All podcasts / People I (Mostly) Admire / Summary

31. Peter Leeson on Why Trial-by-Fire Wasn't Barbaric and Why Pirates Were Democratic

2026-07-18 - source - Read full transcript
Steve Levitt (host)Peter Leeson

Key insights

Medieval trial-by-fire ordeals functioned as a rational, self-revealing signaling mechanism rather than superstitious cruelty.
Leeson's theory: if defendants believed an omniscient God determined the ordeal's outcome, a guilty defendant rationally expected to be burned and so preferred to confess or settle beforehand, while an innocent defendant rationally expected divine intervention and accepted the ordeal. Belief in God's judgment, not the physical test itself, did the sorting.
hidden-rationality-of-weird-institutions
Priests had ample, well-documented opportunity to rig ordeal outcomes, and the church's own doctrine gave them cover to do so.
Priests and defendants were alone together while the ordeal fire was built, letting priests keep it only moderately hot, and instructions to 'sprinkle' holy water on the burned hand could easily become dousing. The Catholic doctrine of in persona Christi (God works through the priest's hands) let priests rationalize manipulation as literally channeling God's will rather than committing fraud.
hidden-rationality-of-weird-institutions
Real ordeal-register data from 13th-century Hungary is consistent with rigging: about two-thirds of the 208 defendants who actually underwent the red-hot-iron ordeal came away unscathed, and 100 of 308 total cases never reached the ordeal at all (likely confessions or plea-bargain-like settlements).
Leeson treats the high exoneration rate among people who chose to undergo the ordeal as evidence the sorting mechanism worked as predicted: those who accepted the ordeal disproportionately turned out to be the ones priests believed (or wanted to believe) were innocent.
incentives-and-self-selection
Ordeals were reserved for Christians and never applied to Jews, a secondary prediction that survives scrutiny and rules out simpler alternative explanations.
Because the mechanism only works if the defendant shares belief in an omniscient, judging God, Leeson's theory predicts legal systems would use alternative methods (oath-swearing, witnesses) for populations lacking that faith. That is exactly the historical pattern for Jewish defendants in medieval Europe.
incentives-and-self-selection
The near-total absence of women from cold-water ordeals, versus their frequent use of hot ordeals, is a signature prediction that alternative theories cannot easily explain.
In the cold ordeal, sinking indicated innocence and floating indicated guilt, but higher female body fat makes floating physiologically more likely regardless of guilt. Historical records show priests overwhelmingly steered men toward cold ordeals and women toward hot ones, matching what a rigging-aware priest would do to avoid systematically convicting innocent women.
incentives-and-self-selection
Modern polygraph lie-detector tests run on the exact same logic as medieval ordeals: scientifically bogus as a truth-detection technology, but functional as a self-revealing signaling device because people believe they work.
Leeson argues the polygraph's continued use by the FBI, military, CIA, and private employers despite courts excluding it as evidence isn't irrational: belief in its efficacy makes people with less to hide more willing to take it, reproducing the same self-selection dynamic that made ordeals effective centuries earlier.
incentives-and-self-selection
Pirate crews in the late 17th and early 18th centuries organized as constitutional democracies precisely because they, not absentee capitalists, owned the ship and shared its plunder.
Merchant and naval ships were owned by landed capitalists who installed autocratic captains to control hired sailors with no ownership stake (an agency problem). Pirates jointly stole their ships and split proceeds via a share system, giving every crew member a direct stake in the venture's success and no need for an externally imposed hierarchy.
piracy-as-proto-democracy
Pirates solved Madison's 'paradox of power' - granting authority while constraining its abuse - through democratically elected officers, divided authority between captain and quartermaster, and written constitutions, more than half a century before America's founders.
Pirates couldn't rely on government enforcement since they were criminals, so crews wrote their own constitutions banning theft and violence among themselves, elected their captain and quartermaster, and split power between a wartime military authority (captain) and a peacetime governance role (quartermaster), reserving the captain's autocratic authority strictly for combat situations.
piracy-as-proto-democracy
It is at least plausible, though unproven, that America's founders were exposed to pirate governance models before drafting the Constitution.
Pirate constitutions were published and widely read in the early-18th-century bestseller 'A General History of the Pyrates.' Leeson notes Madison and Jefferson were voracious readers with large libraries who could plausibly have owned it, but a fire destroyed part of Madison's library catalog, leaving the question unresolved; Leeson explicitly declines to claim direct causal influence.
piracy-as-proto-democracy
Leeson holds a minority, old-school rational-choice position that behavioral economics and psychology add little to explaining aggregate social outcomes, even though he doesn't deny individuals deviate from rationality.
His argument: economics explains group-level, aggregate behavior through the interaction of individuals, so the precise mental processes any one person uses to decide don't need to be modeled correctly for the aggregate-level explanation to hold. Levitt pushes back, citing his own growing interest in anthropology, sociology, and other disciplines' insight into topics like power that economics handles poorly.
economic-imperialism-vs-other-disciplines
Good economic explanation is as much art as science: the same toolkit produces very different quality of explanation depending on who wields it.
Levitt recounts that as a young professor he and colleague Austan Goolsbee would prep economic explanations for current news before lunches with Gary Becker and Kevin Murphy, and were 'almost always wrong' not for lacking tools but for reaching for the wrong model; Becker and Murphy had an 'artful' ability to select the right tool, a skill Leeson says takes a lifetime to approach.
the-art-of-economic-explanation
A single willing gatekeeper - one editor prepared to keep reading past an offbeat opening - can be pivotal to launching a career built on unconventional topics.
Levitt, then editor of the Journal of Political Economy, nearly desk-rejected Leeson's first pirate-economics paper after one paragraph, wrote a rejection letter, then read on out of curiosity, tore up the rejection, and sent it to the toughest historian referees he could find, expecting them to hate it; the paper was published and became heavily cited, and Leeson credits it with giving him 'permission' to pursue weird topics.
the-art-of-economic-explanation

Books referenced

Media referenced

Companies

Techniques and frameworks

Summary

This episode reunites Levitt with economist Peter Leeson, whose research career is built on showing that bizarre-seeming historical and social practices are, on close inspection, rational responses to real constraints. The first half covers medieval judicial ordeals: the practice of determining guilt by having a defendant plunge an arm into boiling water or carry a red-hot iron. Leeson argues these weren't barbaric superstition but a self-revealing signaling mechanism. If defendants believed an omniscient God determined the outcome, guilty defendants rationally expected to be burned and preferred to confess beforehand, while innocent defendants expected a miracle and accepted the test. Real 13th-century Hungarian court data backs this up: roughly two-thirds of defendants who actually underwent the iron ordeal came away unburned, consistent with priests quietly rigging outcomes (controlling fire heat while alone with the defendant, "sprinkling" versus dousing holy water) to spare people whose willingness to undergo the test already marked them as likely innocent.

Levitt pushes Leeson on whether this is just a comfortable story fit to convenient data, and Leeson offers two harder-to-fake secondary predictions that hold up. First, ordeals should only work on believers, and indeed medieval legal systems exempted Jews from ordeals, using oaths and witnesses instead. Second, since women have more body fat and float more easily, a rigging-aware priest steering people toward outcomes he wants would avoid sending women to cold-water ordeals (where floating meant guilt) - and the historical record shows almost no women underwent cold ordeals. Leeson draws an explicit parallel to modern polygraph tests: scientifically worthless as lie detectors, but still used by the FBI, military, and CIA precisely because belief in their validity makes honest people more willing to take them, reproducing the same self-selection logic centuries later.

The conversation broadens into a debate about economics as a discipline. Leeson holds an unusually hardline rational-choice position, arguing economics explains aggregate social outcomes through the interaction of individuals rather than needing correct models of individual psychology, and that other social sciences (aside from evolutionary anthropology, which he considers essentially economics) lack a unifying analytical framework. Levitt pushes back, describing his own evolution from believing economics should have "the final word" on policy to valuing what anthropologists and sociologists notice that economists miss, particularly on questions of power. Both agree that applying economics well is as much art as science: Levitt recalls that he and Austan Goolsbee, prepping economic explanations before lunches with Gary Becker and Kevin Murphy, were "almost always wrong," not for lacking tools but for reaching for the wrong ones in the moment.

The second half turns to Leeson's work on pirates. He argues that pirate ships in the late 1600s and early 1700s organized as near-opposite mirror images of merchant and naval ships: instead of a capitalist-appointed autocratic captain controlling sailors who had no ownership stake (a classic agency problem), pirate crews jointly owned their plunder through a share system and had direct incentive to maximize it, which made democratic governance both viable and necessary. Pirates elected their officers, divided authority between the captain (a wartime military role) and quartermaster (peacetime governance), and wrote constitutions banning theft and violence among themselves, since as criminals they couldn't rely on government enforcement. Leeson connects this directly to James Madison's "paradox of power" - the problem of granting political authority while preventing its abuse - and notes pirates solved it with the same tools (divided power, elections, constitutions) more than fifty years before America's founders. He floats, without claiming to prove, that Madison or Jefferson could plausibly have read the bestselling "A General History of the Pyrates" before drafting the Constitution.

The episode closes on the personal and the meta. Leeson describes proposing to his now-wife Ania by hiding a marriage proposal on a bookmarked page inside a printed copy of his book The Invisible Hook, delivered in a treasure chest at a restaurant. Levitt tells the story, for the first time on record, of nearly desk-rejecting Leeson's first pirate-economics paper as editor of the Journal of Political Economy after reading only its opening paragraph, then reversing course out of curiosity and publishing what became one of his proudest and most-cited editorial decisions. Both frame the through-line of Leeson's career, and Levitt's own, as a permission structure: one willing gatekeeper reading past an unconventional opening can be pivotal for anyone doing weird, unfashionable work.

Notable Quotes

"As long as people believe that they work, then there can in fact be an important use for them in our criminal justice system." - Peter Leeson, on both medieval ordeals and modern polygraph tests

"I think basically that there are only economic answers to social questions." - Peter Leeson

"Follow your weirdness, wherever it may take you - probably within the confines of the law, mostly." - Peter Leeson

"It wasn't that we didn't have the right tools to get to the answer... we'd reach into our economic toolkit and we'd pull the wrong tool for the problem at hand." - Steve Levitt, on prepping economic explanations before lunches with Gary Becker and Kevin Murphy

"Pirates developed their system of constitutional democracy, which looks so much like our own, more than half a century before America's founding fathers put pen to paper." - Peter Leeson