Origin of the state

Why did farming societies end up with rulers and taxes when foraging societies almost never did?

Foraging bands had big men, elders, respected shamans — but almost never anyone who could compel a share of your food year after year. Farming societies produced kings, granaries, scribes and standing armies, repeatedly and independently, on several continents. The obvious story is that farming made people rich enough to support rulers. But that story has a hole in it: several thousand years separate the earliest domesticated crops from the earliest states. If surplus alone did the work, why the long delay?

Ask what a ruler actually needs, and the question sharpens. Not wealth in general — a ruler needs to take a share, repeatedly, from people who would rather keep it, at a cost of collection lower than the take. Two conditions have to hold at once. The share must be findable and countable by an outsider. And the people paying it must be unable simply to leave.

Take findability first. What could you tax from a forager? A band's living is spread across dozens of resources that come in at different times, in different places, mostly in forms that spoil. There is no moment when a year's wealth sits in one heap. An assessor sent to appraise it would spend more than he collected, and the band would have moved by the time he returned.

Now consider cereals — wheat, barley, rice, millet. Notice how many separate properties they combine. They grow above ground, so they are visible. They ripen on a known schedule, so the harvest date is legible to a stranger. They must be cut within a short window, so a year's produce really does sit in one place at one time. And the grain is dense, divisible, countable and storable. This is James Scott's argument in Against the Grain: it is not agriculture that makes states, it is grain. The contrast case makes the point — cassava, yams and potatoes can be left in the ground and dug as needed, and a tax collector standing in a field of them has nothing to appraise and nothing to carry away.

But legibility on its own is not enough. If the assessor arrives and you can walk over the hill to unclaimed land, you do. This is the second condition, and it is Robert Carneiro's circumscription argument: where good land is bounded — by desert, sea or mountain, as in the Nile valley, the Peruvian coastal valleys, the irrigated strip of southern Mesopotamia — the losers of a conflict have nowhere to go. So they stay and submit, and get absorbed rather than displaced. Where land is open, subordination stays optional.

Put the two together and the account predicts its own exceptions. If what matters is storable, appraisable surplus rather than farming as such, then a society with no agriculture at all but with a concentrated, storable harvest should also develop ranks and coercion. The Pacific Northwest coast fits: salmon runs arriving at known places at known times, dried and stored in bulk, alongside hereditary chiefs, marked inequality and slavery, without domesticated crops.

The claim is not that agriculture caused the state; the millennia-long gap forbids that. It is that grain farming and bounded land are enabling conditions — they make a durable extractive apparatus affordable, without which it does not appear.

Three honest qualifications. First, this is contested. David Graeber and David Wengrow argue against the whole ladder from band to state, pointing to large early settlements with no obvious ruling class, and to societies that shifted seasonally between hierarchical and egalitarian arrangements. Carneiro's circumscription has counterexamples too. These are best read as conditions that raise the odds, not a mechanism that fires reliably.

Second, early states were fragile. They collapsed constantly, and what looks like catastrophe in the record is often dispersal — people leaving for hills, marshes and steppe where they could not be assessed, which Scott treats as an active strategy of avoiding the state rather than a failure to achieve one.

Third, none of this settles whether early subjects were coerced or attracted. Walls keep people out and also in, and the evidence rarely distinguishes the two.

• Why so many early states kept slaves and captives rather than merely taxing free farmers.
• What the "barbarian" periphery actually was, and how much of it consisted of former subjects.

Every term the collection defines is gathered in the glossary.