Studies

Counting the Dead of the European Witch Trials

Citation Formats

General Reference

Divinity Atlas. "Counting the Dead of the European Witch Trials." Accessed August 30, 2026. https://divinity.interactivelion.com/studies/counting-the-dead-of-the-european-witch-trials.

APA Style

Divinity Atlas. (n.d.). Counting the Dead of the European Witch Trials. Retrieved August 30, 2026, from https://divinity.interactivelion.com/studies/counting-the-dead-of-the-european-witch-trials

BibTeX

@misc{divinityatlas-counting-the-dead-of-the-european-witch,
  author = {Divinity Atlas},
  title = {Counting the Dead of the European Witch Trials},
  year = {2026},
  url = {https://divinity.interactivelion.com/studies/counting-the-dead-of-the-european-witch-trials},
  note = {Accessed August 30, 2026}
}

A long, distributed programme of archival counting rather than a single study, and the atlas records it as one because the question it answers is one question. Historians working territory by territory count trials and verdicts in whatever court records survive, and the aggregate is then estimated by extrapolating across the gaps. The results have moved steadily downward from the figures asserted in the nineteenth century, and the counting has also overturned several things everybody believed: that the trials were medieval, that they were run by the Inquisition, that they peaked where the church was strongest. None of those hold. The work is unfinished by construction, because some of the records were destroyed and some never existed.

Facts

Scholarship

Who Was Accused

Across Europe as a whole roughly three quarters of the accused were women, most of them poor, and many of them old, widowed or otherwise without a household to protect them. The proportion varies sharply by region, and in several, including Iceland, Estonia and Normandy, most of those accused were men. 1

The regional exceptions matter because they rule out the simplest explanations. Any account that treats the persecution as a straightforward campaign against women has to explain the territories where it was not, and any account that ignores the overall proportion is ignoring its most consistent feature.

Who Was Accused

The people executed were not witches, in any sense of the word. They were not practitioners of a surviving religion, they were not members of a sect, and the crime they were convicted of, harm worked by a pact with the Devil, is not something anybody did. 1

This is the settled position of the field and not a cautious formulation. Some of the accused practised folk healing or charming, which was a common and mostly tolerated trade and was not what the courts were prosecuting. Confusing the two is the commonest error in the subject.

What Counted as Evidence

Judicial torture. Under the Roman-canon procedure used across much of continental Europe a capital conviction effectively required a confession, and torture was the means of obtaining one. England did not use torture in ordinary criminal procedure, and English execution totals are a small fraction of Scotland's and a very small fraction of the worst German territories. 1

The correlation between torture, weak appellate review and high execution counts is one of the strongest findings in the field. Where a small local court could convict and execute without review, totals are high; where a higher court reviewed verdicts, they fall.

What Counted as Evidence

Spectral evidence. Testimony that the shape or spectre of the accused had afflicted a witness, while the accused was demonstrably somewhere else. Admitted at Salem in 1692 and then repudiated, and its withdrawal effectively ended the prosecutions there, because without it almost no case had anything left in it. 2

What Counted as Evidence

The swimming test. A suspect was bound and put into water on the theory that water, the element of baptism, would reject a witch: floating indicated guilt, sinking innocence. It was largely a popular or extra-judicial practice rather than a formal legal proof, and English judges repeatedly rejected it. 1

What Counted as Evidence

The witch's mark. A place on the body held to be insensible to pain, where a familiar spirit was said to have fed, found by searching and pricking. On any adult body a mole, wart, scar or skin tag is available, so the test has no negative result and confirms whatever the searcher already believes. 3

Method

Investigators

Regional archival historians across Europe, with syntheses by Wolfgang Behringer, Brian P. Levack and Julian Goodare. 4

Evidence

State of the Question

Ongoing. Regional counts are still being revised as archives are worked through, and each revision changes the aggregate. There is no prospect of a complete count. 1

Disputed

Estimated Executions

Modern archival estimates put executions across the whole of Europe and its colonies over roughly three centuries in the tens of thousands, with the commonly cited working range running from around forty thousand to around sixty thousand, out of a substantially larger number of trials. The atlas does not assert a total. 4

Regional archival historians across Europe, working jurisdiction by jurisdiction from surviving trial and execution records, and synthesised by Wolfgang Behringer, Brian Levack and Julian Goodare, have each produced an estimate of the total executed across the European witch trials, and the estimates differ because the historians differ on method rather than on any ideological reading of the trials. The disputed methodological choices are how far to extrapolate a death toll for jurisdictions whose own records have been lost, using rates observed in better-documented regions, whether to count deaths in custody or under torture before a verdict, and whether prosecutions under adjacent charges belong in the same total. Behringer's Witches and Witch-Hunts: A Global History (2004) and the comparable syntheses by Levack and Goodare each state their own figure together with its margin of uncertainty and the basis for the extrapolation used to reach it. The method throughout is archival demographic reconstruction from incomplete regional court records rather than a single continent-wide census that never existed. Because the records are genuinely incomplete and the extrapolation choices are stated rather than hidden, any single number quoted without a margin and a stated basis should be treated as unsourced.

Open Questions

Scholarly Note

How many people were executed is not established, and the share of the total that has to be extrapolated across jurisdictions which left no records is itself unquantified. 4

Counting proceeds jurisdiction by jurisdiction and then extrapolates across the territories whose records are lost, so the aggregate carries an unmeasured component that no amount of care in the counted jurisdictions removes. Regional counts are still being revised and each revision changes the total. Systematic archival counts for the jurisdictions not yet worked through, published with their own margins, would reduce the extrapolated share, and a shared convention for counting deaths in custody before verdict would narrow it further. Neither closes it. This is separate from the range itself, on which historians differ by method and which is recorded here as debated.

Debunked

Estimated Executions

Nine million people were burned as witches in Europe. 4

Gottfried Christian Voigt, a jurist in Quedlinburg, originated the nine-million-victim figure for the European witch trials in 1784 by extrapolating from a small set of local Quedlinburg records to the whole of Europe using an arithmetic that does not survive inspection when checked against the same regional records elsewhere. Matilda Joslyn Gage took the figure up in her 1893 Woman, Church and State, and from there it entered feminist, anticlerical and later neopagan writing as the Burning Times. Modern archival counting, jurisdiction by jurisdiction from surviving trial records, summarised by Wolfgang Behringer, Brian Levack and Ronald Hutton, gives a total in the tens of thousands rather than the millions, a method of direct regional record aggregation rather than the single extrapolation Voigt used, and it has thoroughly refuted the nine-million figure, false by a factor of more than a hundred. The claim is recorded here rather than omitted because it remains in wide circulation, and because setting the record straight against a still-current number matters on its own terms.

Learn More

What the Archives Can and Cannot Support

Counting is harder than it sounds and the difficulty is not squeamishness about a number. It is that the number does not exist in one place.

Witch trials were held in an enormous variety of courts across three centuries and dozens of jurisdictions: episcopal courts, secular criminal courts, seigneurial courts, town councils, and in some places bodies with no permanent existence at all. Some of those courts kept careful registers which survive. Some kept registers which were later destroyed by fire, war or indifference. Some kept nothing. A count therefore proceeds territory by territory: a historian works through what survives for one region, counts trials and verdicts, and produces a local figure with a stated margin. Aggregating those local figures into a European total requires extrapolating across the regions where nothing survives, and that extrapolation is where the disagreement lives.

There are further complications that are easy to overlook. Not everyone tried was executed, and acquittal rates varied enormously, so a count of trials is not a count of deaths. People died in custody before verdict and are recorded unevenly or not at all. Some jurisdictions executed for related offences under other names, which is how Anna Göldi was condemned as a poisoner in 1782.

The consequence is that responsible scholarship gives a range, states what it rests on, and does not offer a single figure. So does the atlas.

What the Counting Overturned

The counting did more than shrink a number. It broke several assumptions that are still repeated.

The trials were not medieval. The heaviest prosecution falls in the sixteenth and seventeenth centuries, overlapping the Renaissance, the Reformation and the beginnings of the scientific revolution, not the age of cathedrals. The Middle Ages, for most of their length, held officially that the night flight was a dream.

They were not principally run by the Inquisition. The Spanish Inquisition largely stopped prosecuting witchcraft after 1614, and the Roman and Portuguese tribunals executed very few. The great majority of executions were the work of ordinary secular courts, and the territories with the highest totals are generally those where small local jurisdictions could try and execute without review by a higher court. Strong central review reduced the numbers wherever it existed, which is an uncomfortable finding for several popular accounts at once.

They were not evenly spread. A large share of all European executions falls in the German-speaking lands of the Holy Roman Empire, with intense episodes elsewhere and long periods in which large countries prosecuted almost nobody. England, where judicial torture was not part of ordinary criminal procedure, has totals a small fraction of Scotland's, and Scotland's a small fraction of some German territories.

And the people executed were not witches. That is not a delicate way of saying the evidence was weak. It is the settled position of the field.

Cross-Tradition Connections