What Were the Witch Trials? A European Panic With a Legal Machinery

Belief in harmful magic is ancient and near-universal, and mass prosecution is not, so the question is what altered. Several developments combined in the fifteenth century:

• •A theological shift from treating magic as superstition to defining witchcraft as a pact with the devil, which converted a folk matter into heresy and therefore into a crime against God requiring the most serious procedure
• •The elaboration of the sabbath, a fantasy of night flights, secret gatherings and collective worship of the devil, which meant that any accused person could be asked who else was present, converting individual accusations into cascading ones
• •The adoption of inquisitorial procedure in secular courts, in which the court itself investigates and prosecutes rather than requiring a private accuser to bear the risk
• •Judicial torture, legally permitted in much of continental Europe to obtain the confession that Roman-canon law required for conviction, which reliably produces confessions and names
• •Printing, which spread demonological manuals rapidly, above all the Malleus Maleficarum of 1487, a handbook that was condemned by the theology faculty at Cologne and went through numerous editions anyway
• •Weak central control in fragmented territories, since the worst outbreaks occurred where small jurisdictions had criminal authority with no effective appeal

The pattern is consistent enough to be informative. The large majority of the accused were women, with the proportion varying by region from around half in Iceland and Estonia to over ninety percent in parts of Germany and England. Within that, the recurring profile is a woman who was old, poor, widowed or unmarried, without male protection, and known for a sharp tongue or a reputation for cursing. Midwives and healers were accused, though the claim that they were systematically targeted has not held up under examination of the records. Accusations frequently followed an ordinary quarrel, over a loan, a gift refused, a boundary or a grazing right, after which a subsequent misfortune, a sick child, a dead cow, a failed batch of butter, was attributed to the earlier ill feeling. That mechanism, identified by Keith Thomas and Alan Macfarlane in their studies of English cases, connects the trials to a period when traditional obligations of charity were breaking down and a refused request generated guilt that could be converted into suspicion.

Procedure varied enormously and the differences explain the wildly different death tolls. In inquisitorial systems, a court investigating on its own initiative could apply torture to obtain a confession, which was required for a capital conviction, and could then require the confessing person to name accomplices, which generated the chain reactions that produced the largest panics. Evidence included the testimony of children, spectral evidence in which a witness reported being tormented by an apparition, and physical tests including searching for an insensitive mark on the body and, in some places, swimming, in which floating indicated guilt. England was an exception that proves the mechanism: torture was not permitted in ordinary criminal cases, trials were by jury, and the death toll was consequently far lower, with the worst English episode occurring during the breakdown of order in the civil war under a self-appointed witchfinder. Spain and Italy, where the formal Inquisition had jurisdiction, also had comparatively few executions, because the trained inquisitors were sceptical of the evidence and required standards that local secular courts did not.

Careful archival work since the 1970s has replaced earlier estimates that ranged into the millions. The current scholarly consensus is roughly ninety thousand prosecutions and between forty and sixty thousand executions across Europe and its colonies over three centuries, with around half occurring in German-speaking lands. The figure of nine million, still repeated, originated with an eighteenth-century German estimate extrapolated carelessly and was popularised in the nineteenth and twentieth centuries. Other persistent errors include the belief that the trials were medieval, that they were driven primarily by the Catholic Church, since Protestant territories prosecuted at least as vigorously and the Spanish Inquisition was notably restrained on this charge, that the accused were practitioners of a surviving pagan religion, a theory advanced by Margaret Murray in 1921 and comprehensively rejected, and that burning was universal, since England and its colonies hanged.

The decline was gradual, uneven and driven less by disbelief in witchcraft than by a collapse of confidence in the evidence. Judges and educated elites became increasingly unwilling to convict on confessions extracted under torture and on spectral testimony, and several influential figures argued the procedural case, including Friedrich Spee, a Jesuit who accompanied condemned prisoners and published anonymously in 1631 that torture would make anyone confess to anything. The Salem trials of 1692, small in European terms with twenty deaths, ended when the governor excluded spectral evidence and the convictions collapsed. Courts began requiring standards of proof that witchcraft accusations could not meet, prosecutions fell, and statutes were repealed across Europe during the eighteenth century, with the British Witchcraft Act of 1735 recasting the offence as pretending to have magical powers, which is a fraud statute rather than a religious one. Accusations of witchcraft continue to cause violence and killings in several parts of the world today, which is why the historical episode is studied as a case of how a legal system can generate the evidence it is looking for.

The trials ran mainly between 1560 and 1630, not in the Middle Ages, and killed an estimated forty to sixty thousand people rather than the millions often claimed. They required a specific apparatus: a theology defining witchcraft as a pact with the devil, the fantasy of a sabbath that made accused people name others, inquisitorial procedure and judicial torture that reliably produced both. England's lower toll follows from its prohibition of torture in ordinary criminal cases, and the panic ended when courts stopped accepting the evidence.