Book Reviews

The Species above Constraints: A Review of The Dawn of Everything – A New History of Humanity by David Graeber and David Wengrow

DoE Cover

By Markar Melkonian

The Dawn of Everything Cover

It was with high hopes that I opened this big book, despite its too-ambitious title. David Graeber, after all, was the author of the wonderful book, Debt: The First 5000 Years. I am neither an anthropologist nor an archaeologist, and I came toThe Dawn of Everything with many questions. I wondered if there might be new evidence of resistance to class domination during the millennia separating the Upper Palaeolithic from V. Gordon Childe’s Urban Revolution; I wanted to know what had happened during the long centuries between the adoption of domesticated grains and livestock, on the one hand, and the large-scale ‘domestication’ of humans by humans, on the other, and I wanted to be convinced that there had existed, here or there in the distant past or more recently, large ‘complex’ human settlements without ruling classes and the repressive institutions of states. According to the book’s subtitle, after all,The Dawn was supposed to be anew history of humanity.


It got off to a shaky start on page one, with an epigraph from the Woo-Woo Meister himself, C.G. Jung. But I pressed on, and, in the pages that followed, Graeber and Wengrow (henceforth ‘the authors’) would indeed deliver on some of my hopes. The Dawn of Everything is a long rejoinder to the whiggish prehistories inscribed in high-school textbooks and academic journals alike; it is a challenge to the technological determinism that saturates our conversations, and it is an original alternative to the grand narratives that some (but not all) proponents of evolutionary psychology have been telling for decades. Not all the contentions are new, of course – and, in any case, new is not always true – but the range of scholarship is astonishing, and it is a whirlwind of a read.

There are many reviews of The Dawn of Everything accessible online,[1] and there is no need to repeat what has already been said, pro and con. This discussion will focus on two formulations that deserve special scrutiny, because they have to do with the book’s emphasis on collective decision-making as a hallmark of humanity, and they are central to the authors’ stated mission. They are: (i) the authors’ advice to abandon mode of production as an analytical concept, and (ii) their definition of the state in terms of three ‘principles of domination’. These formulations, I will suggest, as well as the prominent role that the authors ascribe to collective decision-making over the long sweep of prehistory, elevateHomo sapiens, in a familiar but dubious way, above all other terrestrial species.

But, first, a note about the scope of the New History of Humanity. As we will see, this has a bearing on the authors’ philosophical anthropology, and thus on much of their subsequent discussion.

I. Off to a Late Start

The authors begin with a commitment to optimism of the intellect: ‘This book’, they write, ‘is an attempt to begin to tell another, more hopeful and more interesting story; one which, at the same time, takes better account of what the last few decades of research have taught us.’ (p. 3.) Later, they write that, ‘Since this book is mainly about freedom, it seems appropriate to set the dial a bit further to the left than usual and to explore the possibility that human beings have more collective say over their own destiny than we ordinarily assume’ (p. 206).

The authors deny that humans have ever permanently lived in small hunter-gatherer bands.[2] Some reviewers have been surprised that self-identified anarchists would hold such a view, and others have reported this claim uncritically, as a new insight.[3] The claim is less surprising, though, in view of the authors’ contention that Upper Palaeolithic humans should be conceived not as isolated nomads, but as groups interacting over large geographical areas, along the lines of aboriginal Australians and North American peoples at the time of contact with Europeans (pp. 122–3). The argument relies on an analogy with what ethnographers have described as ‘culture areas’. In Chapter 4, the authors argue that the shift away from large-scale culture areas to smaller and higher-density societies took place in the Neolithic, when ‘most people live their lives on an ever-smaller scale as populations get larger’ (p. 121).

It is not true, of course, that Homo sapiens have never permanently lived in small hunter-gatherer bands; however, as we will see, our species has not long been ‘human’, in the sense of the word that the authors require.(The authors also play down the suggestion that humans have much in common with non-Homo sapiens hominids – much less with‘simians’, ‘apes’ and ‘monkeys’, as our evolution-averse authors put it, with what looks like deliberate artlessness.) It would be helpful, then, to gain a clearer view of the historical scope of The Dawn, by asking whenroughly, always roughlyHomo sapiens became ‘human’. And that question hinges on what counts as human.


At times, the authors claim that their new history applies to the last 30–40,000 years of the career of our species,[4] and so this is where they take up their history of humanity. In a brief discussion of our Palaeolithic ancestors in Chapter 3, they note that (‘anatomically modern’?) humans coexisted for millennia with other populations of the genus Homo; however,‘only after those other populations became extinct can we really begin talking about a single, human “us” inhabiting the planet’ (p. 82). This, presumably, was when we had become definitively human and ‘started doing human things’ (p. 82).[5]

It is never entirely clear how it came to pass that we started doing human things. One hypothesis, which Yuval Harari highlights in his bestseller Sapiens (2011), is that our ‘modern brain’ is the result of a relatively abrupt change thatoccurred sometime between 70,000 to 30,000 years ago, perhaps as the result of a genetic mutation. This is the Human (or Cognitive) Revolution hypothesis. According to the hypothesis, this transformed brain accounts for the appearance of handier tools, elaborate burials, and complex symbolic behaviour, and it set our ancestors off to tell stories, to talk about things they had never seen before, to imagine lion-men, and the like.

The Human Revolution hypothesis is supposed to resolve what has come to be called the ‘sapient paradox’, by answering the question: ‘why do so many tens of thousands of years stand between the biological origins of humanity and the widespread appearance of typically human forms of behavior; between when we became capable of creating culture and when we finally got round to doing it?’ (p. 84). Graeber and Wengrow explicitly reject the Human Revolution hypothesis, as well as the sapient paradox. But they also reject the suggestion ‘that for countless millennia we had modern brains, but for some reason decided to live like monkeys anyway; or that we had the ability to overcome our simian instincts and organise ourselves in an endless variety of ways, but for some equally obscure reason only ever chose one way to organise ourselves’ (p. 84). The hundreds of centuries that separate the ticking of our genetic clock from the ticking of our cultural clock, they argue, is an illusion born of the skewed and scanty character of our archaeological record so far (pp. 835).

Thus, the authors seem to be saying that we started doing human things – presumably, including the invention and enactment of diverse social arrangements – tens of thousands of years before the advent of the Upper Palaeolithic (say, 50,000 years ago), and before the extinction of non-sapiens populations of the genusHomo. Leaving it to our archaeologist colleagues to judge the plausibility of that scenario, let us only note that the discussion (in Chapter 3) of human societies during the last glaciation period mostly precedes Graeber and Wengrow’s new history of humanity.

For the most part, the authors focus on a broad variety of cases from the last twelve millennia. The earliest human groups the authors discuss at any length, namely Upper Palaeolithic mammoth hunters, appeared 12–10,000 years ago, and most of the cases of seasonal or permanent settlements that the authors discuss existed at most ten millennia ago. The authors, then, have left out at least 188,000 years of the career of our species,[6] but without thereby denying the possibility that humans were collectively debating social arrangements, and deliberately refashioning them even before the Upper Palaeolithic (pp. 83–4). Even if we accept the 40,000 years claim, howeverand even if we take the authors’ examples of neolithic settlements to be the rule and not the exception (as seems unlikely, in view of the still-accumulating evidence to the contrary)[7]then Graeber and Wengrow’s claim would amount to conceding that ‘anatomically modern’ humans spent at least 80% of their career in hunter-gatherer bands. This is a far cry from the claim that ‘humans’ did not, in fact, spend most of their history in tiny bands. Moreover, the fact that this or that Upper Palaeolithic or early Neolithic settlement might have been in place in Çatalhöyük, or the Trippillya megasites north of the Black Sea, or elsewhere, does not controvert the claim that the vast majority of human groups might well have continued living in hunter-gatherer groups for millennia thereafter.

Even if it could be said that we have little archaeological evidence so far in favour of the claim that humans have been organised in hunter-gatherer bands for the vast majority of their career, there is also little evidence against it.[8] Moreover, there are strong evidence-based arguments from ethnology and the study of non-sapiens species in favour of this claim. Graeber and Wengrow’s confident pronouncements notwithstanding, then, the claim that our species, like other hominids, spent most of its existence in hunter-gather groups is not open to much dispute these days.

‘The only thing we can reasonably infer about social organisation among our earliest ancestors’, the authors write, ‘is that it’s likely to have been extraordinarily diverse’ (p. 82). Indeed, further evidence might well indicate that social organisation among the earliest Homo sapiens was even more diverse than we believe at present, and perhaps in more ways than we imagine today. But the diversity of social organisation before the advent of the Holocene is, so far at least, consistent with the consensus among anthropologists that the mode of subsistence of our earliest ancestors was hunting and gathering. Moreover, to anticipate an implication in the next section below, the diversity occurred within the wide range of social relations of the great variety of ‘primitive’ communal societies.

So why have the authors advanced such a sweeping and unsupportable claim? The answer, perhaps, has to do with their philosophical anthropology, their starting assumptions about ‘what makes us human in the first place’.


The authors discuss an assortment of societies, culled from across the continents and the millennia, that they describe as exceptions to ‘the conventional story’ of human prehistory that we are all supposed to have learned in school, and these cases are taken to show that there are no natural laws that impose marching orders on the ways we have arranged our collective lives. Throughout The Dawn, the authors emphasise collective debate and decision-making, as well as theatricality and make-believe, as in such formulations as‘cities begin in the mind’ (p. 276), and ‘to farm or not to farm: it’s all in your head’ (p. 242). ‘Human beings’, they write, ‘may be (indeed, we’ve argued they are) fundamentally imaginative creatures …’ (p. 121). At the same time, the authors ‘set the dial a bit further to the left’ by downplaying ‘given circumstances transmitted from the past’,[9] notably ecological constraints and the non-intentional, impersonal forces at work in the prehistory and early history of our species (p. 118). The impersonal forces include the constraints that the level of productivity of labour imposes on social relations of production, as well as class conflict, and (as we will see in section III below) state power.

What emerges in The Dawn is a story of the prehistory of humanity (or perhaps just the last 50 to 30 millennia, after humans presumably had become ‘really human’) as a consciously debating species with the capacity to imagine –and to enact – new social arrangements. This is a mind-over-matter conception of our prehistory and early history, which begins with certain mental faculties of folk psychology, notably the imagination and volition, and then imbues these faculties with causal efficacy.When it comes to the hoary choice between freedom or solidarity, the authors write that, ‘what really makes us human in the first place […] is our capacity – as moral and social beings – to negotiate between such alternatives’ (p. 118). To exercise this capacity is to practise ‘the freedom that makes us human in the first place’ (p. 8).[10]

This, then, is what counts as human for Graeber and Wengrow, and it is all too redolent of the idea that free will sets humans above nature. It is an idea closer in time to the first chapters of the book of Genesis than to the myths of Hobbes or Rousseau, and, when connected to the authors’ hypothesis of a post-neolithic failure of our original imagination, it is closer in theme to the expulsion from Eden.[11]

II. The Problem with the Problem with Modes of Production

To make room for the freedom that makes us human in the first place, the authors want to haul away the ‘conventional story’ of human history, which they identify with a theory of stages in history. In a theory of stages, as the authors would have us understand it, the sequence of development mounts up in discrete steps, ineluctably and irreversibly, from lower to higher, according to a criterion such as moral or intellectual progress, efficiency, or complexity. Adam Smith is said to have presented such a theory when he divided human history into the Age of Hunters, the Age of Shepherds, the Age of Agriculture, and the Age of Commerce; and, similarly, in the case of the American anthropologist Lewis Henry Morgan, who identified the three stages of savagery, barbarism, and civilisation. It is not entirely clear that Morgan (or Smith, for that matter) held that these stages were ineluctable and irreversible; Morgan presented them as empirical generalisations drawn from ethnological observations.

Graeber and Wengrow rally evidence from a wide range of sources, to challenge the idea of an irreversible progression of stages.[12] The authors’ objection, however, is not limited to theories of stages, but extends to all depictions of humans as a species caught in an ‘evolutionary straitjacket, their place in history defined by their mode of subsistence, and their role blindly to enact some abstract law of development which we understand but they do not’ (p. 96). Here, we encounter pernicious ‘evolutionism’ (pp. 446 ff.). The authors include Marx’s concept of mode of production[13] among the versions of evolutionism that ‘were basically unworkable, and eventually had to be thrown out’ (p. 446).

As we know, Marx and Engels posited at least four modes of production – ‘primitive’ communalism, slavery, feudalism, and capitalism – plus one hoped-for possibility, the ‘higher stage’ of communism. The first four modes are presented as empirical generalisations, not as a marching order on history.[14] The ‘higher stage’ of communism, by contrast, is a programmatic goal, an aspiration based on what the authors of the Communist Manifesto argued was achievable, thanks to what capitalism has wrought.

In the authors’ view, mode of production is an unhelpful category because it indiscriminately lumps together very different ‘culture areas’. The point can be illustrated by considering two cases that the authors discuss separately: the culture area comprised of fourth-millennium BCE ‘low-density urban settlements’ north of the Black Sea, like Taljanky and Nebelivka, seems to fall into the same ‘primitive’ communal category as ninth-millennium BCE Mesolithic hunter-foragers of the Fertile Crescent.[15] In view of how different these two culture areas are in time, place, and character, however, it might seem pointless to toss them into the same categorial pot, as instances of the same mode of production.

But pointless for what purpose? The concept of mode of production emphasises close causal connections between social relations and productive forces, and it highlights the constraints on social organisation that these interactions pose. This contravenes the authors’ disposition, in the service of ‘freedom’, to disconnect production, distribution, and exchange of the means of subsistence, on the one hand, from considerations of property, custom, lineage, law, language, the gods, and philosophy, on the other hand.

In Chapter 5, Graeber and Wengrow suggest that one reason why humans have got stuck in certain social arrangements to the exclusion of others has to do with why we expend so much effort trying to demonstrate that we are different from our neighbours (p. 166). They illustrate the point by describing the many ways that the linguistically diverse native gatherers of acorns and pine nuts in California distinguished themselves from their ‘complex forager’ neighbours of the Northwest Coast. The striking differences between these two ‘culture areas’ include, of course, the institution of slavery in the Northwest but hardly at all in California. The areas also differed dramatically when it came to the design of lodging, canoes, masks, baskets, clothing, and other artefacts, and to the respective values placed on work and displays of status (p. 166). ‘Environmental determinists’ have explained the differences between these neighbours by emphasising the greater ecological efficiency, in both California and the Northwest coast, of gathering nuts and fishing, respectively, compared to cultivating maize, which prevailed elsewhere in North America. Graeber and Wengrow, by contrast, suggest that these distinctions are largely the result of conscious collective choices that swing clear, pretty much, from considerations of production of the means of subsistence. Proximity to rich sources of salmon, or the profusion of oak and buckeye groves might have defined social production in the respective regions; nevertheless, decisions taken and enacted collectively fashioned their social life. ‘Framed in this way’, the authors write,

the question of how ‘culture areas’ formed is necessarily a political one. It raises the possibility that decisions such as whether or not to adopt agriculture weren’t just calculations of caloric advantage or matters of random cultural taste, but also reflected questions about values, about what humans really are (and consider themselves to be), and how they should properly relate to one another. (p. 175.)

Thanks to ‘political’ decisions collectively taken and enacted, the abstemious and assiduous Californians were stuck in their social arrangements, while the neighbouring Kwakiutl and Tlingit were stuck in their salmon-fishing economy, class divisions, moieties, potlatches, and enslavement. This process of self-differentiation, or schismogenesis, in neighbouring culture areas exemplifies the general circumstance, for which the authors provide further evidence, that: ‘ever since Mesolithic times [9000–4300 BCE], the broad tendency has been for human beings to further subdivide, coming up with endless new ways to distinguish themselves from their neighbors’ (p. 166). At this point, the authors come closest to answering their big question: schismogenesis, they claim, is ‘crucial to understanding how human freedoms, once taken for granted, eventually came to be lost’ (p. 167).

This hypothesis raises more questions than it answers, but let us leave that aside for the moment. The authors argue that, taking these considerations into account, ‘the idea of classifying human societies by ‘modes of production’ looks decidedly naïve’ (pp. 1889). Moreover, they argue, the classification is wrongheaded because it does not acknowledge that various combinations of modes of subsistence and forms of social organisation, including a wide spectrum of authoritarian and egalitarian social arrangements, have existed within a broad range of relations of production. Slavery, for example, existed among hunters and fishers, horticulturalists, semi-settled groups, and neolithic agriculturalists. Foragers have practised flood-retreat farming; multiple groups of hunters and gatherers have come together seasonally in large settlements featuring monumental stone architecture, and part-time hunters, fishers, and horticulturalists have inhabited large settlements, apparently without class divisions.

However, as soon as we distinguish between mode of production, on the one hand, andmode of subsistence, on the other, it becomes clear that this objection misses its target. A mode of subsistence, such as foraging, pastoralism, fishing, field cultivation, or industry, is a set of skills, practices, and technologies by which members of a society acquire and distribute food.[16] A mode of production, by contrast, emphasises the causal relationships among labour power, means of production, and relations of production, notably ownership and control of the means of production. There is no reason why modes of production cannot alternate seasonally, within this or that social formation,[17] nor that given social formations cannot fragment into multiple bands or small groups periodically, and then regularly reconstitute themselves into a single much larger social formation. If other societies move in or out of faming, or engage simultaneously in horticulture, ‘play-farming’, and foraging, then they could be described as social formations within which two or more modes of production coexist or alternate.[18]

Nor is there any obvious reason why interactions on a ‘continental’ scale could not have taken place in the Neolithic, as they do today among societies with different modes of subsistence and dominant modes of production. Moreover, if different modes of production can coexist within the same social formation, then it should not be surprising that they can coexist within the same culture area. Yes, recent research supports the authors’ view that, in the Upper Palaeolithic, humans did not live ‘exclusively in tiny egalitarian bands’. But this does not entail the empirically implausible claim that the communal mode of production did not dominate human communities across ‘culture areas’ at that time, as it had long before that.

This approach emphasises the causal connections between shifting and coexisting modes of subsistence (or productive forces), on the one hand, and shifting and coexisting social relations, on the other. The concept of mode of production, within its proper theoretical framework, can help us describe in a more nuanced way, for example, a group in the Amazon that switches back and forth from slash-and-burn farming to hunting, and that also engages in commodity production and trade.

The relations of production in a particular social formation – the particular relations among people and other things – arise in the process of social production, exchange, and distribution of the means of subsistence, and they obtain whether or not people acknowledge them. These causal relations, however, are precisely what Graeber and Wengrow wish to soft-pedal, in the course of emphasising conscious collective debate and the freedom to imagine and enact new social arrangements. The authors’ rejection of the concept of mode of production, then, is consistent with their emphasis on the malleability of social relations, and the assumption that social relations can be changed as a result of conscious debate and collective choice, even without a transformation of modes of subsistence or the productive forces.

The authors acknowledge that, ‘Perhaps Marx put it best: we make our own history, but not under conditions of our own choosing’ (p. 206). However, they complain that ‘Social science has been largely a study of the ways in which human beings are not free: the way that our actions and understandings might be said to be determined by forces outside our control’ (p. 498). By dismissing mode of production, the authors downplay involuntary, non-representational conditions not-of-our-own-choosing. This serves their hortatory purposes of promoting the basic freedom to collectively refashion social arrangements, but it compromises our ability to discern patterns in history, to describe the causes of social phenomena, and to locate ‘humanity’ within the larger story of one among multiple species of the genus Homo.

III. An Obscure State

Chapter 10, titled ‘Why the State Has No Origin’, is the longest chapter in the book and it is in more than one respect the culmination of arguments presented in earlier chapters. It begins with the claim that ‘the quest for the “origins of the state” is […] a fool’s errand’ (p. 359).[19] Experts agree that Pharaonic Egypt, Shang China, and Tawantinsuyu were all states in some sense of the word, but the authors write that, ‘… with no consensus among social theorists about what a state actually is, the problem is how to come up with a definition that includes all these cases but isn’t so broad as to be absolutely meaningless’ (p. 359). This, they add, ‘has proved surprisingly hard to do’.

According to the authors, influential contemporary conceptions of the state are either too inclusive, or not inclusive enough, or they are circular. For example, the authors claim that the canonical ‘early states’ of early Dynastic Mesopotamia, Old Kingdom Egypt, the Inca Empire, the states of the ‘Classic’ Maya lowlands, and the late Shang Dynasty of China evade easy grouping together, because among them only the kingdom of Egypt and the Inca empire conform to Max Weber’s conception of the state as holding a monopoly over legitimate violence within a geographical area (pp. 40913). In Early Dynastic Mesopotamia, by contrast, the principle of sovereignty over territory was not usually operative, since each of the many city-states there was governed by a charismatic warrior-king, all of whom were vying constantly for dominance over territory.

Graeber and Wengrow recognise that the canonical early states under consideration all deployed violence, and ‘in every case the apparatus of government stood on top of some kind of division of society into classes’ (p. 410). What all these states had in common with non-canonical states was that in each one of them, in different ways, nonproducers systematically extracted the social surplus from those who produced it. The authors acknowledge this; for all of their misgivings about a Marxist theory of the state, they offer the following definition:

Ruling classes are simply those who have organised society in such a way that they can extract the lion’s share of that surplus[20] for themselves, whether through tribute, slavery, feudal dues or manipulating ostensibly free-market arrangements. (p. 128.)

Instead of exploring that generalisation, however, the authors increase the difficulty of coming up with a sufficiently inclusive, noncircular theory of a state by jerry-building a conception that makes no mention of ruling classes or the appropriation of the social surplus: the state, they say, is ‘… not the result of a long evolutionary process that began in the Bronze Age, but rather a confluence of three political forms – sovereignty, administration, and charismatic competition – that have different origins’ (p. 431).

These three political forms of domination, however, do not all appear together in any of the ‘early states’. For example, in the Maya Lowlands, we learn, the rulers (the ajaws) were ‘effectively, like tiny squabbling gods’ (p. 410) who, according to the authors, personified the principle not of competitive politics or sovereignty, but of bureaucracy.[21] In Shang China, by contrast, the principle of bureaucracy was prominent, as was the principle of charismatic leadership, though to a lesser extent. Since the rulers did not claim sovereignty over extensive areas, the authors describe Shang China as a ‘second-order regime’ (pp. 41113), in which only two of the three basic principles of domination (in this case, bureaucracy and charismatic politics, but not sovereignty) prevailed.

Citing another example, the authors write that,

… the rulers of ancient Mesopotamian city-states made no direct claims to sovereignty, which for them resided properly in heaven. When they engaged in wars over land or irrigation systems, it was only as secondary agents of the gods. (p. 507.)

Since these city-states combined charismatic authority with a highly developed administrative authority – but not sovereignty – they qualify as second-order states. Why not sovereignty? Because, when the rulers engaged in violence, they represented themselves as merely secondary agents of the gods, and ‘made no direct claims to sovereignty’.

Following the same line of reasoning, successive administrations in Washington D.C. may also be said to lack sovereignty, because they have engaged in their many far-flung wars and bombing romps only as secondary agents of Democracy, Freedom, Human Rights, and International Law. When it comes to modern states, however, the authors do not deny the principle of sovereignty: modern states, they write, ‘are simply one way in which the three principles of domination happened to come together, but this time with a notion that the power of kings is held by an entity called “the people”’ (p. 431). What distinguishes a modern state, then, is that all three principles of domination operate, and sovereignty is typically represented as residing in ‘the people’.

At this point, we might remind ourselves that, in the modern state, ‘the people’ – whatever that term may be taken to mean – does not actually have sovereignty, outside of the realm of representation and rhetoric.[22] Popular sovereignty, then, does not appear to differ, either as a fiction or in its legitimating function, from, say, the notion of the divine right of kings in pre-modern states.

After thus jerry-building a conception of the state that makes it possible to have monarchs, aristocracies, slavery, and extreme forms of patriarchal domination even without a state, the authors ask:

If it is possible to have monarchs, aristocracies, slavery and extreme forms of patriarchal domination, even without a state (as it evidently was); and if it’s equally possible to maintain complex irrigation systems or develop science and abstract philosophy without a state (as it also appears to be), then what do we actually learn about human history by establishing that one political entity is what we would like to describe as a ‘state’ and another isn’t? (p. 361).

What applies to the sovereignty of states also applies to empires. In Chapter 10, for example, we learn that, if Chavín de Huántar counts as an empire at all, then it was one ‘built on images linked to esoteric knowledge’, whereas the Olmec was ‘an “empire” built on spectacle, competition and the personal attributes of political leaders’ (p. 390).

States built on the personal attributes and self-descriptions of their rulers; empires built on images and spectacle; fictive representations as definitive of state power – these formulations are consequences of rejecting ‘evolutionist’ or ‘environmental determinist’ approaches to the study of history and society. In view of the many aforementioned problems with this approach, it is curious that the authors do not revisit their reasons for rejecting a Marxist theory of a state. Let us review three of their most serious objections.

(1) The authors assert that recent and historically documented hunter-gatherers ‘present an enormous range of possibilities, from assertively egalitarian groups […] to assertively hierarchical ones’ (p. 539, n. 7). This, they hint, should give pause to the Rousseaueans among us.[23] If hierarchiesnotably hierarchies of knowledge and administration, charisma, and violencehave asserted themselves even within hunter-gatherer groups,[24] and if the principles of domination that these hierarchies embody come close to defining (first-order, second-order, or modern) states, then it is easy to describe the search for the origin of states as a fool’s errand.

However, evidence of differential status or of authority, prestige, or leadership, is not itself evidence of class stratification, class rule, or a state. A distinction is called for here: when an Mbuti hunting party designates a temporary leader by consensus, a hierarchy has been established; however, it is not a dominance hierarchy, enforced by coercion and threat. Rather, it is what we might call aconsensual hierarchy. Furthermore, the presence of a proper dominance hierarchy is not a sign of class rule or state power either, as long as systematic extraction of the social surplus is not taking place. There is no state without the presence of a ruling class, which, as Graeber and Wengrow put it,‘can extract the lion’s share of that surplus for themselves’ (p. 128).[25] If this sort of extraction took place systematically in the case of the Olmec, the Maya, the Natchez, the Shilluk, or any other social formation, then a state was in placeand this is the case whether or not we wish to describe the particular forms of domination in terms of sovereignty, bureaucracy, or charisma, and also in cases of seasonal settlements and among nomadic groups, too.[26]

(2) The authors reject theories of the origin of the state because such theories posit ‘the emergence, in embryonic form, of a new and unprecedented institution that would grow and evolve into modern forms of government’ (p. 413). Presumably, such origin stories are teleological because they describe the modern state as the culmination of features that have been projected backwards in time, across the millennia. But the point of Engels’s discussion of early states in The Origin of the Family, Private Property, and the State was that some of their salient features – notably: a dramatic subordination of the roles and status of women, organised violence within the population, systematic expropriation of direct producers, and class rule – arecontinuous with states of affairs in the present. This suggestion was a scandal among Whiggish historians 140 years ago, and evidently it remains so today.

Graeber and Wengrow write: ‘Now it is undoubtedly true that, over the broad sweep of history, we find ever larger and more settled populations, ever more powerful forces of production, ever larger material surpluses, and people spending ever more of their time under someone else’s command’ (p. 133). Historical materialists do not need to advance any claims more ‘teleological’ than these broad empirical generalisations.

(3) The authors mention a further objection to the Marxist definition of the state, namely that it is ‘unpalatable to liberals, ruling out any possibility that the state could ever become a benevolent institution’ (p. 360). A Marxist might reply that a state, any state, is always a ‘benevolent institution’ – benevolent, that is, to a certain part of a population, sometimes a very small part, that extracts the lion’s share of the social surplus. In any case, Marxists do not dismiss a powerful theory just because it is unpalatable to liberals.


The authors do not provide much of an answer to the big question they pose at the beginning of their book and to which they return at the end of it: ‘how did we find ourselves stuck in just one form of social reality, and how did relations based ultimately on violence and domination come to be normalised within it?’ (p. 519). Their most sustained attempt to answer that question is their schismogenesis-and-culture-areas explanation in Chapter 5. In section II above, it was suggested that this explanation is inadequate because it is inconclusive and difficult to generalise across a wide range of cases. On the other hand, the authors’ default explanation, to which they return in their Conclusion, seems to amount to the circular claim that we got stuck because our imaginations failed us.

But even when we exercise our imaginations to reinvent social relations, there is the problem of enacting our visions. Attempts to enact alternative forms of social organisation often fail, and in short order. Graeber and Wengrow remind us that ‘collapses’, ‘Dark Ages’ and interregna can sometimes be occasions of reinvention. But they may also be reversions to the old order.

It is strange that, in a big book that poses the authors’ ‘big question’ thirty years after the final demise of the most massive socialist experiment in history, the authors make no mention of the Soviet Union – neither as a bright but fleeting illo tempore, nor as a false dawn or a cautionary tale. Of course, one cannot say everything, even in a six-hundred-page book, and perhaps the authors judged the story of the rise and fall of the Soviet Union to be at cross-purposes with their hortatory aims: after all, the Soviet Union, a ‘failed experiment’, does not offer much in the way of inspiration for rebels today.

Even so, the October Revolution and its immediate aftermath well illustrate the authors’ point that humans are capable of choosing alternative social arrangements, on a trans-continental scale. If Tripollye, Teotihuacan or Ubaid Uruk were sites of ‘bold social experiments, resembling a carnival parade of political forms’ (p. 4), then surely Petrograd and Moscow were, too, at least in the years immediately following the October Revolution. Under Lenin’s slogan All power to the soviets, hundreds of sites of self-organised ‘neighborhood councils and popular assemblies’ (p. 517) embodied, however briefly, ‘self-conscious social experimentation’ (p. 326) and ‘the freedom to create new and different forms of social reality’ (p. 525): the repressive apparatus of the Czarist order was smashed; Lenin insisted on the right of nations to self-determination, including the right to independent statehood; Bolsheviks like Alexandra Kollontai pushed for legislation that would undermine the patriarchal nuclear family, and male homosexuality was decriminalised in 1922.[27] Cities were redesigned and others were built from the ground up, incorporating bold experiments in urban design, architecture, public housing, childcare, education, transportation, the plastic arts, music, film, and scientific research.

But carnival parades do not last long. Coming to a collective decision to rearrange social relations is a long, long way from following up on that decision and sustaining it from one generation to the next – especially in the face of foreign invasion, civil war, economic embargo, military encirclement, a growing internal class opposition, massive destruction by the Wehrmacht, and an unwinnable arms race. Long before the seventy-fifth anniversary of the October Revolution, stuckness had returned with a vengeance. And, ever since the end of the Soviet interregnum, imagination, by any common usage of the word, has by all appearances been extirpated, along with‘that freedom to imagine and enact other forms of social reality’.


A few words in closing about the problem-of-scale argument, which concludes that any large and complex society necessarily requires a state (pp. 3601). This view prevails, of course, among anthropologists; however, Marx and Engels argued that a future classless and statelesssocial order is achievable, precisely because of the technological, organisational, and democratic achievements of bourgeois society, the rising productivity of labour, the completion of the global market, the revolution in ‘communication and transport’ and a hundred other transformations that the authors of theCommunist Manifesto foresaw.

In Chapter 10 of The Dawn and in the Conclusion, the authors, too,take pains to refute the problem-of-scale argument. If they are right that large urban societies have existed for centuries with neither class stratification nor repressive institutions of class rule,[28] then they have provided precedentsprehistoric, ancient, and more recentthat defy this objection to a communist future. The theoretical question remains whether a generation yet unborn will ever live in large, urbanised, and highly productive societies that are no longer dominated by market relations and repressive institutions of ruling-class power. But, on this point, among others, Graeber and Wengrow have carved out a space for optimism of the will.


Appiah, Kwame Anthony 2021, ‘Digging for Utopia’, New York Review online, 16 December, available at: <>.

Bloch, Maurice 1983, Marxism and Anthropology: The History of a Relationship, Oxford: Clarendon Press.

Childe, V. Gordon 1951 [1936], Man Makes Himself, New York: The New American Library.

Deresiewicz, William 2021, ‘Human History Gets a Rewrite’, The Atlantic online, 18 October 2021, available at:<>.

Diamond, Jared 2011 [2005], Collapse: How Societies Choose to Fail or Succeed, New York: Penguin Books.

Engels, Friedrich 1990 [1884; first English-language edition 1902], The Origin of the Family, Private Property, and the State, inMarx/Collected Works, Volume 26, pp. 129–275, London: Lawrence & Wishart.

Graeber, David 2014 [2011], Debt: The First 5,000 Years, Brooklyn, NY: Melville House.

Graeber, David and David Wengrow 2018, ‘How to Change the Course of Human History (or at Least the Part that’s Already Happened)’, Eurozine online, 2 March, available at: <>, accessed 14 August 2022.

Graeber, David and David Wengrow 2021, The Dawn of Everything: A New History of Humanity, New York: Farrar, Straus and Giroux.

Guha, Sumit 2022, ‘A False Dawn?’, Not Even Past online, 4 March,available at: <>.

Harari, Yuval Noah 2011, Sapiens: A Brief History of Humankind, London: Vintage.

Healey, Daniel 1993, ‘The Russian Revolution and the Decriminalisation of Homosexuality’, Revolutionary Russia, 6, 1: 26–54, <>, accessed 21 August 2022.

Knight, Chris, Nancy Lindisfarne and Jonathan Neale 2021, The Dawn of Everything Gets HumanHistory Wrong’, Monthly Review online, 20 December, available at: <>.

Lewis-Kraus, Gideon 2021, ‘Free for All’, New Yorker online, 1 November, available at: <>.

Morgan, Lewis Henry 1877, Ancient Society, or Researches in the Lines of Human Progress from Savagery, through Barbarism to Civilization, New York: Henry Holt and Company.

Newitz, Annalee 2022, ‘After 200,000 Years, We’re Still Trying to Figure out What Humanity Is All about’, Washington Post online, 26 November, available at: <>.

Rousseau, Jean Jacques 1987 [1754], Discourse on the Origin and Foundations of Inequality among Men, inRousseau’s Political Writings, edited by Alan Ritter and Julia Bondanella, translated by Julia Bondanella, New York: W.W. Norton & Company.

Smith, Adam 1976 [1776], An Inquiry into the Nature and Causes of The Wealth of Nations, Chicago, IL: The University of Chicago Press.

What Is Politics? 2022, ‘10.3 The Ingredients of Hierarchy: Graeber & Wengrow’s Dawn of Everything, Chapter 3’, YouTube, 22 February, available at: <>.


[1] Including Appiah 2021, Knight, Lindisfarne and Neale 2021, Lewis-Kraus 2021, Guha 2022, and the series of five video lectures on The Dawn (more than five hours in total), which is part of the vlogWhat Is Politics? 2022.

[2] Thus, for example, we read: ‘Our world as it existed just before the dawn of agriculture was anything but a world of roving hunter-gatherer bands’. Graeber and Wengrow 2021, p. 164. Also, see pp. 25 and 110. In an interview with Amy Goodman on Democracy Now!, Wengrow doubled down: ‘Firstly, this very conventional idea that before human beings invented agriculture, we lived almost exclusively in tiny egalitarian bands of hunter-gatherers. It’s not true. It’s actually wildly inaccurate.’ (Democracy Now!, 18 November 2021.)

[3]For the latter view, see, for example, Newitz 2022, Deresiewicz 2021, and Lewis-Kraus 2021.

[4] Thus, for example, we read: ‘Overwhelming evidence from archaeology, anthropology, and kindred disciplines is beginning to give us a fairly clear idea of what the last 40,000 years of human history really looked like, and in almost no way does it resemble the conventional narrative. Our species did not, in fact, spend most of its history in tiny bands; agriculture did not mark an irreversible threshold in social evolution; the first cities were often robustly egalitarian.’ Graeber and Wengrow 2018. In The Dawn, the authors first mention the 30,000-year figure, on p. 4.

[5] We now believe that Neanderthals disappeared some 40,000 years ago, and that Denisovans might have gone extinct perhaps as late as 20,000 years ago. We today might share a genetic inheritance with these two earlier groups, and perhaps with others, too.

[6]At least: archaeologists believe that the remains ofHomo sapiens found at a site in Jebel Ighoud in Morocco are 300,000 years old.

[7] Lindisfarne and Neale, among other reviewers, have emphasised that the accumulating anthropological and archaeological evidence tends to conflict with Graeber and Wengrow’s picture (Knight, Lindisfarne and Neale 2021). Sources they provide in their endnotes cast doubt on the contention that we have little archaeological evidence about earlier human societies.

[8] Kwame Anthony Appiah has observed that, for the authors of The Dawn, ‘the absence of evidence routinely serves as evidence of absence’ (Appiah 2021).

[9]The phrase is from a famous passage in Marx’s The Eighteenth Brumaire of Louis Bonaparte.

[10] With reference to dominance–submission behaviour (‘no doubt inherited from our simian ancestors’), the authors write that, ‘what makes societies distinctively human is our ability to decide not to act that way’ (Graeber and Wengrow 2021, p. 86).

[11] The authors write that the Marxist conception of the state, as they describe it on p. 360, is ‘very much in the tradition of Rousseau’, presumably because it posits a pre-class and pre-state idyll from which Man fell long ago. We will return to this theme in part III below.

[12] As they do, for example, with reference to the Poverty Point earthworks (Graeber and Wengrow 2021, Chapter 4), Ubaid village societies of Mesopotamia in the fifth millennium BCE (Graeber and Wengrow 2021, p. 422), and the Hopewell interaction sphere (Graeber and Wengrow 2021, p. 460).

[13] A mode of production comprises the direct preconditions for producing the necessities of life, including theproductive forces (land, labour, instruments, and raw materials), as well as therelations of production (the social structures, institutions, customs, and laws that regulate the relations among humans in the production of goods).

[14] For one thing, Marx and Engels, at various times, posited other modes of production, most famously the Asiatic mode of production. Moreover, they observed that the usual sequence can reverse itself: capitalism on some latifundia in imperial Rome, for example, gave way to feudalism, and as Engels emphasised in The Origin of the Family, Private Property and the State, Germanic foragers defeated Rome and replaced slavery and incipient capitalism with a ‘lower’ mode of production.

[15] In the case of the Taljanky ‘mega-site’, ‘no evidence was unearthed of centralized government or administration – or indeed, any form of ruling class’ (Graeber and Wengrow 2021, p. 289).

[16] The authors do not consistently distinguish between these two terms. At one point, for example, they refer to foraging, pastoralism, farming, and industry as ‘modes of production’ (Graeber and Wengrow 2021, p. 12).

[17] A social formation is ‘society viewed as a system for production and distribution, conceived of independently of the actors’ representations or justifications of the system’ (Bloch 1983, p. 23).

[18] Later in the book, the authors advance a related objection to ‘evolutionism’, namely, that ‘it takes ways of life that developed in symbiotic relations with each other and reorganises them into separate stages of human history’ (Graeber and Wengrow 2021, p. 446). Settled agriculturalists, for example, have interacted closely with ethnically, linguistically, and religiously distinct nomadic pastoralists, and the interactions have taken on the character of economic production, exchange, and tribute. (Marxists, however, do not assume at the outset that these relationships are ‘symbiotic’.) Let us note that, as long as we acknowledge that multiple modes of production often coexist within the same social formation, this objection to ‘evolutionism’, too, does not in any obvious way impugn the concept of mode of production.

[19] ‘Social scientists and political philosophers have been debating the “origins of the state” for well over a century’, the authors write. ‘These debates are never resolved and are unlikely ever to be. At this point, at least we can understand why. Much like the search for the “origins of inequality,” seeking the origins of the state is little more than chasing a phantasm.’ (Graeber and Wengrow 2021, p. 427.)

[20] The authors are referring here to that social surplus, the production of which, they say, ‘sets us apart from other non-human animals’ (Graeber and Wengrow 2021, p. 128).

[21] The authors acknowledge that ‘Most Mayanists would agree that Classic-period rulers lacked a sophisticated administrative apparatus’ (Graeber and Wengrow 2021, p. 410). But what is definitive here, at least for Graeber and Wengrow, is that the Mayans, or at least their rulers, ‘imagined the cosmos itself as a kind of administrative hierarchy … an intricate set of celestial or subterranean wheels, such that it was possible to establish the exact birth and death dates of major deities thousands of years in the past.’ This apparently qualifies them as states in which the bureaucratic principle was prominent, even though they did not manage to count the populations of their actual cities (Graeber and Wengrow 2021, p. 410).

[22] We might also recall a passage from Marx’s Preface to A Contribution to a Critique of Political Economy: ‘Just as one does not judge an individual by what he thinks about himself, so one cannot judge such a period of transformation by its consciousness, but, on the contrary, this consciousness must be explained from the contradictions of material life, from the conflict existing between the social forces of production and the relations of production.’ As we saw in section II above, this is the sort of formulation that qualifies Marx as an ‘evolutionist’.

[23] Including Marxists (see footnote 11 above).

[24] On this point Graeber and Wengrow find themselves in agreement with Friedrich Engels. In The Origin of the Family, Private Property and the State (1884), Engelssuggested that within human groups biological reproduction and the sexual division of labour foreshadowed class divisions, and the appearance of patriarchy foreshadowed the state.

[25] In a later chapter, the authors dispense with a Marxist theory of the state because, it ‘introduced new conceptual problems, such as how to define exploitation’ (Graeber and Wengrow 2021, p. 360). Let us note that the authors’ formulation, just cited, comes very close to a Marxist definition of the state, and it makes no mention of exploitation.

[26] According to this conception, there is no obvious reason why Weber’s ‘within a given geographical area’ qualification must always obtain for a state to be in place.

[27]Healy 1993. After the October Revolution in 1917, the Bolsheviks scrapped and rewrote the country’s laws. They produced two Criminal Codes – in 1922 and 1926 – and an article prohibiting gay sex was left off both.

[28] Here is one exemplary passage, among many: ‘All the evidence suggests that Teotihuacan had, at the height of its power, found a way to govern itself without overlords – as did the much earlier cities of prehistoric Ukraine, Uruk-period Mesopotamia and Bronze Age Pakistan’ (Graeber and Wengrow 2021, p. 330).