The Black Death: The Pandemic That Remade Europe

The agent of the Black Death was Yersinia pestis, a gram-negative bacterium. The confirmation comes from ancient DNA extracted from plague pit burials across Europe, reported by Bos and colleagues in 2011 and by Spyrou and colleagues in 2019. This is the single strongest fact in this article and the one place where the evidence really is closed.

The foundational study has a specific address. Bos, Schuenemann, Golding and colleagues published a draft genome of Yersinia pestis from victims of the Black Death in Nature in 2011, sequenced from skeletal remains excavated at the East Smithfield burial ground in London, the emergency cemetery on the site of the Royal Mint where more than 2,000 plague victims were buried in 1348 and 1349. A correction to that paper was published in December 2011. It is a correction, not a retraction, and it does not affect the finding. Two things are worth saying plainly about the citation: our own file attaches a DOI to this paper that resolves to a post-publication review record rather than to the paper, and the article's strongest claim was therefore resting on a pointer to the wrong document. The correct identifier is used in the sources below.

The genomes also date the family tree. The Black Death strain is ancestral to most modern Yersinia pestis lineages, and our file describes the pandemic as a 'Big Bang' event for plague evolution. Spyrou and colleagues put a face on that phrase in 2022: sampling two Nestorian Christian cemeteries in the Chuy Valley near Lake Issyk-Kul in modern Kyrgyzstan, at Kara-Djigach and Burana, they recovered Yersinia pestis from the teeth of seven individuals, and the strain sits at the origin point of that diversification, immediately ancestral to the strains that reached Europe from 1346.

Our own research file lists the transmission route inside its Tier 1 verified table, in the same four rows as the ancient DNA identification: primary vector Xenopsylla cheopis, the oriental rat flea, carried by Rattus rattus, the black rat. Its summary states that the plague was transmitted primarily through flea bites from infected rats, with no hedge at all. We carry that tier here because tiers in this library come from the document, never from the writer. We also say, in the same breath, that it does not belong in that table. Nothing about the rat-flea mechanism is in doubt. Its primacy in fourteenth-century Europe is.

A second route needs no rodent at all. Pneumonic plague transmits directly between people by respiratory droplets, with no flea or rodent intermediary. That is not contested by anyone; the question is only how much of the fourteenth-century spread it accounts for.

The strongest attempt so far to decide between the routes did not choose the rats. Dean, Krauer, Walloe and colleagues, publishing in the Proceedings of the National Academy of Sciences in 2018, modelled three candidate routes against mortality curves from nine European outbreaks: rat fleas biting people, human ectoparasites (human fleas and body lice) biting people, and person-to-person pneumonic spread. The human-ectoparasite model gave the best fit for seven of the nine outbreaks, with Eyam and Givry the two exceptions, and the authors concluded that human ectoparasites were the primary vectors during the Second Pandemic, which includes the Black Death.

That model is not the last word either, and the objection to it is biological rather than statistical. Subsequent vector-competence work has found the human flea to be a poor transmitter of plague bacilli, which is a real argument against a model that needs it to be an efficient one. So the honest state of the question is three live candidates with different kinds of evidence behind them: a mechanism that is demonstrably efficient in the laboratory but fits the epidemic curves worse, a model that fits the curves better but leans on a vector that may not be up to the job, and a route that is undisputed as a mechanism and insufficient as a whole explanation. This article crowns none of them.

That blackening in the third row is where our own document makes a plain factual error, and it is worth correcting in the open precisely because it is ours. E_2_06 appends the parenthetical 'origin of Black Death' to the skin blackening. Historians reject that derivation. Fourteenth-century writers did not use the phrase at all: they called the pandemic the Great Pestilence or the Great Mortality. The name descends from the Latin atra mors, in which atra means both black and terrible, so the colour reading is most likely a mistranslation. The Montpellier-trained physician Simon of Couvin, writing while it was happening, used mors nigra with no suggestion that bodies turned black. The phrase entered print as a name for this specific pandemic only in the seventeenth century and became standard in English in the nineteenth. The clinical description in that row is sound. Only the etymology hanging off it is wrong.

The earliest evidence in the chain sits at Lake Issyk-Kul in modern Kyrgyzstan, around 1338 to 1339. The tombstones in the Chuy Valley cemeteries carry Syriac inscriptions giving the name of the deceased, the year of death and often the cause, and they record a spike of deaths attributed to pestilence in exactly those two years. Ancient DNA from the teeth beneath them supplies the organism. Documentary evidence and genetic evidence, from the same graves, agreeing.

Not every specialist reads that as the origin. The historian Monica H. Green, in 'The Four Black Deaths' in the American Historical Review in 2020, argues that the genetic divergence of Yersinia pestis into four branches happened in Central Asia at some point before the Black Death, and that the Mongols moved plague through Central Eurasia in the thirteenth century rather than the fourteenth. On that reading the Issyk-Kul outbreak is a late episode in a longer dispersal, not the starting gun. Green's argument does not contradict the genetic result so much as reframe what it is evidence of, and this article carries both without choosing. It is absent from our own file entirely.

Between 1343 and 1346 the plague spread westward along the Silk Road and the Mongol trade routes. In 1346 a Mongol army was besieging Caffa, the Genoese trading post in Crimea that is now Feodosia, and plague broke out in the besieging camp. In October 1347 Genoese ships arrived at Messina in Sicily and the plague entered Europe.

What happened at Caffa next is the most quoted episode in the whole pandemic and the least secure. Gabriele de' Mussi, an Italian notary writing around 1348, recorded that the Mongol besiegers catapulted plague-infected corpses over the walls. If the account is accurate it is one of the earliest documented instances of biological warfare. De' Mussi was not an eyewitness and his account is secondhand; our own file's judgement is that there is nonetheless no compelling reason to doubt the core narrative, which is a warmer verdict than the same file's own header gives, since that header lists the Caffa origin story as one of only two items in the entire document rated at medium confidence.

There are two deflations of the story, and our file supplies both itself. The first is mechanical: bubonic plague requires flea vectors, and contact with corpses could transmit pneumonic plague but does so inefficiently, so catapulted bodies are a poor delivery system for the disease they are supposed to have delivered. Plague may simply have entered Caffa with rats crossing the siege lines, corpses or no corpses. The second is historical: the plague would have reached Europe along the trade routes regardless, so Caffa may have accelerated European transmission rather than initiated it. The siege is a vivid scene. It is not a cause.

From Messina the sequence is well documented. During 1348 the plague moved through Italy, France and the Iberian Peninsula, and reached England by that summer. During 1349, the peak mortality year, it reached Germany, Scandinavia and Eastern Europe. Between 1350 and 1353 it reached Russia and the remote areas, and the pandemic waned. The initial wave lasted roughly six years. The recurrences continued for centuries.

For Europe overall our file gives an estimated death rate of 30 to 60 percent, adding that most estimates converge around 40 to 50 percent. Read that as a range with a method behind it rather than a count. Nobody counted the dead of fourteenth-century Europe. Every figure in the table below is reconstructed from something else: a tax roll, a manorial account, a churchyard, an abandoned farm, or another region's numbers borrowed and scaled.

The top of that European range has an author and a method. Ole J. Benedictow's The Black Death, 1346 to 1353: The Complete History (Boydell Press, 2004) argued for roughly 60 percent, and that figure is derived rather than measured: it rests on an estimated pre-plague European population near 80 million and an estimated 50 million dead in the first wave, with rates from the better-documented regions extrapolated onto the poorly documented ones. Benedictow later revised his own estimate upward rather than down. It is precisely that practice, predicting one region's experience from another's, that the next challenge targets. We carry these figures as independent summaries of his argument report them; this pass did not obtain the monograph itself.

In 2022 a study went looking for the dead in the pollen. Izdebski, Guzowski, Poniat and colleagues, in Nature Ecology and Evolution, analysed 1,634 fossil pollen samples from 261 radiocarbon-dated sites across Europe covering roughly 1250 to 1450, reading agricultural activity and land abandonment as a proxy for population loss. The result was strikingly uneven. Southern Sweden, central Italy and Greece show dramatic contraction of the cultivated landscape. Catalonia and Czechia show no discernible decrease in pressure on the land. In Poland, the Baltic countries and central Spain, labour-intensive cultivation actually increased. Ireland and much of Iberia show continuity. The authors' conclusion is that mortality was far more spatially heterogeneous than assumed, and neither universal nor universally catastrophic.

That result did not go unchallenged either. Daniel R. Curtis published a formal Matters Arising in the same journal and the same volume, arguing that a reduction in grain pollen indicates a fall in cultivation and population, which can have several causes, and does not by itself measure Black Death mortality. The original authors replied in the same volume. The dispute is live and unresolved, which leaves the reader in an unglamorous but accurate position: the traditional continental figure is under active methodological challenge, the challenge is under active challenge in turn, and nobody currently gets to state one number for Europe as though it were measured. The regional rows in the table above are the honest way to feel the scale. Florence at roughly 60 percent and the Middle East at 25 to 40 percent in the same breath say something a single European average hides.

The end of serfdom is our file's headline consequence. Its stated mechanism is a massive labour shortage: peasants could demand higher wages and better conditions, and lords were forced to compete for workers. Its stated evidence is the Statute of Laborers of 1351 in England, which attempted to cap wages and largely failed, and the peasant revolts that followed, including the Peasants' Revolt of 1381. The documents are real and dated. The causal arrow is an argument, not a measurement, and it needs a caution our file does not give: serfdom weakened or ended on very different timetables in different places, and in parts of eastern Europe labour coercion tightened rather than loosened over the same period.

Wages are the link a reader most expects, and they are true at the endpoint. English manorial records show real wages doubled by 1450, and our file states that the rise lasted more than a century.

They are false in the 1350s, and that lag is where the interesting history is. Work on the English wage series by John H. Munro finds that the real wages of building craftsmen in southern England, rural and urban alike, had peaked in 1336 to 1340, fell through the 1340s, and continued to fall after the plague arrived in 1348, into the 1360s. Part of the delay is legal: the Ordinance of Laborers and then the Statute of Laborers froze wages by law, and manorial wages in many places stayed low into the early fifteenth century. So the mid-fifteenth-century peak is real, and it arrived decades late, against organised legal resistance, rather than as an immediate market response to a labour shortage. This finding comes from a working paper rather than a peer-reviewed article, and is the most lightly sourced external claim in this file.

Two further consequences sit on ordinary record evidence. Land redistribution: abandoned estates were consolidated and survivors inherited multiple properties, with tax records showing property concentration and abandonment. Urban growth: rural-to-urban migration accelerated as cities offered opportunities, and urban population recovery ran faster than rural.

The bottom of the ladder is technological innovation, and our file marks it soft itself. The proposed mechanism is that labour scarcity drove labour-saving inventions, with the printing press, water mills and mechanical clocks named. The evidence column, in our file's own words, reads: 'Correlation supported by economic historians (Pamuk, 2007).' Two of those three examples strain the chronology on their own. Water mills long predate 1347. Movable-type printing in Europe arrives around 1450, a full century after the first wave. Treat this as a hypothesis economic historians have argued, with Pamuk named, and not as a chain of cause.

The Church could neither explain the plague nor stop it, and it paid for that in authority. Many priests died or fled, and our file records that some religious orders lost 40 to 60 percent of their members. Alongside the institutional crisis came the flagellant movement: penitent processions through European cities, with self-flagellation offered as atonement, eventually condemned by Pope Clement VI in 1349.

The search for an explanation found a target. Jews were accused of poisoning wells and massacred across Germany, France and Switzerland, with Basel, Strasbourg, Mainz and Cologne among the named sites; entire communities were destroyed, and our file records attacks on around 200 or more Jewish communities. Clement VI issued papal bulls defending Jews and condemning the violence. Our file records that they were largely ignored.

The well-poisoning accusation was a conspiracy theory, not a historical fact: a medieval explanation manufactured to justify violence that was already under way. It belongs in this article's Tier 4 section beside the modern fringe claims, and the two are worth reading together. A population under a catastrophe it could not explain produced a story about a hidden group doing it deliberately. That story has exactly the shape of the modern claim that the plague was engineered depopulation. The medieval version got people killed.

What Europe made of it afterwards is visible in its art. The Danse Macabre, the Dance of Death, showed Death as the great equalizer taking rich and poor alike, and the tradition flourished from the fourteenth century into the sixteenth. Memento mori culture spread with it: skulls, skeletons and imagery of decay became pervasive in European art and literature.

That last row closes a loop worth naming. The bacterium was identified in 1894, during the Third Plague Pandemic, and took its name from the man who identified it. Everything this article says about what killed people in 1348 depends on the ancient DNA work of 2011 and after carrying that nineteenth-century identification backwards, out of the shipping lanes of Yunnan and into a London burial ground. The name came from the modern pandemic. The proof came from the medieval one.

The hypothesis as our file frames it: the Black Death created the conditions for the Renaissance by concentrating wealth among survivors through inheritance windfalls, disrupting traditional institutions and authority structures, creating labour scarcity that drove innovation, and weakening the Church's monopoly on intellectual life. David Herlihy argued in 1997 that the plague was a 'creative destruction' event for European civilization.

Our file supplies its own counter-arguments, and they are good ones. The Renaissance had multiple causes, including the rediscovery of Classical texts, competition between the Italian city-states, and trade wealth. Plague alone is an insufficient explanation. Its own verdict is the one this article adopts: the connection is supported but is not a simple causal chain, and the plague was one factor among many.

So the title stands on the lower rungs, not this one. Europe was remade in ways this file can evidence at Tier 1 and Tier 2: a mortality event our file puts at somewhere between 30 and 60 percent of the continent's people, the collapse and reordering of the labour market, the persecutions, the loss of clerical authority, an art that changed its subject. The road to the Renaissance is a separate and softer claim, and it is not what carries the title.

The Black Death arrived during the transition from the Medieval Climate Anomaly to the Little Ice Age. The proposed mechanism is that cooling stressed populations, reduced crop yields and concentrated rodent populations near human settlements, and the Great Famine of 1315 to 1322, caused by cold and wet conditions, had already weakened European populations before the plague arrived. Our file's own verdict is that climate and pandemic synergy is plausible but difficult to prove directly, and it offers one comparison: the volcanic winter of 536 CE preceding the Justinianic Plague. That comparison belongs to another file in this wing, and we leave it there.

The genetic thread starts with a hypothesis our own file then overturns. The CCR5-delta-32 mutation, which confers resistance to HIV, was once hypothesized to have been selected for by the Black Death through plague resistance, which would have meant the pandemic permanently altered European genetics. More recent evidence suggests the mutation predates the plague and may have been selected by smallpox or other pathogens instead (Galvani and Slatkin, 2003). Our file's summing up is that pandemics exerting genetic selection pressure is well established, while specific plague-to-gene links remain unresolved.

That summing up has become more right, not less, and our file's genetics section stops twenty years before the evidence that makes it so. Klunk, Vilgalys, Demeure and colleagues, in Nature in 2022, examined ancient DNA around immune-related genes from individuals in London and Denmark before, during and after the Black Death, and reported four candidate loci (ERAP1 and ERAP2, NFATC1, CTLA4, TICAM2), with a variant near ERAP2 the strongest signal and homozygotes for the protective allele reported as roughly 40 percent more likely to survive. Barton, Santander, Skoglund and colleagues then published a formal Matters Arising in Nature in 2025, finding that the reported enrichment does not survive a proper randomization test, reporting a p-value inflation factor of 11.80, and arguing that the apparent signal is consistent with artefactual differences between immune and neutral loci arising from different sequencing protocols and batches, with none of the four candidate variants clearing statistical thresholds once allele frequencies are correctly estimated. Their stated conclusion is that selection on immune genes during the Black Death remains plausible while its magnitude and the genes involved are unknown. Do not carry that survival advantage away as a fact. It is a contested finding under a live formal objection.

Two different things sit on that paper's record and they are routinely confused. Nature published an Author Correction to Klunk and colleagues in January 2025: an author-issued correction, not a retraction, and the original paper stands. Separately, the Barton Matters Arising is a formal scientific dispute filed against the paper in the same journal. A correction is a repair. A Matters Arising is an argument. Neither is a withdrawal, and this article treats them as what they are.

The claim: that the Black Death was not plague at all but a viral hemorrhagic fever similar to Ebola. Our file files this under Tier 4, No Credible Source, and the specific claim is indeed refused by the evidence. But the two works it names are not what that heading implies, and the reader should know precisely what is being refused. Samuel K. Cohn Jr. is a professional historian at the University of Glasgow, and The Black Death Transformed (2002) argued from a re-reading of the medieval sources that the described disease did not match modern bubonic plague. Cohn explicitly declined to name a replacement: his own conclusion states that in place of Yersinia pestis he offers no alternatives. The Ebola-like filovirus proposal is not his. It belongs to Susan Scott and Christopher J. Duncan, in Biology of Plagues (2001), published by Cambridge University Press. Our file conflates the two positions, and this article does not.

The refutation is clean and it is physical. Ancient DNA has definitively resolved the debate, with Yersinia pestis DNA recovered from dozens of plague pit burials across Europe. Our own file's qualification is the right one: the non-plague hypothesis is now obsolete, though it raised valid questions about plague ecology. That is worth sitting with. Refuted is not the same as unsupported, and grouping academic work that was beaten by later evidence into the same bucket as a claim that never had any is a category error, whoever makes it. The reason we know it was Yersinia pestis is that somebody pushed hard enough on the medieval sources to make somebody else go and check the teeth.

The other occupant of that tier is a different animal entirely. Fringe claims that the Black Death was a deliberate population reduction carried out by elites have no supporting evidence of any kind, and they fail on capability before they get anywhere near evidence: no medieval authority had the biological knowledge or the means to engineer a pandemic. This is the same story shape as the well-poisoning accusation of 1348, told about the same event by a much later century, and it is refused on the same grounds: no evidence, and no capability.