David Graeber and David Wengrow recently wrote a long piece in the New Humanist, Are we city dwellers or hunter-gatherers? New research suggests that the familiar story of early human society is wrong – and the consequences are profound. What follows is my critical review of it. The structure that I adopt is quoting large chunks from their essay followed by my commentary (I don’t usually quote at length, but since my take is quite critical, I chose to let the authors speak in their voices, rather than paraphrase).
Graeber is an American anthropologist and anarchist activist. I read three of his books, including Debt. Wengrow is an archaeologist specializing in the early history of Egypt. I read two of his books, including What Makes Civilization? Both aim at the “public intellectuals” status, with Graeber, clearly, much farther along the way to fame. Let’s see how well their latest piece works.
The authors start their essay with the following passage:
For centuries, we have been telling ourselves a simple story about the origins of social inequality. For most of their history, humans lived in tiny egalitarian bands of hunter-gatherers. Then came farming, which brought with it private property, and then the rise of cities which meant the emergence of civilisation properly speaking. Civilisation meant many bad things (wars, taxes, bureaucracy, patriarchy, slavery) but also made possible written literature, science, philosophy and most other great human achievements.
Hold on a second! I object to putting taxes and bureaucracy into this list of “bad things.” Without taxes we could have no government, and without government we would have no public goods that it produces, which is what really makes possible high standard of living we enjoy in reasonably well-governed societies (which include Western Europe and North America). Of course, taxes can be used not to elevate general well-being but for other less productive, or outright malignant purposes. But it would be like saying that fire is a bad thing because people get burned to death in forest fires. Same for bureaucracy. We love to hate bureaucrats, but large-scale societies cannot function without professional administrators. Take a look at this graph from our recent article analyzing Seshat data:
What it says is that once the polity population gets to roughly 200 thousand (and certainly by the time you exceed a couple of millions), it must have sophisticated government institutions, including professional bureaucrats. A society numbering in millions simply can’t function without specialized administrators. Societies that try to do it, instead fall apart, which is why we don’t see them today, or (much) in history.
The conclusion from this is that the way forward to sustaining and increasing the well-being of large segments of population is not to abolish government, but to evolve institutions that keep bureaucrats working for the benefit of the population, rather than themselves.
But let’s get on with the essay.
Almost everyone knows this story in its broadest outlines. Since at least the days of the 18th-century philosopher Jean-Jacques Rousseau, it has framed what we think the overall shape and direction of human history to be. This is important because the narrative also defines our sense of political possibility. Most see civilisation, hence inequality, as a tragic necessity. Some dream of returning to a past utopia, of finding an industrial equivalent to “primitive communism”, or even, in extreme cases, of destroying everything, and going back to being foragers again. But no one challenges the basic structure of the story.
There is a fundamental problem with this narrative: it isn’t true. … Those writers who are reflecting on the “big questions” of human history – Jared Diamond, Francis Fukuyama, Ian Morris and others – still take Rousseau’s question (“what is the origin of social inequality?”) as their starting point, and assume the larger story will begin with some kind of fall from primordial innocence.
Graeber and Wengrow employ here an old and highly effective, but intellectually dishonest rhetorical device. They present the reader with a caricature, and then associate it with authors who actually say quite different things. Ian Morris, for example, is hardly a Rousseauian. Just read his War, What Is It Good for? Not that I necessarily agree with everything he says (see this blog post).
As to their substantive critique, let’s read more before I address it. Skipping a few paragraphs, we get to the main thesis:
What we’re going to do in this essay, then, is two things. First, we will spend a bit of time picking through what passes for informed opinion on such matters, to reveal how the game is played, how even the most apparently sophisticated contemporary scholars end up reproducing conventional wisdom as it stood in France or Scotland in, say, 1760. Then we will attempt to lay down the initial foundations of an entirely different narrative. This is mostly ground-clearing work. The questions we are dealing with are so enormous, and the issues so important, that it will take years of research and debate to even begin to understand the full implications. But on one thing we insist. Abandoning the story of a fall from primordial innocence does not mean abandoning dreams of human emancipation – that is, of a society where no one can turn their rights in property into a means of enslaving others, and where no one can be told their lives and needs don’t matter. On the contrary. Human history becomes a far more interesting place, containing many more hopeful moments than we’ve been led to imagine, once we learn to throw off our conceptual shackles and perceive what’s really there.
Well, let’s throw off our conceptual shackles and follow the authors on their intellectual journey.
I again skip quite a bunch of paragraphs, in which Graeber and Wengrow first expand on their caricature and then criticize Jared Diamond and Francis Fukuyama. A few of their criticisms I actually agree with—for my alternative view on the rise of complex societies during the last 10,000 years, see my 2016 book Ultrasociety.
A dismal conclusion, not just for anarchists, but for anybody who ever wondered if there might be some viable alternative to the status quo. But the remarkable thing is that, despite the smug tone, such pronouncements are not actually based on any kind of scientific evidence. There is no reason to believe that small-scale groups are especially likely to be egalitarian, or that large ones must necessarily have kings, presidents or bureaucracies. These are just prejudices stated as facts.
In the case of Fukuyama and Diamond one can, at least, note they were never trained in the relevant disciplines (the first is a political scientist, the other has a PhD on the physiology of the gall bladder).
Ouch! What could a gall bladder specialist possibly tell us about the evolution of human societies?? Seriously, what does the topic of dissertation that Diamond defended in 1961 have to do with the validity of his ideas four decades later? Diamond is a broad-band thinker, who during his long career contributed to diverse fields (including, for example, community ecology). His ideas may be controversial (and I don’t agree with everything he says), but they have been influential and productive — that is, they lead to new empirical and theoretical research.
Still, even when anthropologists and archaeologists try their hand at “big picture” narratives, they have an odd tendency to end up with some similarly minor variation on Rousseau. In The Creation of Inequality: How our Prehistoric Ancestors Set the Stage for Monarchy, Slavery, and Empire (Harvard University Press, 2012), Kent Flannery and Joyce Marcus, two eminently qualified scholars, lay out some 500 pages of ethnographic and archaeological case studies to try and solve the puzzle.
Here the Davids actually make a good point. I myself slammed Flannery and Marcus for dragging in Rousseau in an otherwise positive review of their book in the Times Literary Supplement (unfortunately behind a paywall, but the preprint is here).
After several more paragraphs, Graeber and Wengrow finally get to presenting their alternative understanding about the course of human history.
So, what do we actually know about this period of human history? Much of the earliest substantial evidence for human social organisation in the Palaeolithic derives from Europe, where our species became established alongside Homo neanderthalensis, prior to the latter’s extinction around 40,000 BC. (The concentration of data in this part of the world most likely reflects a historical bias of archaeological investigation, rather than anything unusual about Europe itself.) … Prehistorians have pointed out for some decades – to little apparent effect – that the human groups inhabiting these environments had nothing in common with those blissfully simple, egalitarian bands of hunter-gatherers still routinely imagined to be our remote ancestors.
To begin with, there is the undisputed existence of rich burials, extending back in time to the depths of the Ice Age. Some of these, such as the 25,000-year-old graves from Sungir, east of Moscow, have been known for many decades and are justly famous. Felipe Fernández-Armesto, who reviewed The Creation of Inequality for The Wall Street Journal, expresses his reasonable amazement at their omission: “Though they know that the hereditary principle predated agriculture, Mr. Flannery and Ms. Marcus cannot quite shed the Rousseauian illusion that it started with sedentary life. Therefore they depict a world without inherited power until about 15,000 B.C. while ignoring one of the most important archaeological sites for their purpose.” Dug into the permafrost beneath the Palaeolithic settlement at Sungir was the grave of a middle-aged man buried, as Fernández-Armesto observes, with “stunning signs of honor: bracelets of polished mammoth-ivory, a diadem or cap of fox’s teeth, and nearly 3,000 laboriously carved and polished ivory beads.” And a few feet away, in an identical grave, “lay two children, of about 10 and 13 years respectively, adorned with comparable grave-gifts – including, in the case of the elder, some 5,000 beads as fine as the adult’s (although slightly smaller) and a massive lance carved from ivory.”
Indeed, this is an amazing display of wealth. In The Upper Paleolithic of the Central Russian Plain (p. 456), Olga Soffer estimates that just the beads alone represent over 2,500 hours of human labor. But this is only a single datum to support a lot of theory. Furthermore, we don’t really know what role the children played. Were they “princes”? Or sacrifices, as some archaeologists suggested?
Such findings appear to have no significant place in any of the books so far considered. Downplaying them, or reducing them to footnotes, might be more easy to forgive were Sungir an isolated find. It is not. Comparably rich burials are by now attested from Upper Palaeolithic rock shelters and open-air settlements across much of western Eurasia, from the Don to the Dordogne. Among them we find, for example, the 16,000-year-old “Lady of Saint-Germain-la-Rivière”, bedecked with ornaments made of the teeth of young stags hunted 300 km away, in the Spanish Basque country; and the burials of the Ligurian coast – as ancient as Sungir – including “Il Principe”, a young man whose regalia included a sceptre of exotic flint, elk antler batons and an ornate headdress of perforated shells and deer teeth.
These other examples don’t sound very compelling to me. It would be good to get a professional archaeologist estimate of the amount of human labor needed to produce these ornaments, but I doubt it would approach the Sunghir grave. Still, the point is well taken that we do see substantial displays of wealth in pre-Neolithic societies. This is not news for archaeologists, however, because pretty much everybody I talk to agrees that it’s not really agriculture that leads to the rise of inequality. Foraging societies in highly productive areas, like Northwest Indians, have developed highly unequal societies with high differentials in wealth, slavery, etc.
No less intriguing is the sporadic but compelling evidence for monumental architecture, stretching back to the Last Glacial Maximum. The idea that one could measure “monumentality” in absolute terms is of course as silly as the idea of quantifying Ice Age expenditure in dollars and cents. It is a relative concept, which makes sense only within a particular scale of values and prior experiences. The Pleistocene has no direct equivalents in scale to the Pyramids of Giza or the Roman Colosseum. But it does have buildings that, by the standards of the time, could only have been considered public works, implying sophisticated design and the coordination of labour on an impressive scale. Among them are the startling “mammoth houses”, built of hides stretched over a frame of tusks, examples of which – dating to around 15,000 years ago – can be found along a transect of the glacial fringe reaching from modern-day Kraków all the way to Kiev.
Until this paragraph I’ve been half-willing to go for a ride with Graeber/Wengrow. Yes, they caricature the views of their opponents, and are not above ad hominem attacks, but I agree with some of the points they make, such as that there was no sharp transition in human social evolution with the adoption of agriculture (and it’s worth pointing out that this is not a novel idea for most Neolithic archaeologists). With this passage, however, Graeber and Wengrow themselves become guilty of monumental silliness.
An example of a “monumental” mammoth bone hut in the Vienna Museum of Natural History (photo by the author)
First, their dismissal of the possibility of measuring monumentality. In my experience, refusal to quantify is usually the last refuge of those who don’t want to see their pet theories rejected. G&W are not above quantifying when it suits their needs. Clearly 5,000 laboriously carved and polished beads represent a much more massive investment of human labor than 5 beads. Three orders of magnitude, to be precise (an order of magnitude is a 10-fold change). This is a truly big difference.
Second, archaeologists are already quantifying monumentality — by how much labor, in people-hours of work, it takes to construct the monument in question. Human labor is a universal coin. Yes, there is some variation in how much different people value an hour of work (and it depends, of course, on the kind of work). But there are ways to incorporate such factors into our estimates. An hour of construction work is a pretty good unit. What’s important, is that any variations in the value of this unit in different cultures and different periods of human evolution are dwarfed by the many orders of magnitude in the sheer number of work hours needed to construct different monuments in human history.
I used this device at the beginning of my book Ultrasociety to trace the scale at which people cooperate to construct impressive buildings. Looking at such monuments as the Empire State Building, the Amiens Cathedral, Egyptian Pyramids, and Gobekli Tepe, I showed the social scale implied by the scale of the monument diminishes as we go back into the past. And this is a change of many orders of magnitude. For example, it required roughly 400,000 people-years of work to construct the Great Pyramid of Giza, but only 300 people-years to build each Gobekli Tepe temple (for details and labor estimates for other famous monuments, see Chapter 1 of Ultrasociety). Let’s add mammoth bone houses to this sequence:
I doubt that it took one person more than a day to construct a mammoth hut (after you have hunted down and butchered the mammoth, of course). It would take much more time (and people) to eat such a mountain of meat! The difference between one day of work and 300 people-years, required for Gobekli Tepe, is 5 orders of magnitude (365 days/year x 300 people-years = 109,500 people-days). And there is another jump of 3 orders of magnitude from Gobekli to the Great Pyramid. Claiming that the monumentality of a mammoth house is not really different from that of the Great Pyramid is, well, silly.
Still more astonishing are the stone temples of Göbekli Tepe, excavated over 20 years ago on the Turkish-Syrian border, …
Indeed, 5 orders of magnitude more astonishing than a mammoth hut. And note that, although the builders of Gobekli Tepe did not practice agriculture, crops such as wheat and barley were already cultivated in areas only a couple hundred kilometers away.
What, then, are we to make of all of this? One scholarly response has been to abandon the idea of an egalitarian Golden Age entirely, and conclude that rational self-interest and accumulation of power are the enduring forces behind human social development. But this doesn’t really work either. Evidence for institutional inequality in Ice Age societies, whether in the form of grand burials or monumental buildings, is nothing if not sporadic. Burials appear literally centuries, and often hundreds of kilometres, apart. Even if we put this down to the patchiness of the evidence, we still have to ask why the evidence is so patchy: after all, if any of these Ice Age “princes” had behaved anything like, say, Bronze Age princes, we’d also be finding fortifications, storehouses, palaces – all the usual trappings of emergent states.
This passage seems to contradict what has come before…
A wider look at the archaeological evidence suggests a key to resolving the dilemma. It lies in the seasonal rhythms of prehistoric social life. Most of the Palaeolithic sites discussed so far are associated with evidence for annual or biennial periods of aggregation, linked to the migrations of game herds – whether woolly mammoth, steppe bison, reindeer or (in the case of Göbekli Tepe) gazelle – as well as cyclical fish-runs and nut harvests. At less favourable times of year, at least some of our Ice Age ancestors no doubt really did live and forage in tiny bands. But there is overwhelming evidence to show that at others they congregated en masse within the kind of “micro-cities” found at Dolní Věstonice, in the Moravian basin south of Brno, Czech Republic, feasting on a superabundance of wild resources, engaging in complex rituals and ambitious artistic enterprises, and trading minerals, marine shells and animal pelts over striking distances. Western European equivalents of these seasonal aggregation sites would be the great rock shelters of the French Périgord and Spain’s Cantabrian coast, with their famous paintings and carvings, which similarly formed part of an annual round of congregation and dispersal.
Why are these seasonal variations important? Because they reveal that from the very beginning, human beings were self-consciously experimenting with different social possibilities. Anthropologists describe societies of this sort as possessing a “double morphology”. … Most striking, in terms of political reversals, were the seasonal practices of 19th-century tribal confederacies on the American Great Plains – sometime or one-time farmers who had adopted a nomadic hunting life. In the late summer, small and highly mobile bands of Cheyenne and Lakota would congregate in large settlements to make logistical preparations for the buffalo hunt. At this most sensitive time of year they appointed a police force that exercised full coercive powers, including the right to imprison, whip or fine any offender who endangered the proceedings. Yet as the anthropologist Robert Lowie observed, this “unequivocal authoritarianism” operated on a strictly seasonal and temporary basis, giving way to more “anarchic” forms of organisation once the hunting season and the collective rituals that followed were complete.
Scholarship does not always advance. Sometimes it slides backwards. A hundred years ago, most anthropologists understood that those who lived mainly from wild resources were not, normally, restricted to tiny “bands”. That idea is really a product of the 1960s, when Kalahari Bushmen and Mbuti Pygmies became the preferred image of primordial humanity for TV audiences and researchers alike. As a result we’ve seen a return of evolutionary stages, really not all that different from the tradition of the Scottish Enlightenment: this is what Fukuyama, for instance, is drawing on, when he writes of society evolving steadily from “bands” to “tribes” to “chiefdoms”, then finally, the kind of complex and stratified “states” we live in today – usually defined by their monopoly of “the legitimate use of coercive force”. By this logic, however, the Cheyenne or Lakota would have had to be “evolving” from bands directly to states roughly every November, and then “devolving” back again come spring.
Here we go again. To equate late summer congregations of the Cheyenne or the Lakota to complex stratified states does as much violence to data as equating a mammoth hut to a great pyramid. The social scale of such seasonal congregations of hunter-gatherers was a few thousand people. Complex large-scale societies organized as states, such as we find in Ancient Egypt, have populations counted in millions, tens of millions, and more. That’s a difference of 3-4 orders of magnitude. States are also organized in a centralized fashion. There is a supreme ruler (a king, an emperor, or a president) at the top of deep vertical hierarchy with 4, 5, 6 and more levels of control. Even more importantly, states are characterized by a internally specialized governance. This means that we have people who specialize as administrators (the bureaucrats), others as military leaders (officers), yet other as ideological leaders (priests). There was nothing like that in the Cheyenne society. Of all American Indian societies on the Great Plains, it was the Comanche who approached a politically centralized society the closest, but even they did not have a supreme leader (read the great book The Comanche Empire by Pekka Hamalainen for this fascinating story). A monopoly on legitimate use of coercive force is also a huge stretch. Tribal police of the Plains Indians was a community-based force whose purpose was to control non-cooperators. In fact, small-scale societies can control the behavior of their members much more effectively (and oppressively) than the state’s police, as anybody who lived in a small village can attest.
Modern authors have a tendency to use prehistory as a canvas for working out philosophical problems…
Indeed. And Graeber and Wengrow provide us with a striking example of how ideology can override common sense.
…the real question is not “what are the origins of social inequality?” but, having lived so much of our history moving back and forth between different political systems, “how did we get so stuck?” All this is very far from the notion of prehistoric societies drifting blindly towards the institutional chains that bind them. It is also far from the dismal prophecies of Fukuyama, Diamond et al. where any “complex” form of social organisation necessarily means that tiny elites take charge of key resources, and begin to trample everyone else underfoot.
Jared Diamond notwithstanding, there is absolutely no evidence that top-down structures of rule are the necessary consequence of large-scale organisation. Walter Scheidel notwithstanding, it is simply not true that ruling classes, once established, cannot be gotten rid of except by general catastrophe.
In fact, there is absolutely no evidence that any large-scale human society can be organized in any other way than hierarchically. We are not ants! (I expand on this theme in the Pipe Dream of Anarcho-Populism)
And social movements organized on the principle of anarcho-populism invariably fail. As did the Occupy Movement in which David Graber played an important role. Where are the “occupiers” today? What have they accomplished? Nothing.
The only way to achieve a lasting positive change at the society’s level is through effective political organization, which in humans means chains of command (I explain why in Ultrasociety). Of course, once leaders emerge there is a terrible temptation for them to subvert their social power to their selfish purposes. This is why the first centralized societies quickly became despotisms. But then cultural group selection started weeding out the most despotic societies, resulting in the evolution of norms and institutions that began to restrain the worse excesses of power abuse. I tell this story in Ultrasociety.
The pieces are all there to create an entirely different world history. For the most part, we’re just too blinded by our prejudices to see the implications.
If you are interested in an alternative view of social evolution, different from both Graber/Wengrow’s and their caricature of the conventional narrative, read Ultrasociety. As I explain in the book, there is a way forward to a peaceful, affluent, and just society. But anarchism is not such a way — it’s a blind alley.