All podcasts / Dwarkesh Podcast / Summary

Why Leonardo was a saboteur, Gutenberg went broke, and Florence was weird - Ada Palmer

2026-03-06 - 122 min - source - Read full transcript

Dwarkesh Patel (host)Ada Palmer

Key insights

Petrarch's project to revive Roman virtue through classical education produced a world he never intended but that still went well by his own deeper hopes.

Petrarch wanted libraries and classical education to raise leaders who resembled the noble Romans he read about. Instead it produced worse wars, the Borgias, and eventually a scientific and political culture unrecognizable to him - one he'd be horrified by (democracy) but also weep with joy over (a cure for the Black Death he watched kill his friends). Palmer's distinction is between a project going 'your way' versus going 'well.'

renaissance-classical-revival

Classical trappings - togas, Latin, Greek-speaking tutors, faux-Roman statues - functioned as manufactured legitimacy for rulers with no real claim to power.

Palmer's extended set piece has a French ambassador, primed to view Florence as 'a pit of scum and villainy' run by low-status merchants, walking into Cosimo de' Medici's courtyard full of lifelike bronze statues, Roman-emperor busts, and a ten-year-old grandson reciting Greek poetry. The display flips the power dynamic: the condescending nobleman ends up in awe of the 'merchant scum,' which is the entire point of the theater.

renaissance-classical-revival

Machiavelli abandoned the prevailing theory that reading about virtuous ancients makes rulers virtuous, proposing instead an evidence-based method of comparing historical outcomes - effectively founding political science.

After watching classically-educated princes like Cesare and Lucrezia Borgia produce bigger, bloodier wars than before, Machiavelli concluded that moral 'osmosis' from reading Cicero didn't work. His alternative: line up historical examples (e.g., battles fought near rivers), see which decisions succeeded, and imitate the choices rather than the character. Palmer notes contemporaries described him as a historian, not a philosopher, for this reason.

renaissance-classical-revival

Gutenberg, his creditor bank, and his own apprentices all went bankrupt because a mass-produced commodity existed before any distribution network could sell it beyond a single small market.

Gutenberg printed 300 Bibles in a landlocked German town where only seven people were legally allowed to read one, so he had 293 unsellable copies and went bankrupt; the bank that seized his press failed the same way. The press only became profitable once his fleeing apprentices reached Venice, the Mediterranean's shipping hub, where ten copies could be handed to each of thirty ship captains bound for thirty different cities.

printing-press-economics

The print revolution was a single continuous technological upheaval with successive applications over roughly 150 years, directly paralleling computers' path from mainframes to social media - and each new medium arose to solve the trust problem the previous one created.

Books (slow, expensive) gave way to pamphlets (fast, cheap, hard to censor, and how Luther's Ninety-five Theses reached London in seventeen days), which in turn created a fake-news problem that the newspaper's serialized reputation-based accountability solved; when newspapers started contradicting each other, The Gentleman's Magazine was invented specifically to fact-check them against one another.

printing-press-economics

Pre-print Europe's reliance on parchment (animal skin, roughly the price of a leather jacket per sheet) capped its libraries at a few hundred books while contemporaneous Middle Eastern and Chinese libraries, using cheap papyrus or rice paper, held thousands.

Europe lost access to warm-weather papyrus after the fall of Rome and had no substitute until paper arrived from Asia around 800 AD, then took another four hundred years to be trusted for official documents. The University of Paris's library, the largest in Europe at the time, held six hundred books - fewer than an airport bookstore - while Middle Eastern and sub-Saharan African libraries of the same era held thousands. This was a material bottleneck, not a literacy or curiosity gap.

printing-press-economics

Florence ran its commoner republic by sortition, drawing nine merchants at random to rule from a locked tower for short terms specifically to prevent bribery and tyranny - until Cosimo de' Medici realized he could capture the system statistically rather than abolish it.

If a third of the city's population worked for Cosimo, roughly a third of any random nine-person draw would too, and directing that plurality was often enough to control policy. When bad luck once produced a council that arrested and imprisoned him instead, he bribed his way out with the equivalent of a million dollars and returned to shift from merely controlling a plurality to bribing the people who ran the elections themselves.

florentine-republic-and-power

The Florentine and Machiavellian term 'popolo' ('the people') meant only the enfranchised top roughly 4% of the population (merchant-guild members), not the citizenry at large.

This mistranslation trap led 19th-century readers, and modern American readers who see resonance with founding-era rhetoric about arming and trusting 'the people,' to mistake a narrow oligarchic republic for something closer to democracy. Palmer says historians only caught the error after reading more of the period's actual administrative documents.

florentine-republic-and-power

A losing resistance can still permanently constrain the winner: because Florence's republic fought hard for its institutions before finally falling to the Medici, the resulting dukedom stayed meaningfully less tyrannical than confidently absolute neighbors.

When the first Medici duke wanted to demolish part of a wealthy family's centuries-old tower to build a private walkway, he built an awkward detour around it instead rather than risk provoking rebellion - a restraint the openly tyrannical Duke d'Este of Ferrara, who felt free to murder servants and kidnap rival musicians, never had to show. Palmer draws a direct parallel to modern debates about how strong US institutional resistance would remain even under a worst-case erosion of democratic norms.

florentine-republic-and-power

Whether censorship works is bounded by an intersection of legality and raw technological speed: the Inquisition could regulate printed books but never pamphlets, because pamphlet news-runners moved information faster than enforcement could travel.

Printers who employed the same news-runners the Inquisition relied on for information would hear about their own impending arrest before inquisitors arrived, and simply relocate. Palmer draws the direct modern parallel: platforms say the same thing about pornography or hate speech moderation today - 'we just can't, it's too fast, there's too much.'

censorship-and-information-control

Censors are reliably wrong, in hindsight, about which ideas are actually dangerous.

The 1545 Index of Banned Books put minor Protestant theologians' names in all caps as 'arch-heretics' while Machiavelli's name appeared in plain type, unflagged. Two centuries later, French censors ordered to ban the wildly popular, genuinely radical Encyclopedie staged a ceremonial burning of harmless Jansenist theological tracts instead, specifically to avoid destroying a book everyone from the king to the queen loved.

censorship-and-information-control

The sense that history 'sped up' after 1600 is partly a hindsight illusion, and what really changed was that knowledge production became deliberately cumulative rather than individually hoarded.

Palmer says zooming into any past decade she studies feels just as fast-moving to the people living through it as the present does; textbooks and games like Civilization impose a false 'slow past, fast present' shape by only counting technologies we still care about. What genuinely shifted after 1600 was Francis Bacon's 'ant, spider, bee' framework displacing secretive artisan-inventors - Palmer calls Leonardo da Vinci a progress 'saboteur' for encoding his discoveries in mirror-writing specifically so no one else could use them - with an ethic of publishing results so knowledge compounds across a community.

pace-of-technological-change

Books referenced

Inventing the Renaissance - Ada Palmer - The book under discussion; Palmer's history of the Italian Renaissance frames the whole conversation, from Petrarch's classical revival through Machiavelli and the printing press.
The Renaissance in Italy - Guido Ruggiero - Recommended on Palmer's website; source of the question Dwarkesh raises about why the Industrial Revolution didn't start in Italy despite its advanced metalworking, milling, and gear technology.
The Prince - Niccolo Machiavelli - Referenced as the culmination of the classical-revival project; discussed as Machiavelli's proposal to learn from historical outcomes rather than imitate virtuous ancients.
The Swerve - Stephen Greenblatt - Palmer notes it covers similar ground to her own dissertation on Lucretius, and recommends Inventing the Renaissance as a more detailed 'zoom-in' on the same material.
On the Nature of Things (De rerum natura) - Lucretius - Subject of Palmer's dissertation; its 1417 rediscovery is cited as an example of an idea sitting dormant until a critical mass of readers (30,000 within a century, via 30 print editions) could test its atomist claims and eventually help produce germ theory.
The Specter of Skepticism in the Age of Enlightenment - Anton Matytsin - Source of the account of a raid on a clandestine Enlightenment-era bookshop, used to show how badly censors misjudged which texts (Jansenist tracts, not Voltaire or the Encyclopedie) were actually dangerous.
Areopagitica - John Milton - Milton's failed treatise arguing for freedom of the press, written just before England adopted systematic pre-publication censorship.
Paradise Lost - John Milton - Published under the censorship regime Milton's own Areopagitica had failed to prevent; the one line censors flagged was about astrology, not its anti-monarchical rhetoric or unflattering portrayal of God.
Encyclopedie - Denis Diderot (with d'Holbach) - The universal-knowledge project so popular in France that when Rome ordered it banned, French censors staged a ceremonial burning of Jansenist tracts instead, so as not to actually destroy it.

Media referenced

Dwarkesh Podcast episode with Joseph Henrich - podcast - Referenced as a previous guest whose theory credits the Catholic Church's breakdown of kinship networks and promotion of guilds for enabling the Scientific Revolution; Dwarkesh raises it as a competing explanation to Palmer's classical-revival account.
Dwarkesh Podcast episode with Daniel Yergin - podcast - Referenced for Yergin's book on the history of oil; used as a parallel example of slow technology diffusion, since crude oil sat with no real market for decades before the internal combustion engine created demand for gasoline.
Civilization (Civ) - other - Palmer is writing a paper on Civ as 'the number one teacher of history in the world' (70 million copies sold); she criticizes its turn-length structure (fifty years per turn in antiquity, one year in modernity) for reinforcing the false idea that history used to move slowly.
The Gentleman's Magazine - other - Cited as the first magazine, invented in the 18th century specifically to fact-check contradictory newspaper reports against each other.
Romeo and Juliet - other - Used as an analogy for pre-Renaissance elite selfishness (Lord Montague and Lord Capulet caring about their feud, not Verona's welfare) that Petrarch's classical-revival project was meant to correct.

Companies

Medici Bank - The Florentine banking dynasty at the center of the episode; Cosimo de' Medici's role as the papacy's banker, funneling church tax revenue through his accounts, funded his takeover of Florence's lottery-based government.
Standard Oil - Referenced in Dwarkesh's aside about slow technology diffusion; its main business, kerosene for lighting, nearly went obsolete when the light bulb arrived, just before the automobile created a new market for gasoline.
Amazon - Cited via the Kindle as the example of a technology (e-books) existing for decades before a commodity distribution mechanism created an actual market for it - Palmer's parallel to the printing press.
Jane Street - Sponsor; segment on its ongoing 'backdoor puzzle' contest and summer ML internship applications.
Labelbox - Sponsor; pitched for tailoring agent training scenarios to specific domains so models learn real understanding rather than overfitting to prompts.
Mercury - Sponsor; pitched for its personal banking accounts, including a fee waiver for existing Mercury Business users.

Techniques and frameworks

Sortition (lottery-based officeholding) - Florence's method for selecting its ruling council: nine merchants drawn at random from a bag and locked in a tower for a two-to-three-month term, designed to prevent bribery and tyranny; Palmer notes it is an old, generally effective form of government also used in ancient Athens.
Bacon's ant, spider, and bee - Francis Bacon's framework distinguishing two failure modes of knowledge-seeking - the ant who only hoards facts, the spider who spins ungrounded theory - from the honeybee, who processes real-world observation into something useful for humanity; the rhetorical basis for founding England's Academy of Sciences.
Podesta system - Florence's practice of hiring an outside nobleman for a fixed term to serve as police chief and judge, since no local commoner had the noble blood needed for people to accept arrest orders, then banishing him for life to prevent him exploiting the position afterward.
Casebook method of learning from history - Machiavelli's alternative to Petrarch's 'osmosis' theory of moral education: instead of assuming that reading about virtuous ancients makes you virtuous, place historical examples side by side, identify which choices worked, and imitate those - the seed of what Palmer calls political science.

Summary

Dwarkesh Patel's first conversation with Ada Palmer (Renaissance historian, novelist, and composer at the University of Chicago) is built around her book Inventing the Renaissance, and it opens with the biggest thread of the episode: what happens when a multi-generational project to reshape civilization succeeds technically but fails on its own terms. Petrarch, traumatized by the Black Death and endemic banditry, decided Europe's selfish leaders needed to be raised on Cicero and Livy so they'd rule like virtuous Romans. Two centuries of pouring money into libraries and classical tutors later, the result wasn't philosopher-kings but Cesare Borgia, bigger and bloodier wars, and eventually Machiavelli's blunt rejection of the whole premise. Palmer's framing - a project can fail to go "your way" and still go "well," since that same classical-education machinery eventually produced the science that cured the Black Death Petrarch fled - recurs throughout as her central lens on unintended consequences, and Dwarkesh repeatedly pulls it toward the present, drawing an explicit parallel to nobody having a track record of steering AI's second-order effects.

A long middle section works through Florence's genuinely strange government: a commoner republic that massacred its own nobility, then ruled by sortition, drawing nine merchants at random from a bag to govern from a locked tower for two-to-three-month terms specifically to prevent bribery and capture. Palmer traces how Cosimo de' Medici captured this tyrant-proof system not by breaking it but by exploiting its statistics - enough of the city worked for him that a plurality of any random draw did too - and how the term "popolo" in Machiavelli's writing meant the enfranchised top 4%, not the citizenry, a mistranslation that made later readers see false democratic resonance in what was really an oligarchic republic. A striking detail closes this thread: even after the Medici formally took over, Florence's history of hard-fought resistance meant the resulting dukes stayed meaningfully less tyrannical than confidently absolute rulers elsewhere, respecting property rights (down to routing a private corridor around one family's ancestral tower) in a way neighboring Ferrara's duke, who wandered his city naked with a sword to prove nobody would dare harm him, never had to.

The episode's second half turns to information technology and its economics, drawn out in vivid, physical detail (Palmer produces actual samples of papyrus, parchment, and printed pamphlets during the recording). Gutenberg's famous bankruptcy, she argues, wasn't a personal failure but a distribution problem: he had a mass-production technology in a world with no mass-distribution network, and only turned profitable once his fleeing apprentices reached Venice's shipping hub and could scatter small print runs across dozens of cities. From there, Palmer builds out the case that the print revolution wasn't one event but a 150-year sequence of successive applications - books, then pamphlets, then newspapers, then magazines - each solving a trust problem the last one created, an argument she draws as an explicit parallel to computers moving from mainframes through PCs, the internet, phones, and social media as one continuous revolution rather than several discrete ones.

This economics-of-information thread connects directly to censorship: pamphlets proved permanently uncensorable because pamphlet-runners moved news faster than any Inquisition enforcement apparatus could travel, a speed asymmetry Palmer compares directly to platforms today saying they simply can't keep up with the volume of content needing moderation. She's equally interested in how consistently censors misjudge danger in hindsight - 1540s inquisitors obsessing over minor Protestant theologians while ignoring Machiavelli, and 18th-century French officials staging a ceremonial burning of harmless Jansenist tracts specifically to avoid burning the far more radical Encyclopedie everyone from the king down actually loved. The episode closes on the theme that pace itself is often an illusion: Palmer argues any decade she studies closely felt just as fast to the people living it as ours does now, and that what genuinely changed with the scientific revolution wasn't the pace of invention but whether inventors like the secretive, code-writing Leonardo da Vinci (whom she provocatively calls a progress "saboteur") shared their discoveries or hoarded them.

Notable Quotes

"Petrarch thought he would make a world which shared his values. Instead, he made a world that doesn't share his values but is capable of curing a disease he never imagined would be curable." - Ada Palmer

"What does Leonardo do? He writes everything he discovers down in coded mirror writing so that nobody but him can possibly use it... This is actually a saboteur of human progress, if anything, who deliberately makes progress and then tries to cut it off at that point so that no one else can be his peer." - Ada Palmer

"You have the Industrial Revolution when you feel you need to. That's the thing about Gutenberg as well that a lot of people don't think about. People are like, 'Gutenberg was an inventor and invented a thing, and then it had an impact.' No." - Ada Palmer

"But an era is always wrong about what ideas and what circulation and what changes are the really big ones." - Ada Palmer

"It's important to remember they were nobody when this set up. They were a minor important family." - Ada Palmer, on the Medici