Decentralization Papers I. Before the State: 300,000 Years Without Kings

For most of our existence, humans actively prevented anyone from accumulating power. Hierarchy is the recent invention, not equality.

Every argument about how society should be organized smuggles in an assumption about what is natural. Defenders of concentrated power (kings once, states and platforms now) have always leaned on the same claim: hierarchy is simply what humans do. Someone must be in charge. It has always been this way.

It has not. Anatomically modern humans have existed for roughly three hundred thousand years. States, kings, and standing bureaucracies appear in the last five thousand, less than two percent of that span. For the rest, our ancestors lived in small, mobile bands of foragers, and the ethnographic record of surviving mobile forager bands is strikingly consistent: they are fiercely, deliberately egalitarian. Not, on the best ethnographic evidence, because resources were too thin to hoard, but because the group refused to let anyone hoard them.

The anthropologist Christopher Boehm spent a career assembling this record, and his conclusion deserves to be more famous than it is. Egalitarianism among foragers is not the absence of politics; it is politics. Boehm called it a reverse dominance hierarchy: the many, acting in coalition, dominate anyone who tries to dominate them. The tools are graduated and universal: ridicule for the boastful hunter, gossip about the greedy, disobedience toward the self-appointed, exile for the incorrigible. An aspiring strongman in a forager band does not face an army. He faces everyone.

Some societies ritualized this vigilance. Among Kalahari foragers, a hunter returning with a large kill is expected to disparage his own meat, and his campmates will do it for him if he forgets. It is a practiced ceremony against the conversion of skill into rank. Talent was welcome; the parlaying of talent into permanent authority was not. These were societies that had seen what we would call leadership capture, and had engineered institutions against it, institutions plausibly far older than any written constitution.

Scale mattered. The evolutionary psychologist Robin Dunbar argued that human cognition comfortably tracks stable social relationships numbering only in the low hundreds; beyond that, we cannot personally know whom to trust, and we need abstractions to stand in for acquaintance. Below that threshold, accountability is ambient: everyone sees what everyone does, and reputation is enforcement enough. The deep human toolkit for keeping power distributed (watchfulness, reputation, coalition) was built for groups where those tools work automatically.

None of this romanticizes the past. Forager life was materially poor, medically precarious, and sometimes violent; nobody at this publication proposes returning to it. The point is narrower and sharper: the claim that concentrated power is the natural human condition is false on the evidence. Everything we can see of our species' longest-running social experiment points in the opposite direction. And that experiment ran for three thousand centuries.

That inverts the burden of proof for every debate that follows in this series. Centralized power is not the default that decentralizers must argue against; it is the anomaly that requires justification: an invention with a birth date, adopted for particular reasons, carrying particular costs. What that invention was, why it won, and what it traded away is the subject of Part II.

This series, the Decentralization Papers, is the founding argument of this publication. Across six parts it will trace how power centralized, what democracy actually decentralized and what it failed to, why the great centralizing project of the twentieth century broke, which traditions decentralization genuinely belongs to, and what the way forward looks like now that our institutions can be written in software. The register you are reading it in (sequenced, hashed, append-only, publicly correctable) is a small piece of that argument, practiced rather than stated.