowing they would be repaid with interest from the proceeds of royal largesse. The king knew this too, but didn’t mind, since the whole point was to show that his wealth exceeded all possible notions of equivalence. If necessary, he could always expropriate the bankers. They, in turn, were aware that the important game here was not the economic one (of linear inequalities) but one of status; and the king’s (taxonomic) status was always absolute. Of course, this also applies to:‍
12. Architecture (II): Let’s assume, for a moment, that house size = wealth inequality, so we can now read this off certain parts of the archaeological record, as if we were navigating a modern homebuyers’ or real estate website (Zillow, Zoopla, Rightmove, etc.). Even then, problems remain, because house size is just one factor guiding value, and usually not the most important, which is location: a two-bedroom house in the suburbs of Birmingham costs a fraction of a similar property in the suburbs of London, which in turn costs a fraction of the same kind of property near the centre of town, or close to a good high street, school, or other ameneties that people today value. This kind of consideration may seem very obvious, but it plays no role in current applications of the Gini co-efficient to archaeology, which are not in any way integrated with or adjusted for landscape studies, of which many exist. This, in turn, would require engagement with questions of taxonomic value: what constitutes a ‘centre’ (e.g. a palace, temple, calendrical earthwork, vacant plaza?), or a ‘periphery’? Instead, what studies of the kind we are discussing here do, for analytical purposes, is to strip away the taxonomic dimension entirely, in the hope of deciphering some deeper pattern along the linear one. Do they succeed?‍
13. How to evaluate the claims: perhaps one good measure is if the findings depart in some meaningful way from the purely speculative, and ideologically freighted notions of economic growth advanced by political economy’s founding fathers. Against this criterion, it is surely a step forward that the new research confirms a lack of correlation between the adoption of agriculture and the rise of wealth inequality; but it does so, I fear, only then to take two steps back, by attributing the inception of enduring wealth inequality to the adoption of the plough, or other technologies of agrarian extensification. ‍
14. Inequality and the plough: on this point, the authors evoke the work of Andrew Sherratt, who drew on Jack Goody, who drew on V. Gordon Childe. But Goody (e.g. 1971; 1976) also argued that the histories of Europe and Asia had far more in common than Childe thought, because their societies shared bilateral systems of property inheritance that allowed arable land to be held within status groups, defined by elaborate codes of class or caste. He thought this made them different from African societies south of the Sahara, except for Ethiopia. A key factor in creating such differences, Goody argued, was the early adoption of animals for plough-based agriculture in Eurasia. Another was the growth of literate bureaucracy. In most of Africa, by contrast, inequality was about “wealth in people” rather than in land. It’s intriguing to see aspects of Goody’s model revived by Sam, Amy, and colleagues but I think it’s important to recall that this was only somewhat less than half of his original model. The other major part was about African state formation, including systems of enduring wealth inequality in labour-limited economies. In Africa, Goody saw this as being based less on access to the means of agricultural production than the capture and ownership of slaves, control over foreign trade, and a monopoly on what he called the ‘means of destruction,’ especially horses and guns. Of course, such forms of domination were also typical of Eurasian states, which developed them to much greater heights, eventually turning abundance into scarcity. This afternoon, Walter Scheidel will be arguing—somewhat counter-intuitively—that in the broad sweep of human history, ‘Violence has been the most important means of levelling wealth and income inequality in human history.’ In debating that point, I think we’d do well to reinstate the other half of Goody’s model, because it reminds us that organised violence, plunder, and predation—most systematically in the form of empire—was ultimately the most universal means of creating such overarching inequalities, in the first place.

DW. London, 29^th June 2024.

‍

References

• Ames, K. 2001. ‘Slaves, chiefs, and labour on the northern Northwest Coast.’ World Archaeology 33(1): 1-17.
• Bogaard, A. Fochesato, M. and S. Bowles. 2019. ‘The farming-inequality nexus: new insights from ancient Western Eurasia.’ Antiquity 93(371): 1129-43.
• Bowles, S. and M. Fochesato. Forthcoming, 2024. ‘The origins of enduring economic inequality.’ Journal of Economic Literature.‍
• Donald, L. 1997. Aboriginal Slavery on the Northwest Coast of North America. Berkeley: University of California Press.
• Fochesato, M., A. Bogaard and S. Bowles. 2019. ‘Measuring ancient inequality: The challenges of comparability, bias, and precision.’ Antiquity 93(370): 853-69.
• Gini, C. 1927. ‘The scientific basis of fascism,’ Political Science Quarterly 42(1): 99-115.
• Goody, J. 1971. Technology, Tradition and the State in Africa. Oxford: Oxford University Press.
• Goody, J. 1976. Production and Reproduction: a Comparative Study of the Domestic Domain. Cambridge: Cambridge University Press.
• Graeber, D. 2011. Debt: The First 5,000 Years. New York: Melville House.
• Graeber, D. and D. Wengrow. 2021. The Dawn of Everything: A New History of Humanity. London: Penguin.
• Kohler, T.A., M. E. Smith, A. Bogaard, G. M. Feinman, C. E. Peterson, A. Betzenhauser, M. Pailes, E. C. Stone, A. M. Prentiss, T. J. Dennehy, L.J. Ellyson, L. M. Nicholas, R. K. Faulseit, A. Styring, J. Whitlam, M. Fochesato, T. A.Foor, and S. Bowles. 2017. ‘Greater post-Neolithic wealth disparities in Eurasia than in North America and Mesoamerica.’ Nature 551 (7682): 619-622.
• Santos-Granero, Fernando. 2009. Vital Enemies: Slavery, Predation, and the Amerindian Political Economy of Life. Austin: University of Texas Press.
• Scheidel, W. 2018. The Great Leveler: Violence and the History of Inequality from the Stone Age to the Twenty-First Century.Princeton, NJ: Princeton University Press.
• Sherratt, A. 1997. Economy and Society in Prehistoric Europe:Changing Perspectives. Edinburgh: Edinburgh University Press.
• Wengrow, D. and D. Graeber. 2018. ‘Many seasons ago: slavery and its rejection among foragers on the Pacific Coast of North America.’ American Anthropologist 120 (2): 237–49.