What If There Was No Eden? Graeber, Wengrow, and the Case Against Our Part I

Part I of the Decentralization Papers opened this publication with a claim about the deep past. Anatomically modern humans have existed for roughly three hundred thousand years; states and standing bureaucracies for about five thousand of them. For the rest, we argued, people lived in small mobile bands that were deliberately, actively egalitarian, and that egalitarianism was not the absence of politics but its product: Christopher Boehm's reverse dominance hierarchy, the many acting in coalition to prevent anyone from ruling them. We called that the default social order, and we used it to invert the burden of proof. Concentrated power is not the human baseline. It is an invention with a birth date, and it is the thing that requires justification.

There is a well-known book that says we got the framing wrong, and the honest thing to do with it is not to footnote it.

What the book argues

David Graeber and David Wengrow published The Dawn of Everything with Allen Lane on 19 October 2021: an anthropologist and an archaeologist, seven hundred pages, a sixty-three page bibliography. Their target is not the claim that hierarchy is natural. Their target is the idea that there was any default at all.

The evidence they assemble runs in three directions. Some societies switched political form by season, authoritarian in one part of the year and communal in another, which suggests populations that understood political arrangement as a choice rather than a condition. Some early cities show no palaces, no temples, no central storage and no written administration, and at least one, Teotihuacan, appears to have turned away from concentrated authority rather than toward it. And underneath both, they propose three freedoms as the real deep-historical constants: the freedom to move away, the freedom to disobey, and the freedom to remake social relations. The Wendat leader Kondiaronk, as rendered in Lahontan's dialogues, is their emblem of the third: an indigenous critique of European hierarchy, patriarchy, punitive law and profit that Europeans then had to answer.

[UNVERIFIED: this publication has not read the book cover to cover. The account above is drawn from its encyclopaedia summary and from the characterisations offered by its reviewers, including two who are hostile to it. Readers who want the argument in its own words should go to the book.]

Where it lands on Part I

It lands, and the phrase it breaks is "the default social order".

If political form varied seasonally inside a single society, then "default" is the wrong category. A default is a resting state that obtains unless something disturbs it. A society that is hierarchical for the hunt and communal for the gathering season has no resting state; it has a repertoire. Boehm's reverse dominance hierarchy survives this perfectly well as a description of a mechanism, one that people demonstrably used. It does not survive as a description of a universal baseline, and Part I used it as both.

So the correction is real and we should write it down. What it does not do is help the centralizer. Part I's actual work was the burden of proof, and Graeber and Wengrow make that burden heavier, not lighter. Boehm says people suppressed hierarchy when it appeared. Graeber and Wengrow say people chose between political forms and knew they were choosing. Both accounts refuse the claim that concentrated power is simply what humans do. The second refuses it more comprehensively than the first, because it denies the centralizer even the consolation that history was running in one direction.

The case against the book, which is serious

This is where an editorial of this kind usually stops. It should not.

Walter Scheidel reviewed the book in Cliodynamics in April 2022, and he concedes a great deal before he attacks. He accepts that foragers were "not necessarily confined to small bands", that monumental building does not require chiefs or states (Poverty Point in Louisiana, Jomon-era Japan), that farming spread slowly and with reversals, that agriculture was "much less of a rupture than one might imagine", and that there were real lag times between farming and the emergence of inequality. He also grants that classicism has fetishised state power and that rescuing the pre-literate millennia from invisibility is a genuine service.

Then he says the book has an ideology problem. He fastens on the authors' own admission that where one sets "the dial between freedom and determinism" is largely a matter of taste, and argues that ecology, technology, demography and warfare are systematically sidelined so that the dial can stay where they like it. On the agricultural trap, his line is hard to shake off: "even a trap that was slow in closing was, in the end, a trap." On the claim that linear narratives are refuted by slow, uneven transitions, he calls it "a red herring: trajectories can be both linear and very long."

He also disputes particulars. He says the earliest large-scale societies appeared where storable domesticates appeared early, and that regions whose staples were poorly storable (banana, yam, taro, manioc) did not generate states. He says Babylon and large urban sites in South and East Asia predated Teotihuacan, against the book's chronology, and that Teotihuacan remained an outlier in Mesoamerica until Tenochtitlan in the fifteenth century. On North America he argues that Cahokia, at roughly fifteen thousand residents, was followed by dispersal because sparse population and the absence of horses made exit cheap and power projection hard, which is a material explanation of a supposedly political choice. And he offers a distributional check on the whole argument: around the beginning of the Common Era, he estimates, something like three-quarters of humanity lived inside four Eurasian empires.

He is not alone. Gary Feinman charges the book with "cherry-picked and selectively presented examples". Kwame Anthony Appiah found a "discordance between what the book says and its sources say", while judging the argument against historical determinism "immensely valuable". Arjun Appadurai described it as "swerving to avoid a host of counter-examples and counter-arguments". The historian David Bell, on the specific claim that indigenous critique shaped the Enlightenment, went as far as "scholarly malpractice".

And the objection from the left, which is the one aimed at us

Peter Kulchyski's review in Historical Materialism is the one this publication should take most personally, because his political objection is directed at exactly the use we would make of the book.

His argument is that by rejecting egalitarianism as a defining feature of gatherer-hunter societies, Graeber and Wengrow remove a load-bearing beam from the politics of Indigenous bush peoples, whose claims rest on demonstrated historical alternatives to hierarchy. The book, he writes, "will demoralise many of the readers it hoped to inspire". His theoretical objection is the companion of that: the authors discard mode of production without putting anything in its place, and what is left is voluntarism, a world in which "hey ho, everything goes". To say a society has a structure, he insists, is to say there are limits to how far it can vary. Collapse the structure-agency dialectic and you have not liberated anyone; you have only stopped explaining anything.

That charge has force against us specifically. A publication that argues for decentralization has an obvious motive to believe that political form is chosen, because a chosen form can be rechosen. If everything is choice, though, then nothing accounts for why the same choice keeps getting made, and we would be left telling readers that the world could be otherwise without being able to say what is holding it in place.

What we amend, and what we keep

Part I should not say that egalitarianism was the default social order. It should say that egalitarian arrangements were achievable, common, deliberately maintained and repeatedly rebuilt, and that societies knew how to do it because they kept doing it. That is a weaker claim and a more defensible one, and it does the same work in the argument: hierarchy still holds no writ from nature.

Part II comes out of this better than Part I. Its claim is that storable, countable surplus is what states form around, and that the ledger was the first instrument of centralization. That is a materialist claim, and Scheidel's own storability argument, that societies scaled where the crops kept and did not where they rotted, is closer to supporting it than to refuting it. The grain bargain does not need a Garden of Eden before it.

What we take from Graeber and Wengrow, in the end, is not their history. It is their test. Boehm's framework asks whether a society is egalitarian, which is a question about outcomes and is hard to answer from outside. The three freedoms ask something better, because they ask about the structure rather than the result: can you leave, can you refuse, and can you rewrite the terms.

That converts unexpectedly well to the things this publication actually covers. Can a household leave its distribution utility, or is the exit priced out by an additional surcharge. Can a rival assistant refuse the platform's terms, or must it apply for certification to the firm it competes with. Can a cooperative's members rewrite their own rules, or does the rulebook live somewhere they cannot reach. Those are the same three questions in a different century, and they are considerably more useful to us than an argument about what our ancestors did.

Which leaves the uncomfortable part, and it should stay uncomfortable. The strongest criticism of The Dawn of Everything is that its authors set the dial by preference. We have the same preference. The defence against it is not to insist we are right about the Pleistocene. It is to publish the criticism, write the amendment in the open, and let both stand in the record.

What is still unknown

How reliable the Kondiaronk material is as history rather than as Lahontan's literary construction. Bell's "scholarly malpractice" charge is aimed here, and this publication has not read the underlying dialogues or the specialist literature on them.

Whether Boehm's reverse-dominance evidence and Graeber and Wengrow's account actually conflict, or whether a repertoire that includes egalitarian arrangements and a mechanism for maintaining them are complementary claims that only look opposed because both sides needed a fight.

What archaeology has said since 2021 about the Ukrainian mega-sites and the cities-without-rulers reading of them. No post-2021 assessment was located for this piece. [NEEDS DATA: peer-reviewed work after 2021 on the Trypillia mega-sites and on Teotihuacan's political organisation.]

Whether any of the critics quoted here have revised their positions since their reviews appeared. Only Appiah's later and more favourable judgement was found.

Sources

Walter Scheidel, review of The Dawn of Everything, Cliodynamics, April 2022, hosted at davidgraeber.org: https://davidgraeber.org/reviews/walter-scheidel-review-of-the-dawn-of-everything/

Peter Kulchyski, "Everything Goes: Three Problems with The Dawn of Everything", Historical Materialism: https://www.historicalmaterialism.org/everything-goes-three-problems-with-the-dawn-of-everything-a-review-of-the-dawn-of-everything-by-david-graeber-and-david-wengrow/

The Dawn of Everything, encyclopaedia entry, used for publication details and for the reception survey naming Appiah, Feinman, Appadurai, Bell, Smith, Handler, Morris, Suzman and De Cruz: https://en.wikipedia.org/wiki/The_Dawn_of_Everything

Christopher Boehm, for the reverse dominance hierarchy as cited in Part I: https://en.wikipedia.org/wiki/Christopher_Boehm