Day 31 · Food & Empire

The Imperial Ledger of Four Crops

Monday, 27 July 2026 · BigCat's Time Machine
A price gap on one nutmeg bought a massacre. A cup of sweetened tea kept a factory running. One tuber fed hundreds of millions more people — and starved a million. Crops make no decisions; they merely re-rank the costs and payoffs of the decisions people make.
EVENT · 01

Emptying an Archipelago for One SpiceThe Banda Massacre · 1621

Feb–May 1621Banda IslandsVOC

Nutmeg then grew only in the Banda Islands — ten small islands holding roughly 15,000 people — and sold in Amsterdam for hundreds of times its price at source. The Dutch East India Company (VOC, founded 1602) was the first joint-stock company ever authorized to declare war, build forts and sign treaties: sovereignty outsourced to a balance sheet. Governor-General Jan Pieterszoon Coen (1587–1629) judged that a monopoly could not be held by contract, only by holding the ground.

In February 1621 Coen landed with some 2,000 men on Banda Besar — the Bandanese elders (orang kaya) had also signed with the English, which the VOC read as breach. Between March and May, forty-four elders were beheaded by Japanese mercenaries as a public display; by year's end the islands' population had fallen below a thousand. The VOC promptly carved the land into some sixty plantations (perken), allotted them to Dutch perkeniers, and worked them with enslaved people shipped in from elsewhere — a population replacement performed to nail down a price curve. In the 1667 Treaty of Breda, England traded Banda's Run island away for Manhattan; on the books of the day, that was the shrewd side of the deal.

Had the VOC signed only purchase contracts, the Bandanese would have kept selling to the English and to Javanese traders alike, and the monopoly rent would have gone to zero — while the VOC's credit rested on high dividends. Coen's atrocity was not a loss of control; it was an inference from the business model. The dispute is over how good that ledger really looked: Giles Milton's Nathaniel's Nutmeg (1999) plays the drama, while Femme Gaastra's long-run accounting of the VOC shows that by the eighteenth century profits came far more from intra-Asian shipping and textiles — the spice monopoly had already depreciated.

Geographic monopolies over critical minerals (cobalt, rare earths) pose the same problem: the return on controlling the node falls as substitutes appear, while the cost of holding it rises with time. The two curves eventually cross.

The cost of a moat is permanent; the rent it collects depreciates.
Which barrier are you defending whose upkeep is now growing faster than the returns it earns?
EVENT · 02

Writing People into Property: Barbados SugarSugar and the Barbados Slave Code · 1640s–1661

1661Barbadoschattel slavery

England seized Barbados in 1627 and first planted tobacco — poor in quality and hard to sell. In the late 1630s the Dutch brought sugar technology and credit from Brazil, and the planter James Drax (c. 1609–1662) was among the first to switch to cane. Sugar's decisive property is not sweetness but that it must be processed to a factory rhythm: cane not crushed and boiled within hours of cutting is ruined. That physical constraint forced a labor organization that was continuous, synchronized and coercible.

Sugar consumed land and labor together: between 1640 and 1660 smallholders were bought up into large plantations, and the number of trafficked Africans rose from a few hundred to over twenty thousand, surpassing the white population by the 1660s. In 1661 the Barbados assembly passed the Act for the Better Ordering and Governing of Negroes, redefining Africans from "servants" into heritable, mortgageable chattel. The code was then copied clause by clause into Jamaica (1664), Antigua and South Carolina (1691) — what got exported was not sugar technology but this legal template. Sidney Mintz's Sweetness and Power (1985) supplies the other end: sugar became cheap calories for workers, and sweetened hot tea let factory hands take in energy and stay alert in minimal time. Two labor regimes on opposite shores of the Atlantic, buckled together by one commodity.

Had Barbados stayed with tobacco, sugar capital would have flowed to Jamaica or Saint-Domingue — the institution would not have vanished, only changed address. The genuinely forking step was legal: writing bondage as a heritable property right tied to skin color was not deduced from the physics of cane. The debate still circles Eric Williams's twin theses in Capitalism and Slavery (1944) — that slave-trade profits financed the Industrial Revolution, and that abolition followed West Indian decline. Seymour Drescher's Econocide (1977) refutes the second with trade data: around 1807 the British Caribbean was at a profit peak, and abolition came out of political and moral mobilization.

"Crush the cane the same day" defined the human working day; in today's algorithmically scheduled warehouses and delivery fleets, what sets the pace is likewise the system's tempo rather than an employer's will — and once the tempo is fixed, institutions grow into the shape it requires.

Economic pressure decides which kind of institution will be wanted; the version actually written into law is always chosen by people.
Who sets the tempo in your system — and what is it quietly deciding about who must obey whom?
EVENT · 03

The Tuber That Fed Millions, and the Ones It StarvedThe Potato: Population and Famine · 1756–1852

1756Prussia / Ireland1845–1852

Brought from the Andes via Spain, the potato was long distrusted in Europe (a nightshade, and absent from Scripture). The hard numbers overrode the prejudice: two to four times the calories per acre of grain, viable on poor cold soils, and grown underground — hard for a passing army to requisition or burn. From 1756 Frederick II of Prussia issued a series of "potato orders" compelling cultivation.

Nathan Nunn and Nancy Qian, writing in the Quarterly Journal of Economics in 2011, used local suitability for potato cultivation as an instrumental variable and attributed roughly a quarter of world population growth and a quarter of urbanization between 1700 and 1900 to the potato — the caloric surplus released people from the land, and industrial labor came from there. The bill arrived in 1845: about a third of Ireland's population lived almost entirely on potatoes, concentrated in a single variety, the Lumper — clonally propagated, so the island's crop was near-genetically identical. Phytophthora infestans landed in autumn 1845 and took the whole 1846 season. Between 1845 and 1852 roughly a million people died and another million emigrated; the population fell by a quarter. Throughout the famine, Ireland went on exporting grain and livestock to Britain.

"The tenants should have planted more varieties" does not hold: under subdivided holdings and rent pressure, only the highest-calorie-per-acre Lumper kept a family alive to the next year — diversification is a strategy for those with slack. What is genuinely counterfactual is policy: in 1846 the Assistant Secretary to the Treasury, Charles Trevelyan, halted the Peel government's food purchases in favor of laissez-faire public works. Cormac Ó Gráda (Black '47 and Beyond, 1999) argues a different relief regime could have changed the death toll by an order of magnitude, yet classes the failure as ideologically driven dereliction rather than deliberate extermination; others (Tim Pat Coogan among them) press a genocide charge. The real question underneath: can motive for inaction be inferred backward from its consequences?

Monoculture is a single point of failure. Today's counterparts are the Cavendish banana facing TR4, and one base package, one cloud availability zone, one model architecture. Efficiency rewards homogeneity, robustness demands variety, and the two cannot be maximized at once.

Betting a system on the single optimal answer buys efficiency in normal times with fragility in catastrophic ones.
Which "everyone uses it" choice in your stack is precisely the point at which everyone fails together?
EVENT · 04

A Penny for a Seat at the TableLondon Coffeehouses · 1652–1675

1652London29 Dec 1675

In 1652 Pasqua Rosée, servant to a Levant merchant, opened London's first coffeehouse in St Michael's Alley off Cornhill; by the 1660s the city had hundreds. The rule was simple: a penny at the door, and anyone could sit down and join the conversation — what it sold so cheaply was not coffee but access to information unsorted by rank.

Different houses quickly specialized into different information markets: Lloyd's on Tower Street (1686) drew shipowners and underwriters and grew into Lloyd's of London; Jonathan's in Exchange Alley took in the stockjobbers expelled from the Royal Exchange and grew into the London Stock Exchange. But concentrated information is also a hazard: on 29 December 1675 Charles II issued a Proclamation for the Suppression of Coffee Houses, calling them places where malicious and scandalous reports were spread. Merchants and public opinion pushed back hard, and the proclamation was withdrawn on 8 January 1676, before it took effect — it lasted ten days.

Why did the ban die so fast? The binding constraint was enforcement cost: hundreds of scattered private premises would each have to be policed, while the coffee duty was itself Crown revenue — the ban contradicted the treasury. The retreat was not royal enlightenment but arithmetic. The historiographical fight aims straight at Habermas's "bourgeois public sphere": Brian Cowan's The Social Life of Coffee (2005) shows coffeehouses were far from egalitarian — women were largely excluded and most houses had de facto in-groups. Their real contribution may lie not in democracy but in lowering the cost of verifying strangers: underwriting, share dealing and scientific refereeing all need a place where you meet the same people repeatedly.

Lloyd's model is now any platform that gathers counterparties in one place; the 1675 proclamation is structurally the same problem as today's platform governance — the returns and the risks of aggregated information come from the identical source.

A venue's value lies not in what it sells but in which otherwise-unacquainted people it seats together, repeatedly.
Which "table" supplies your highest-value information? If it vanished, how long to rebuild input of equal quality?

Four Crops: Where the Rent Was Locked

The fork depends on where scarcity is locked: in geography, in institutions, or in a network.
Crop / Scene
What the empire did
Where the rent sat → outcome
Nutmeg · Banda 1621
Clear the natives, own the source
Geography → void the moment seedlings leak out
Sugar · Barbados 1661
Legislate people into chattel
Institutions → copyable, travels with capital for 200 years
Potato · Europe 1756–1845
Push one high-yield variety
Yield → efficiency bought with homogeneous fragility
Coffee · London 1652
Try to suppress the meeting place
Network → policing costs exceed gains; ban dead in ten days

Going Deeper

One — Cross-case analogy: nutmeg's monopoly held the source, sugar's held a labor regime. Why did the second last so much longer?
Nutmeg's rent was locked to geographic coordinates and could be routed around by transplantation: in the 1770s the Frenchman Pierre Poivre smuggled nutmeg seedlings to Mauritius, and the monopoly began leaking. Sugar's rent was locked into a replicable regime of labor and law, rebuilt wherever the capital went. Ask of any barrier: is it locked to a coordinate, or to a transferable organizational capability? The first gets bypassed; the second propagates itself.
Two — The long wave: why did each of these foods exert its real influence only after it stopped being scarce?
In the luxury phase, effects are confined to elite consumption and trading profit — small in magnitude, large in visibility. Once the price drops into an everyday budget, the good starts rewriting caloric structure, working hours and social venues: the system's infrastructure layer. Technology behaves the same way. When compute is scarce it changes who can train a model; only when it is cheap enough to be ordinary does it change how work itself is organized.
Three — Decision science: Coen's massacre, the 1661 code, Trevelyan's relief policy were each backed by the "rationality" of their moment. What mechanism blocks such decisions in advance?
All three share one structure: the decider and the bearer of the cost are fully separated, with no channel folding the cost back onto the decider's own ledger. Retrospective accountability only takes effect in the next round. What works ex ante is moving the cost into the same books — pricing externalities, seating the affected party in the decision, and using reversible procedure instead of one-shot discretion. What has ever worked historically is not moral suasion but changing who pays for the outcome.