EVENT · 01
The Captain Could Be Voted Out, and Often WasThe Pirate Articles · Bartholomew Roberts, 1721
1716–1722Caribbean · West AfricaRediker · Leeson · Johnson 1724
Background and Key Figures
The War of the Spanish Succession ended in 1713, throwing tens of thousands of privateersmen and navy sailors out of work overnight. The merchant-ship order they knew ran like this: the captain held near-absolute power over wages, rations and the lash, and once at sea no institution could check him. Bartholomew Roberts, a Welshman, was third mate on a slave ship until pirates captured him in 1719 — six weeks later his captors elected him captain.
What Happened
His articles survive because the 1724 General History of the Pyrates (published under the name Captain Charles Johnson) copied them down, and they are startlingly specific: every man has a vote in affairs of the moment; of the plunder the captain and quartermaster take two shares each, the gunner and boatswain one and a half, everyone else one; a man crippled in action draws 800 pieces of eight from the common stock first. Two provisions are the real machinery: the captain wielded absolute command only in chase and battle, while the division of spoils and all discipline fell to a quartermaster elected by the whole crew — and the captain could be voted out at any time.
Counterfactual + Historians' Debate
Marcus Rediker, in Villains of All Nations (2004), reads this as the world turned upside down: class revolt forced out by merchant-ship tyranny. Peter Leeson, in The Invisible Hook (2009), objects that pirates lived outside the law — a captain could neither call on state violence to punish desertion nor sue anyone in court, so divided power and written articles were simply the cheapest available solution to a principal-agent problem, and had nothing to do with ideals. The counterfactual runs into a hard constraint: the pirate economy was parasitic. All income came from plundering ships someone else had loaded; it could not produce, could not scale, and could not survive without ports that would buy stolen cargo. Woodes Rogers reached Nassau in 1718 carrying the King's pardon; on 10 February 1722 Roberts was killed by grapeshot off Cape Lopez in West Africa, and that April 52 of his crew were hanged at Cape Coast Castle. Cut off the fencing ports while opening an exit, and it was over in three years.
Modern Parallel
The benevolent dictator for life and the right to fork in open source; early-stage startups buying trust with vesting, veto rights and exit clauses. Divided power is not a virtue. It is collateral.
One-Line Lesson + Question
Where nothing can be enforced from outside, governance rests on credible commitment — and the price of credible commitment is giving power away.
In your organization, what actually constrains the person at the top: institutions, or the people they cannot afford to lose?
EVENT · 02
First Define Who Counts as a Vagrant, Then Decide Whom to HelpFrom the Ordinance of Labourers to the Settlement Act · 1349–1662
1349 · 1547 · 1601 · 1662EnglandPolanyi · Blaug · Hindle
Background and Key Figures
The Black Death of 1348–49 took thirty to forty percent of England's population. Labour turned scarce overnight and wages shot up. In 1349 Edward III issued the Ordinance of Labourers: every landless person under sixty had to accept work at pre-plague rates, and could not leave the parish to seek a better price elsewhere. These are the earliest laws in English that criminalized movement itself.
What Happened
Edward VI's Vagabonds Act of 1547 pushed it to the extreme: an able-bodied person found idle for three days was branded with a V on the chest and enslaved for two years, and for life if he ran. It was repealed within two years — local justices simply refused to enforce it. Elizabeth's Poor Relief Act of 1601 then set the framework for the next two centuries: a poor rate levied by parish, with the poor sorted into three categories — relief for the impotent, compulsory work for the able-bodied, the house of correction for those who refused. The Settlement Act of 1662 added the last piece: a parish could return anyone lacking local settlement rights to their place of origin. Relief and removal were always two faces of one system. Speenhamland tied wage subsidies to the price of bread from 1795; the New Poor Law of 1834 replaced it with the workhouse.
Counterfactual + Historians' Debate
Karl Polanyi's The Great Transformation (1944) held that Speenhamland turned relief into a hidden subsidy for employers and destroyed workers' bargaining power. Mark Blaug's 1963 essay The Myth of the Old Poor Law answered with parish accounts: the old system mainly absorbed seasonal agricultural unemployment and did not depress productivity, and the 1834 reform led with ideology rather than evidence. What if relief had been organized nationally rather than by parish? Steve Hindle's On the Parish? (2004) points at the crux: the parish's advantage was not stinginess but screening capacity — neighbours knew who genuinely could not work. A state has no such information and can only set cruder thresholds. The price was that the poor were pinned to their birthplace: security traded against mobility.
Modern Parallel
Locality-bound social insurance and public services, out-of-province medical reimbursement, points-based residency; platform gig workers whose coverage evaporates when the place they pay into and the place they work come apart.
One-Line Lesson + Question
Every welfare system must first draw a line around who belongs here — and that line turns around and restricts where people can go.
Which of your own protections is bound to one specific place?
EVENT · 03
Confess and You Are Guilty; Stay Silent and the Devil Is Helping YouBamberg's Drudenhaus and the Cautio Criminalis · 1626–1631
1626–1631Bamberg · WürzburgBehringer · Levack · Spee
Background and Key Figures
Between 1400 and 1775 roughly 90,000 people across Europe were tried for witchcraft and 40,000 to 60,000 executed (Brian Levack's estimate, far below the millions of popular legend). The peak fell between 1560 and 1630 — not in the "dark" Middle Ages but between the Renaissance and the Scientific Revolution. Wolfgang Behringer's Weather, Hunger and Fear lines that curve up against the failed harvests of the Little Ice Age.
What Happened
A late frost on 27 May 1626 destroyed the vines and grain of Franconia. Over the next five years the prince-bishop of Bamberg, Johann Georg II Fuchs von Dornheim, built a purpose-made witch prison, the Drudenhaus, and executed roughly a thousand people; neighbouring Würzburg killed some nine hundred. The accused began as poor widows and quickly reached councillors and the mayor himself, Johannes Junius — the letter he smuggled to his daughter from prison in 1628 survives: every confession was invented, he wrote, because he could not bear the torture. In 1631 the Jesuit Friedrich Spee, who had heard the confessions of many condemned prisoners, published the Cautio Criminalis anonymously. His argument was not theological but formal: under torture, silence is read as the Devil's assistance, confession as self-incrimination, and naming accomplices manufactures the next batch of defendants — a procedure that returns "guilty" for every possible input proves nothing at all.
Counterfactual + Historians' Debate
Levack's The Witch-Hunt in Early Modern Europe reaches a counterintuitive conclusion: the Inquisitions were not the great killers. The Spanish Inquisition had all but stopped executing after 1614, and the Parlement of Paris reviewed every capital sentence automatically, driving death rates sharply down. The heaviest tolls came from small territories with no appellate review. Nor did Enlightenment thinking end the Bamberg hunt: in 1630–31 the Imperial Aulic Council took up appeals, Swedish armies drew near, and the bishop fled. Behringer stresses that crop failure created demand for scapegoats; Levack stresses that procedure determined how far it could spread. Demand lights the fire; procedure decides how big it gets.
Modern Parallel
Any accusation structured so that denial is itself evidence — from trial by public opinion to internal investigations that reverse the burden of proof. The technical version is the same: a detector that scores every rebuttal as a suspicious signal has stopped being a detector.
One-Line Lesson + Question
A procedure that returns the same output for every input is not verifying anything. It is manufacturing a conclusion.
Do you run any assessment whose verdict would not change no matter what the person being assessed did?
EVENT · 04
He Stepped Forward and Said "I Started It" So No One Else Would Be ChargedGe Cheng and the Suzhou Weavers' Uprising · 1601
Wanli 29 (1601)SuzhouWu Jen-shu · Scott · Ray Huang
Background and Key Figures
From 1596 the Ming court dispatched eunuch tax commissioners across the empire to collect commercial and mining levies, bypassing local magistrates and reporting straight to the imperial privy purse. Suzhou was the silk-weaving capital of China, where workshop owners and hired weavers lived day to day on daily pay — no work today meant no food today. The commissioner Sun Long imposed a surcharge per loom and per bolt of silk and hired local toughs to man street checkpoints. Looms fell idle by the thousand, and thousands of weavers lost their income.
What Happened
In the sixth month of 1601 an ordinary weaver named Ge Cheng raised a banana-leaf fan as a signal and gathered several thousand men. The action was tightly self-limited: they struck only the commissioner's enforcers, killing the tax bully Huang Jianjie and others, and looted no shops, entered no homes, burned no government offices. After three days the crowd dispersed on its own. When the authorities began hunting for ringleaders, Ge Cheng turned himself in — I am the one who started it, he said; charge no one else. He spent thirteen years in prison, was released in 1614, and Suzhou remembered him as "General Ge," later building him a shrine.
Counterfactual + Historians' Debate
Earlier Chinese scholarship read the episode through the "sprouts of capitalism" frame, as a bourgeois revolt against feudalism. Wu Jen-shu's Provoking the Good People (2011) revises this: violence in late-imperial urban collective action carried sharp moral boundaries, and participants deliberately framed themselves as good subjects removing a menace rather than rebels, because that framing determined how the court would classify them — as James C. Scott puts it more bluntly in Domination and the Arts of Resistance (1990), open resistance from below usually wears the ruler's own language. The counterfactual is equally constrained: had Ge Cheng not confessed, killing officials in a crowd fell close to treason under Ming law, with vast collective liability; had the court answered with a massacre, Suzhou was precisely the tax and silk hub financing the Wanli emperor's three great campaigns, and the cost would have been unbearable. Both sides were doing the same arithmetic, which is why "one man takes the blame" was the equilibrium.
Modern Parallel
Self-limitation and legitimacy framing in protest; and the classic problem of who goes first in collective action — the first person carries all the risk and everyone after pays less. Internal dissent works the same way.
One-Line Lesson + Question
The weak do not win on force. They win by fitting their action inside language the other side cannot repudiate.
The last time you disagreed openly at work, were you speaking your language or theirs?
Four Marginal Groups, Four Mechanisms Laid Bare
The usual explanations talk about character; what actually drives these is structure.
Case / Date
Usual Explanation
Actual Mechanism
Pirate articles · 1721
Outlaws with no respect for rules
Without outside enforcement, credible commitment is bought with divided power
Vagrancy law · 1349–1662
People then simply lacked compassion
Define belonging first, allocate relief second: relief is mobility control
Witch-hunt peak · 1626
A product of superstition
Crisis lights it; procedure without review amplifies accusation into cascade
Suzhou riot · 1601
Oppression breeds rebellion
Legitimacy arbitrage: package resistance in official language to cap the charge
Going Deeper
Question 1: Are the pirates' divided power and the Suzhou weavers' restraint the same kind of wisdom?
Opposite in form, identical in logic: both trade self-imposed limits for room to survive. The pirates limited the captain's power to keep the crew from deserting; Ge Cheng limited the scope of the violence and the number of people liable to keep the court from a massacre. Each was predicting the opponent's response, conceding something in advance, and steering the other side toward the cheaper option. The difference is that the pirates wrote their commitment into enforceable articles, while Ge Cheng could pledge only his own life — with no institution to lean on, the price of a credible commitment is personal risk.
Question 2: Levack argues procedure explains death tolls better than belief. Where else does that transfer?
To every situation where good intentions produce runaway outcomes. Two regions holding identical beliefs about witchcraft, one with automatic review and one without, differ by an order of magnitude in deaths — so the variable driving the runaway is not what people believe but whether a correction loop exists and how long it takes. That matches the intuition from distributed systems: failures are inevitable, and what determines the system's fate is isolation and rollback, not how reliable any single node is. The question is not whether your judgment can be wrong, but which layer catches it once it is.
Question 3: Why is the documentary record on marginal people often unusually rich?
Because what recorded them was the judicial machine: trial transcripts, poor-rate accounts, admiralty court files. The nameless enter the archive only when prosecuted, relieved or registered. That creates a bias worth watching: the marginal people we can see are almost entirely those the system caught, while everyone who successfully disappeared left no trace. Any inference drawn from surviving data should begin with one question — who generated this data, and for what purpose?