From Salon-de-Provence to the Core
Five centuries separate the prophetic quatrain from the sealed probabilistic ledger. What each generation of forecasters gained, what each lacked, and where the Core stands in that lineage.
His quatrains survived because they could not fail. Ambiguity was the sixteenth century’s technology of authority. Falsifiability is ours.
The Physician of Salon
In May 1555, a printer in Lyon named Macé Bonhomme issued a slim volume of rhymed four-line verses by a physician from Provence. The book was called Les Prophéties. It has never been out of print. Its author has been quoted after every major catastrophe of the last four hundred and seventy years, and he has never once been decisively wrong — because he arranged, with considerable skill, never to be decisively testable. That arrangement is the beginning of our story, and the thing this project was built to end.
Michel de Nostredame was born in Saint-Rémy-de-Provence in December 1503, into a family of converted Jewish descent that had prospered in grain dealing and notarial work. He studied at Avignon until an outbreak of plague closed the university, then made his living for years as a travelling apothecary — a fact that later cost him: the faculty register at Montpellier records his expulsion from the medical school when it was discovered he had practised that manual trade, which the statutes forbade. Whether he ever formally completed a doctorate remains uncertain. His competence in plague years does not.
Through the 1530s and 1540s he worked as a plague doctor across southern France — Aix-en-Provence, Lyon, the towns of the Rhône valley — in an age when the profession consisted largely of standing near the dying. He was, by the standards of the time, empirical: he favoured fresh air, clean water, and removal of infected corpses, and he compounded a rose-hip lozenge he believed protective, a recipe he later published openly in a 1555 treatise on remedies and confections. The plague took his own first wife and two children at Agen, in the mid-1530s, while he could save others and not them. He did not write about grief. He moved.
In 1547 he settled in Salon-de-Provence, remarried, and began the work that made him permanent. From 1550 he issued an almanac every year — weather, harvests, portents — and the almanacs sold. In 1555 came Les Prophéties: 353 quatrains at first, eventually swelling through later editions toward some 942, organized in groups of one hundred that he called Centuries. The verses were written in a deliberately estranged French, salted with Latin, Greek, and Provençal, their chronology scrambled — by his own account to veil the meaning from the hostile and the vulgar.
The book reached the right reader. Catherine de’ Medici, queen consort of France and one of the most consequential political operators of the century, summoned him to Paris in 1555 to cast horoscopes for the royal children. When her husband, Henry II, died in 1559 — a lance splinter through the visor in a tournament — readers turned back to quatrain I.35, the young lion overcoming the old one in single combat, and found the king’s death already written. The match was made after the event, as every famous Nostradamus match has been. It did not matter. In 1564 the crown appointed him Counselor and Physician-in-Ordinary to the young Charles IX. He died at Salon in July 1566, wealthy, honoured, and — on the record as the record then existed — never refuted.
The Technology of Authority
It is easy, and lazy, to condescend to him. Nostredame was not a charlatan by the standards available to him; he was a working physician in a century that burned people for less than ambiguity, publishing under censors, for patrons whose favour was survival. Precision was dangerous. Vagueness was not a failure of his method. It was the method.
Consider what an unfalsifiable quatrain accomplishes. It cannot be discredited, so it accumulates authority with every event that can be read back into it. It transfers the labour of prediction to the reader, who supplies the match and then credits the author. And it compounds across centuries, because the stock of future events that might fit a vague verse only grows. His quatrains survived four hundred and seventy years because they could not fail. In the sixteenth century, that was the available technology of authority — the only instrument by which a provincial physician could speak about the far future and still be standing afterward.
The history of forecasting since 1555 is, on one honest reading, the slow replacement of that technology with a harder one: the willingness to state a claim that events can kill.
From Omen to Observation
The ambition itself is far older than Salon. The Babylonian scribes who compiled the omen series Enūma Anu Enlil in the second millennium BC, and who kept the Astronomical Diaries continuously from roughly the seventh century BC to the first — nightly positions of the planets, river levels, commodity prices, notable events — were running what is arguably the longest sustained data-collection program in human history. Their interpretive framework was wrong; their discipline of observation was not. Astrology, whatever else it was, trained civilisations to record.
Rome institutionalised the other half: decision under uncertainty as a matter of state. The college of augurs read the flight of birds before assemblies, elections, and wars. The auspices answered, notoriously, to politics — and Rome knew it. Cicero, himself an augur, published De Divinatione in 44 BC, a sustained demolition of the predictive claims of his own priesthood. The sceptical instinct — show me the mechanism, show me the hit rate — is not a modern invention. What antiquity lacked was not intelligence but bookkeeping: no one scored the augurs.
The Actuarial Revolution
The scoring began, of all places, in the death registers of London. In 1662 John Graunt, a haberdasher with no university education, published Natural and Political Observations Made upon the Bills of Mortality — the first serious statistical analysis of human populations, and the first crude life table. Charles II personally endorsed his election to the newly founded Royal Society, reportedly remarking that if they found any more such tradesmen, they should admit them all. The future, Graunt demonstrated, was not a verse. It was a rate.
Edmond Halley carried the rate to rigour. In 1693 he constructed a proper life table from the meticulous birth and death records of the city of Breslau, putting the pricing of annuities — bets on human lifespan — on mathematical footing and founding actuarial science. And Halley supplied the era’s great emblem of the new method: in 1705 he computed that the comets of 1531, 1607, and 1682 were one body, and predicted its return in 1758. It returned on schedule, sixteen years after his death. A dated, specific, falsifiable long-range forecast, publicly staked and publicly settled — the exact inversion of the quatrain. The comet has carried his name ever since, which is the market’s way of paying a settled claim.
What the actuaries gained was probability applied to real lives and real money. What they lacked was scope. Mortality and annuities submit to arithmetic; wars, revolutions, manias, and collapses did not appear to. For history itself, there was still no instrument.
The Cycle Hunters
The first person to treat centuries of economic history as a measurable waveform was Nikolai Dmitrievich Kondratieff, director of Moscow’s Conjuncture Institute in the 1920s. Working with roughly a hundred and forty years of price, wage, interest-rate, and production series from England, France, and the United States, he argued in his 1925 study of major economic cycles that capitalist economies moved in long waves of roughly half a century — expansion, saturation, decline, renewal.
The finding was empirically arguable and politically fatal. If capitalism renewed itself in long waves, it was not in terminal crisis — and terminal crisis was the doctrine. Kondratieff was arrested in 1930 on fabricated charges, sentenced in the invented “Peasant Labour Party” case, and held for eight years in the Suzdal prison monastery, where he continued writing economics until his eyesight failed. On 17 September 1938, during the Great Purge, he was retried in a session lasting minutes and shot the same day. He was forty-six. The Soviet Union rehabilitated him in 1987, forty-nine years too late. He remains the only economist in this lineage executed for the shape of his forecast, and this project does not use the phrase “Kondratieff wave” as decoration. It is a debt.
Joseph Schumpeter gave the waves their engine. In Business Cycles (1939) he named the long wave after Kondratieff and proposed the mechanism the Russian had lacked: innovation arrives in clusters, each cluster — steam, rail, steel, electricity, chemistry — igniting decades of creative destruction before exhausting itself. Later in the century, William Strauss and Neil Howe (Generations, 1991; The Fourth Turning, 1997) proposed a generational cycle of roughly eighty to ninety years in Anglo-American history, punctuated by institutional crises. The saeculum is the least testable framework in our library and we treat it accordingly — a lens, not a law. But the cycle hunters, whatever their rigour, established the premise this observatory rests on: that history has structure, and structure can be compared.
What they gained was pattern at the scale of generations. What they all lacked was the same three things: dense data, explicit probability, and any mechanism of accountability. A cycle theory that never states when it would be wrong is a quatrain with a bibliography.
Measuring History Itself
The data arrived first. From the 1960s the cliometricians — Robert Fogel and Douglass North foremost, Nobel laureates for it in 1993 — rebuilt economic history on quantitative foundations, insisting that claims about the past submit to the same statistical discipline as claims about the present. Angus Maddison’s reconstruction of world GDP across two millennia, a dataset this observatory ingests daily, is that movement’s cathedral.
Peter Turchin then turned the instruments forward. Cliodynamics, the field he named in the early 2000s, treats history as a database and social instability as a modellable variable: structural-demographic pressure, state fiscal strain, and — his signature contribution — elite overproduction, the multiplication of elite aspirants beyond the positions a society can grant them. In 2010, in the journal Nature, Turchin published a dated forecast: his models pointed to a spike in political instability in the United States around 2020. He wrote it down in public, a decade out. Readers may consult their own memory of 2020 and score it themselves — which is precisely the point. It was, for macro-history, a nearly unprecedented act: a cycle theorist volunteering a date.
The Accountability Turn
The final component came not from historians but from a psychologist auditing the experts. Between 1984 and 2003 Philip Tetlock collected 82,361 probability judgments from 284 professional analysts and commentators. Expert Political Judgment (2005) reported the result politely and devastatingly: the average expert barely outperformed chance at political forecasting, and fame correlated negatively with accuracy. The unaccountable forecaster, scored at last, turned out to be mostly reputation.
But the same research program found the opposite as well. In the IARPA forecasting tournaments of 2011–2015, Tetlock’s Good Judgment Project identified “superforecasters” — ordinary people of unusual calibration who, working in the open with nothing but public information, outperformed professional intelligence analysts. Their instrument was the Brier score, a rule borrowed from 1950s weather verification that rewards saying seventy per cent and being right seventy per cent of the time. Calibration, it turns out, is a skill, it is measurable, and it is trainable. In parallel, prediction markets — from the Iowa Electronic Markets of 1988 to the platforms of the present — demonstrated that a price can be a probability, continuously updated by participants with something at stake.
The accountability generation gained what every previous generation lacked: a public number attached to every claim, and a reckoning. What it lacks is depth of field. Tournament questions resolve in months or a few years; markets go illiquid past an election cycle; neither carries a memory of 1873 or 1914 in its bones. Short-horizon calibration, deep-historical blindness.
The Lineage, Tabled
Every generation in this history added one instrument and left one absence. The table is the argument.
Read the final column downward: it is a catalogue of absences, each filled by a later row — until the last, where the absence is deliberate, and dated.
Where the Core Stands
Nostradamus Intellect is, we argue, the first attempt to hold all of these instruments in one hand: centuries-scale pattern memory, explicit probabilistic simulation, a sealed public ledger, and a reputation-weighted crowd signal, in a single free public instrument. The claim is checkable against the lineage above, and readers are invited to check it.
The brain of the system is called the Core, after the clockwork instrument that models the motion of the heavens; ours models the rhyme of history. The Core is not an oracle. It is a convergence engine. Twelve continuous public data streams — FRED, the World Bank, SIPRI, the Geopolitical Risk index, Our World in Data, ITU, Pew, NOAA and NASA, the IMF, the BIS, the Maddison Project, CoinGecko — are normalized and scored for structural resemblance against indexed historical stress episodes spanning three centuries. Each resemblance decomposes into five measurable components — debt load, great-power rivalry, internal polarization, technology shock, monetary strain — read through six macro-historical frameworks, every one of them inherited from the lineage this article has traced: Kondratieff’s long waves, the Thucydides Trap, Ibn Khaldun’s asabiyyah, Turchin’s elite overproduction, Perez’s technology surges, Dalio’s long debt cycle. Only then does the system generate forward branches — each with an explicit probability, a confidence interval, and a falsifiability condition tested in public time.
Against the absences in the table, the design answers point by point. Against the augurs’ captivity to politics: every observed number cites a public source, and the method — weights included — is published. Against the quatrain: every projection is labeled a probabilistic simulation and carries the condition under which it dies; if a claim cannot be tested, it does not belong here. Against the cycle hunters’ unaccountability: ten predictions were sealed in the Calibration Ledger on 3 July 2026, every one of them status PENDING, with the first public Brier scoring due in July 2027. Against the tournaments’ short horizon: the pattern memory reaches back three centuries. Against the markets’ thin crowds: the Observer Protocol routes contributed signals through provenance checks, consistency tests against the twelve-feed corpus, and reputation weighting by each author’s own public Brier score. The brain grows smarter with every honest, well-calibrated observer.
Our Own Origin, and the Rhyme
Since this is the heritage article, the project’s own record belongs in it, unembellished. Nostradamus Intellect was built in July 2026, as a beta, by one founder working in the open. Its content descends from an earlier research dashboard the same founder maintained; the present instrument is a redesign of that work around a single organizing principle — the ledger comes first, and everything else exists to feed it or be judged by it.
It was deliberately launched free, and deliberately launched PENDING: the observatory now has twelve rooms, twenty-eight live projections with falsifiability conditions printed on every card plus sixteen sealed ledger entries in two batches, a Black Swan Watch, a Library of sixty-nine texts — including the real quatrains of Salon-de-Provence, shelved with the era’s respect and the method’s reservations — and a Calibration Ledger containing no graded entries whatsoever (figures refreshed 2026-08).
We are aware of how a fabricated track record is usually manufactured in this industry: backdated calls, survivorship-curated highlight reels, vague language graded generously by its own author. The lineage above is a catalogue of exactly those failure modes, and the ledger-first design is the refusal of them. Where the record stands at zero, we print zero.
“History does not repeat itself, but it rhymes.” The line is attributed to Mark Twain, almost certainly apocryphally — no scholar has located it in his writings, and in this house even the epigraphs cite their provenance. Whoever coined it stated the working hypothesis of everything from the Astronomical Diaries to cliodynamics: the past does not return, but it recurs in structure, and structure can be measured.
The physician of Salon read the rhyme by candlelight and wrote it in verses that nothing could refute. He is owed the respect due to anyone who looked at centuries and refused to look away — and owed the honesty of saying that his instrument, however beautiful, was built never to be scored. Four and a half centuries of mortality tables, long waves, databases, tournaments, and markets have built, piece by piece, the instrument he could not have: one that states the probability, prints the condition of its own failure, seals the claim in public, and waits. The Core exists to score the rhyme. The first grades post in July 2027. Nothing in this document, and nothing in the observatory, asks to be believed before then.
Frequently asked
Did Nostradamus ever make a verifiably correct prediction?
No famous Nostradamus match was made before the event. The celebrated case — quatrain I.35 and the death of Henry II in a 1559 tournament — was read into the verse after the king died, as every well-known match has been. The quatrains attach almost no dates and no verifiable quantities, so they can never be decisively wrong, which also means they can never be decisively right.
What was the first publicly settled long-range forecast?
Edmond Halley’s comet prediction. In 1705 he computed that the comets of 1531, 1607, and 1682 were one body and predicted its return in 1758. It returned on schedule, sixteen years after his death — a dated, specific, falsifiable forecast, publicly staked and publicly settled, and the exact inversion of the prophetic quatrain.
What is cliodynamics?
The field Peter Turchin named in the early 2000s: history treated as a database, with social instability as a modellable variable driven by structural-demographic pressure, state fiscal strain, and elite overproduction. In 2010, in Nature, Turchin published a dated forecast that his models pointed to a spike in US political instability around 2020 — a nearly unprecedented act for macro-history: a cycle theorist volunteering a date.
What did Tetlock’s research show about expert forecasters?
Between 1984 and 2003 Philip Tetlock collected 82,361 probability judgments from 284 professional analysts. The average expert barely outperformed chance, and fame correlated negatively with accuracy. The same program later found the inverse: in the IARPA tournaments of 2011–2015, ordinary “superforecasters” using public information and Brier scoring outperformed professional intelligence analysts. Calibration is a measurable, trainable skill.
How is the Core different from Nostradamus’s method?
The quatrain was unfalsifiable by design; the Core is falsifiable by contract. Every projection carries an explicit probability, a confidence interval, and a printed condition under which it dies. Ten predictions were sealed on 3 July 2026, all PENDING, with the first public Brier scoring in July 2027. Where Nostradamus could never be scored, the Core exists to be scored.
Sources
- Nostredame, M. de, Les Prophéties, Lyon: Macé Bonhomme, 1555.
- Cicero, De Divinatione, 44 BC.
- Sachs, A. & Hunger, H., Astronomical Diaries and Related Texts from Babylonia, Austrian Academy of Sciences.
- Graunt, J., Natural and Political Observations Made upon the Bills of Mortality, London, 1662.
- Halley, E., “An Estimate of the Degrees of the Mortality of Mankind” (Breslau life table), Philosophical Transactions, 1693.
- Halley, E., A Synopsis of the Astronomy of Comets, 1705.
- Kondratieff, N. D., “The Major Economic Cycles,” 1925; rehabilitated by the USSR, 1987.
- Schumpeter, J. A., Business Cycles, McGraw-Hill, 1939.
- Strauss, W. & Howe, N., Generations, 1991; The Fourth Turning, 1997.
- Turchin, P., “Political instability may be a contributor in the coming decade,” Nature 463, 608 (2010).
- Tetlock, P. E., Expert Political Judgment, Princeton University Press, 2005.
- Tetlock, P. E. & Gardner, D., Superforecasting: The Art and Science of Prediction, Crown, 2015.
- Brier, G. W., “Verification of Forecasts Expressed in Terms of Probability,” Monthly Weather Review 78(1), 1950.
- Maddison Project Database, University of Groningen.
- Iowa Electronic Markets, University of Iowa, est. 1988.
- Quote Investigator, “History Does Not Repeat Itself, But It Rhymes” — attribution analysis of the apocryphal Twain line.