The witch trials

Not nine million, not medieval, not pagans, and not over — the numbers, the dating, the victims and the present tense are all commonly got wrong.

What it is

The prosecution and execution of people for witchcraft in Europe and its colonies, chiefly between about 1450 and 1750.

It is among the most misremembered episodes in Western history, and nearly every element of the popular account is wrong.

Where it came from

Not from the Middle Ages. The trials are an early modern phenomenon, peaking between roughly 1560 and 1630 — the era of the Reformation, the Counter-Reformation, the wars of religion, the Little Ice Age and the beginnings of the scientific revolution. Medieval church authorities had generally regarded belief in witches' powers as itself a superstition to be corrected.

The Malleus Maleficarum (1487), written by the inquisitor Heinrich Kramer, was influential and frequently reprinted — and was never official doctrine. The Inquisition distanced itself from Kramer, and his own order regarded him poorly.

The geography is very uneven. The Holy Roman Empire accounts for roughly half of all executions. Switzerland, Scotland, France, Poland and the Low Countries have substantial numbers. Spain, Italy and Portugal — where the Inquisition was strong — have strikingly few, because the inquisitorial courts were sceptical. After the Basque outbreak at Zugarramurdi, the Spanish inquisitor Alonso de Salazar Frías investigated exhaustively, concluded in 1612 that there had been no witchcraft, and the Spanish Inquisition effectively stopped executing witches. Ireland and Russia had very few trials.

How it is used

The trials are invoked constantly — in feminist history, in pagan self-understanding, in political rhetoric about persecution. Getting the facts right matters for all of those uses.

What we can and cannot say

The number is not nine million. That figure originates with an eighteenth-century German writer, Gottfried Christian Voigt, who extrapolated wildly from local records; it was picked up by the American suffragist Matilda Joslyn Gage in 1893, entered twentieth-century feminist and pagan writing, and circulates still.

The historical consensus, from actual court records, is roughly 40,000 to 60,000 executions across the whole period, from perhaps 90,000 to 110,000 prosecutions.

That is a horror. It is also two orders of magnitude less than the figure commonly repeated, and using the inflated number damages the case it is meant to serve.

The victims were not pagans. Margaret Murray's thesis that the accused practised a surviving pre-Christian religion — the foundation of Wicca's original origin story — is rejected. The accused were Christians, accused by Christians, in Christian communities. The confessions that appear to describe pagan ritual were extracted under torture and follow the shape of the interrogators' questions.

They were mostly but not only women. Around 75 to 80 per cent of the accused across Europe were women, and the misogyny in the demonological literature is explicit. But the regional variation is striking and is usually omitted: in Iceland roughly 90 per cent of the accused were men; in Estonia and in Normandy men were also the majority. A purely gendered explanation has to account for that.

They were mostly not cunning folk. As that article sets out, the village magical practitioner was generally the person consulted to identify witches. The accused were more often the poor, the old, the socially isolated, the quarrelsome, and people with a reputation for cursing neighbours who then suffered misfortune.

Salem was small. The 1692–93 Massachusetts trials killed twenty people — nineteen hanged, one pressed to death — with several more dying in jail. It is culturally enormous in the English-speaking world and numerically minor.

And the trials are not over. This is the part most often left out.

Witchcraft accusations continue to kill people. In parts of sub-Saharan Africa, accusations against children and older women result in expulsion, torture and murder, documented by UNICEF and by national human rights bodies. In Papua New Guinea, sanguma accusations produce killings severe enough that the government repealed its Sorcery Act in 2013 and has struggled to prosecute the resulting cases. In India, particularly in Jharkhand, Assam and Odisha, hundreds of people — overwhelmingly women, frequently widows with land — are killed each year following witchcraft accusations, and several states have passed specific legislation. In Saudi Arabia, sorcery remains a capital offence and executions have been carried out.

The mechanisms are recognisable from the European record: misfortune requiring explanation, a socially marginal target, a specialist who identifies them, and a community that acts.

We cannot say the European trials had a single cause. Religious conflict, legal changes permitting torture, climatic stress and crop failure, local feuds and the ordinary dynamics of accusation all contributed, and historians weight them differently.

What it is

Where it came from

How it is used

What we can and cannot say

Further reading