Archive/ The Early Modern /The Witch Trials

Interpretive illustrationThe scales of justice, thrown out of balance — a persecution that wore the robes of law to punish an imaginary crime. An original figure, chosen to mark the trials as a miscarriage of justice, not to depict their victims.

Early Modern · The Witch Trials

The Witch Trials

In the archive

The two-and-a-half centuries in which early modern Europe hunted, tortured, and killed tens of thousands of its own people — overwhelmingly poor women — for a crime that, as its persecutors imagined it, did not exist; told as the documented history of a persecution, not as evidence that its victims were what they were accused of being.

The confession

Somewhere in the German lands around 1600, an old woman — a widow, perhaps, or the village healer people came to for a charm against fever — is denounced by a neighbor after a cow sickens or a child dies. She is arrested, stripped, and searched for the Devil's mark. She denies everything. She is tortured — the thumbscrews, the strappado that pulls the arms from their sockets — until she confesses what her interrogators already know she must confess: that she made a pact with Satan, flew by night to the witches' sabbath, feasted and copulated with demons, and blighted her neighbors' crops and cattle. Then she is made to name the others she saw there. Some of them will be arrested next. And then she is burned.

She was not a witch. There was no secret society of Devil-worshippers flying to sabbaths; the crime she confessed to was, in that form, a fantasy — assembled by learned men and extracted by pain. This is the hardest thing to hold steady about the witch trials, and the most important: the persecution was utterly real, and the conspiracy it punished was not.

What the witch trials were

Between roughly 1450 and 1750, the courts of Christian Europe and its colonies tried perhaps 100,000 people for witchcraft and executed somewhere between 40,000 and 60,000 — the great majority in the fragmented Holy Roman Empire. The "witch" they hunted was a specific and terrible invention: not merely someone who worked harmful magic, but a person who had renounced God, sworn allegiance to the Devil, and joined a vast secret conspiracy against Christendom. It is this second, diabolical element — the pact, the sabbath, the anti-church of Satan — that was, as modern scholarship overwhelmingly agrees, largely fabricated by the persecutors themselves.

The making of the witch: from delusion to conspiracy

For most of the Middle Ages the Church had been, remarkably, the skeptic. The early-medieval Canon Episcopi taught that women who believed they flew by night with the goddess Diana were deluded by the Devil — the flight was a dream, not a fact, and to believe it real was itself an error. The witch hunts required the reversal of that doctrine: the night-flight had to become real, and the lone worker of maleficium had to become the sworn member of a satanic army.

That reversal was a welding-together of two very different things. The first was ordinary belief in maleficium — harmful magic — which almost everyone in pre-modern Europe accepted, and which shaded into the genuine world of village cunning-folk: healers, charmers, and finders of lost things who really existed and mostly did good. The second was an elite theory of demonology, worked out by theologians and lawyers, that reinterpreted this folk magic as Devil-worship. The two fused into what historians call the "cumulative concept" of the witch. In 1484 Pope Innocent VIII's bull Summis desiderantes affectibus endorsed the Dominican inquisitor Heinrich Kramer, who two years later produced the movement's infamous handbook, the Malleus Maleficarum ("Hammer of Witches," 1486) — a venomous, deeply misogynistic treatise arguing that women, as weaker vessels, were especially prone to the Devil. It was a printing-press best-seller for two centuries — yet, tellingly, the Malleus was never officially authorized by the Catholic Church and was condemned by leading theologians of its own day for bad doctrine and illegal methods.

The great hunts

The worst outbreaks came where central legal authority was weakest, local zeal unchecked, and torture freely used — above all in the patchwork ecclesiastical states of Germany. The Trier trials (1581–93) killed hundreds. The Würzburg and Bamberg hunts of 1626–1631 were the most devastating of all: in Bamberg, the "witch-bishop" Johann Georg Fuchs von Dornheim built a dedicated witch-prison and executed perhaps 900 people, including his own officials, priests, and children. From that hunt survives one of history's most harrowing documents — the secret letter of the Bamberg mayor Johannes Junius to his daughter in 1628, in which he swears his innocence, describes the tortures that broke him, and explains that he confessed to flying to the sabbath only because "I must say something, whether it be true or not." It is the clearest surviving proof, in the victim's own hand, that torture manufactured the crime.

Elsewhere the pattern varied. In Scotland, the North Berwick trials of 1590–91 drew in King James VI himself, who believed witches had raised storms against his ship; his book Daemonologie (1597) made him Europe's royal witch-theorist. The New England town of Salem produced a small, late colonial flare in 1692, hanging nineteen and pressing one man to death. But the counter-cases matter as much: the Spanish and Roman Inquisitions, often imagined as the cruelest, were in fact among the most skeptical — the Spanish inquisitor Alonso de Salazar Frías, after investigating the Basque panic at Zugarramurdi (1609–14), concluded there was no real evidence of witchcraft at all and effectively halted the hunts in Spain. And in the Friuli of northern Italy, the historian Carlo Ginzburg uncovered the benandanti — a genuine folk fertility cult who believed they left their bodies at night to fight witches for the harvest, and whom the Inquisition slowly, over decades, persuaded and pressured into calling themselves witches: a documented case of the persecutors' fantasy overwriting a real folk belief.

The machinery

What turned belief into slaughter was procedure. Where courts adopted inquisitorial methods (the judge investigates and prosecutes) and permitted torture, two things followed: confessions could be produced to order, and tortured suspects could be forced to name accomplices — so that a single accusation could chain-react into a panic consuming dozens or hundreds. Where older accusatorial procedure survived, or where higher courts reviewed convictions and limited torture (as in England and Spain), the death tolls stayed far lower. The trials were not the work of one church: Catholic and Protestant courts both hunted witches with equal conviction, in an age when both were certain the Devil walked abroad.

Who died, and why

About three-quarters to four-fifths of those executed across Europe were women — most of them old, poor, widowed, or marginal, and sometimes the very healers and midwives their communities had once relied on. That pattern makes misogyny a central thread of the hunts, and the Malleus its charter. But it was not the whole story: in Iceland some ninety percent of the accused were men, and men were the majority in Finland, Estonia, and Russia too. Historians such as Brian Levack therefore reject any single cause and stress a convergence of factors: the legal machinery of torture; the demonological theory of the learned; the social stress of a Europe wracked by plague, war, and famine; and the religious terror of the Reformation age. The historian Wolfgang Behringer added a striking material trigger — the Little Ice Age, whose brutal cold destroyed harvests in the very decades of the worst hunts, giving desperate communities a supernatural culprit for weather that "should not" have happened. Beneath them all lay the ancient human reflex to find, in a frightened time, someone to blame.

The end of the hunts

The trials did not stop because Europe stopped believing in the Devil. They stopped because the machinery lost its legitimacy: judges grew skeptical of confessions won by torture, higher courts demanded evidence that spectral flights and pacts could not provide, and voices of protest — most bravely the Jesuit Friedrich Spee, whose Cautio Criminalis (1631) exposed from the inside how torture made the innocent confess — slowly turned educated opinion. The rising Enlightenment finished the work. By the eighteenth century the last legal executions had ceased, and the witch passed from the courtroom into folklore.

The believer's lens

To understand the trials at all, one must grant how real and coherent the witch was to the people who hunted her — for this is a chapter whose "believers" are the persecutors and their whole society. In their world the Devil was an active, present enemy; Scripture commanded "thou shalt not suffer a witch to live" (Exodus 22:18); and the recent horrors of plague, war, and failing harvests seemed exactly the work a malicious supernatural conspiracy would do. A judge like the Bamberg bishop was not, in his own eyes, a murderer but a defender of Christendom against a satanic fifth column; a peasant who denounced a quarrelsome neighbor after his child died was reaching for the only explanation his cosmos offered. And beneath the elite fantasy lay a genuine popular world of magic — real cunning-folk, real charms, the real benandanti — in which the power of maleficium was simply assumed by nearly everyone. The trials were monstrous, but they were not, to their age, insane: they were the terrible logic of a worldview taken to its conclusion.

The skeptical lens

The modern historical verdict is unambiguous, and it is the frame of the whole chapter. There was no satanic witch-cult: the "witch-religion" thesis of Margaret Murray — that the accused were surviving pagans — is thoroughly discredited (and, ironically, became the founding myth of modern Wicca, Chapter 38). The confessions describe not events but the contents of the interrogators' own manuals, fed back under torture, which is why they repeat the same script across a continent. Maleficium — harmful magic — did not work. The persecution, in short, punished an imaginary crime, and the historian's task is to explain the persecutors, not the witches: the legal machinery, the demonological theory, the social and climatic stress, the misogyny, and the collective panic. The one genuine reality buried under the fantasy was the ordinary folk-magic and healing of the accused — real practices, harmless or helpful, reinterpreted as devilry by men who had decided in advance what they would find.

Symbology and sacred encoding

The witch trials ran on a dark and revealing symbolic system — much of it projected onto the accused rather than found in them.

  • The Devil's mark and the witch's mark were read on the body as encoded proof: any mole, scar, or insensitive patch could be "the Devil's seal," and searchers pricked the accused for a spot that would not bleed. The evidence was written on the flesh because it could be found nowhere else.
  • The confession itself was an encoded, formulaic text: the sabbath, the pact, the flight, the demonic sex recur in trial after trial in the same shape — not because witches did these things, but because interrogators, working from the same demonological manuals, asked the same leading questions and tortured until they got the same script (as Junius's letter proves).
  • The imagined sabbath was a systematic inversion of Christian symbol — a black anti-Mass, prayers said backward, the cross trampled, the world turned upside down — the projection of a society's fears about its own most sacred things (the same inversion Chapter 39 traces into later "Satanism").
  • And the familiar iconography of the witch — the pointed hat, the broomstick, the cauldron and black cat — is very largely later folklore and art, not the reality of the trials: even our picture of the persecution is a construction.

Connections

The witch trials belong to the anxious, violent world of the Reformation (Chapter 31) — an age when both churches were certain the Devil was loose. They inherited the machinery of the medieval Inquisition and the demonology of the schoolmen (Chapter 28), while reversing the older skepticism of the Canon Episcopi. Crucially, they are the historical root of a distinction this book draws carefully in the modern era: between fabricated, accusatory "Satanism" — the imaginary Devil-cult of the trials and its descendants in later panics (Chapter 39) — and the real, self-identified Satanism of the twentieth century, which is an entirely different thing. And they are the (largely mythical) charter of modern Wicca (Chapter 38), whose twentieth-century founders claimed descent from a surviving witch-religion — a claim examined, and doubted, there.

✦ The evidence, honestly

Well-supported by evidence

The witch trials themselves — their chronology (c. 1450–1750), their concentration in the Holy Roman Empire, and a death toll now soberly estimated at 40,000–60,000; the papal bull of 1484 and the Malleus Maleficarum (1486); the great German hunts (Trier, Würzburg, Bamberg) and the Junius letter; North Berwick and James VI; Salem (1692); the role of torture and inquisitorial procedure; the predominance of women; the skeptical restraint of the Spanish Inquisition (Salazar); and the reality of village cunning-folk (and the benandanti).

Not supported (as history)

That the accused were actually members of a satanic witch-cult — Margaret Murray's "witch-cult" thesis is thoroughly discredited. That the confessions, extracted under torture, describe real events. And that maleficium — harmful magic — actually worked. The crime, as charged, was imaginary; the persecution was not.

Genuinely open

The precise death tolls, estimates from incomplete records (the "nine million" figure sometimes claimed is pure myth). The relative weight of the many causes (misogyny, torture-law, the Little Ice Age, confessional conflict). And, in any given trial, how much a real tradition of folk healing or magic underlay the accusation before torture buried it under fantasy.

Sources

Walk this chapter’s room in the Museum (beta) →