Every form you have ever filled in opens with the same boxes: name, address, identity number. It all looks self-evident. Yet fixed, inherited surnames are only a century or two old in much of the world, and the reason they appeared was stated plainly at the time — to collect taxes and to conscript soldiers.
This page takes apart three things: what governments have actually done to make their territory visible; what the knowledge that will not fit into a form looks like; and whether anyone has simply arranged their life so as not to be worth registering.
In late-eighteenth-century Prussia and Saxony, foresters acquired a new instrument: the Normalbaum, a standard tree. Measure the timber volume of one standard specimen and you could extrapolate what the whole stand was worth. It was an elegant invention — for the first time a forest could be written into a budget.
The next step is the one that mattered. To make the ledger accurate, improving measurement is laborious; remaking the forest is cheap. So the undergrowth was cleared, the mixed trees felled, the leaf litter raked away, and in their place went single-species, same-age plantations in straight rows, mostly spruce and pine. Walking into one is like walking down a supermarket aisle. The first rotation grew superbly, yields were predictable, and officials were pleased.
The second rotation failed. About a century later the same ground was yielding twenty to thirty percent less, and German forestry science coined a word for the worst cases: Waldsterben, forest death. The reconstruction runs like this: the first crop had been living off soil capital the old forest had banked over centuries — humus, mycorrhizal fungi, insects, birds, fallen wood. None of it appeared in the ledger, because none of it yields timber. The ledger had not forgotten these things. It had no column that could hold them.
Evidence strength should be stated: this is a historical case with a retrospective reconstruction, not a controlled experiment. And the word Waldsterben in the 1980s was used mainly for forest damage attributed to acid deposition, which is a different matter from the monoculture story and should not be run together with it.
James C. Scott's 1998 book distils this into a single mechanism he calls legibility. The point is not that states want to know more; everyone wants to know more. The point is that usable state knowledge has to be countable, comparable and verifiable from an office two hundred kilometres away. A fact that holds only for people who live there does not exist for that office.
Surnames came from exactly this. In 1849 the Spanish colonial administration in the Philippines issued the Clavería decree, accompanied by a catalogue of more than sixty thousand surnames, distributed province by province, town by town, alphabetically, one per household. The decree stated its reasons flatly: to register the population, to tax it, and to conscript from it.
The whole thing has to be said, though. The same apparatus is the precondition for public health, mass schooling, vaccination records and disaster relief — nobody can vaccinate a person they cannot find. Legibility is not a bad thing. It is a costly thing. The cost is in the next section.
There is a kind of strike in which nobody stops working. People come in as usual and simply carry out every regulation to the letter — the French call it grève du zèle, a strike of zeal — and production collapses. Nothing makes the point more directly: written procedure has never been self-sufficient. It has always been running on top of a large body of judgement that was never written down.
Scott gives that body of judgement a name: mētis, borrowed from ancient Greek and taken from Detienne and Vernant's 1974 study of Greek cunning intelligence. Its marks are easy to spot — local, situational, learnable only by doing, and half-destroyed the moment it is turned into a manual. The classic example is river pilotage. A large ship takes a local pilot aboard on the way into harbour, not because the visiting captain is a worse sailor, but because the sandbars in that channel move every year, and where they are this year is on no chart.
Agriculture supplies a solid piece of fieldwork. Paul Richards's Indigenous Agricultural Revolution (1985) describes intercropping among West African farmers, mostly in Sierra Leone: several crops in one plot, ripening at different times, rooting at different depths, the tall ones shading the short ones. Colonial agronomists, and their successors, long read this as backward and pushed single crops sown in rows. The reasons were not purely prejudice — a monocrop can be yield-measured, compared across sites, and supervised by an inspector, and a mixed plot allows none of the three. Richards argues that in local soils and rainfall, intercropping is frequently the more robust system: against drought, against disease, and in spreading labour across the season. Evidence strength: field research plus agronomic comparison, not a universal result — under other conditions monocropping does yield more.
The largest case of all is Tanzania's villagisation between 1973 and 1976. Dispersed households were moved into planned villages; the counts differ a great deal, with estimates running from somewhat over five million people to above ten million. The aims were not bad ones, and were partly achieved: you can only build a school, pipe water and send out an agricultural adviser where people live close together. But the village sites were fixed from plans and road lines, not from what each household had accumulated over decades about which plot to plant in which year, which slope dries out first, and which hollow actually carries a crop through a wet season. Yields fell.
The book's most transferable passage is its four conditions, which produce catastrophe only when they arrive together: administrative simplification, a confidence that the thing is now understood, a power able to impose it, and a society unable to resist. Take away one and you usually get an ugly project rather than a disaster.
The criticism belongs here too. Some readers find Scott too pessimistic about state capacity; others find he romanticises mētis — local knowledge also gets things wrong, also clings to practices that do harm, and being hard to articulate is not the same as being right. No side is taken here: both readings have cases that stand up.
Start with two kinds of crop. Grain grows above ground, ripens all at once, and stores and travels well: a tax collector riding out can estimate a field's yield at a glance, and has to turn up during the few days of harvest. Tubers like yams and cassava grow underground, get dug when they are wanted, and keep badly once lifted — collecting them means digging plot by plot. The same farming labour, and yet one of them is almost purpose-built for taxation and the other is almost impossible to tax.
The name Zomia was proposed in 2002 by the historian Willem van Schendel, from a Tibeto-Burman word for highlander. It refers to the belt of mainland Southeast Asian uplands above three hundred metres — roughly 2.5 million square kilometres, spanning several countries, home to something like a hundred million people. Scott's The Art of Not Being Governed (2009) offers a provocative reading of it: the traits that lowland states read as backward — shifting cultivation rather than settlement, oral transmission rather than writing, genealogies revised at will, dispersed social structures with no centre — may not be stages not yet reached, but secondary and strategic. Each one lowers the odds of being taxed, conscripted, or seized for forced labour.
The criticism is heavy and deserves to be reported as it stands. Historians point out that uplands and lowlands traded, paid tribute and intermarried over long stretches of time, that a number of upland groups built polities of their own, and that evasion is nowhere near the only available explanation. Reading hundreds of differently descended groups as one category of evaders is itself a simplification — precisely the kind that the forestry section criticises. Scott himself holds that the argument largely stops applying after 1945, once roads, communications and survey had flattened the friction of terrain. Evidence strength: an interpretive argument in historical anthropology, not a testable causal claim.
What travels is not admiration for highlanders; swapping backward for clever still leaves them without a history of their own. What travels is colder: once something is used for extraction or assessment, it starts deforming towards whatever is hard to measure. The piece on organisations and institutions covered why organisations come to resemble each other; this is the other half of it — the party being measured moves too, and often moves faster.