The Years of Magical Thinking: Explaining the Salem Witchcraft Crisis
by Mary Beth Norton

Most Americans’ knowledge of the seventeenth century comes from semi-mythical events such as the First Thanksgiving at Plymouth, Pocahontas purportedly saving Captain John Smith from execution in early Virginia, and Salem witchcraft. This witchcraft scare, and the trials that followed, have especially seized the popular imagination.

Separating the myths from the reality of the Salem witchcraft episode is the historian’s task. In large part, students learn about the Salem witchcraft trials from reading Arthur Miller’s play, The Crucible, frequently assigned in high school classes. Miller’s play is a work of fiction, not history, but its enormous popularity has effectively distorted what really happened in Essex County, Massachusetts, in 1692. Even though Miller drew on original legal documents, he gave his own twist to the evidence. Most notably, he transformed Abigail Williams, an accuser who was actually eleven years old, into an older servant who had had an affair with his hero, John Proctor, and who was seeking revenge for Proctor’s return to his wife. Although Proctor’s actual servant, Mary Warren, accused him and his wife Elizabeth of being witches, no record implies a romantic relationship between Warren and Proctor.

Miller’s play perpetuates myths about the 1692 crisis that were initially created in the nineteenth century. He begins the play with a dramatic scene of the later accusers dancing in the forest and dabbling in magic under the direction of Tituba, the African slave of the Reverend Samuel Parris. That scene is entirely fictional. No seventeenth-century source describes the teenaged accusers engaging in magic of any sort as a group, and no source describes any involvement by Tituba in conjuring with the accusers. In addition, Tituba was not African but rather Native American; she probably had been captured by England’s Indian allies in a raid on one of the Spanish missions in the region that is now northern Florida or southern Georgia, for one reliable source terms her a “Spanish Indian,” as such captives were known in New England. (Nineteenth-century authors concluded she was African or half-African because they knew she was a slave, and at that time historians did not realize how many enslaved Indians lived in New England.)

Even our common name for the crisis—Salem witchcraft—is geographically inaccurate. The accusations began in Salem Village, an area distinct from Salem Town and now known as the town of Danvers; and by the time the crisis had ended, more people had been accused of witchcraft in neighboring Andover than in Salem Town or Salem Village. Of the approximately 150 people formally charged during the crisis, only twenty-four resided in Salem Village. The witchcraft crisis in fact enveloped much of Essex County, the entire northeastern portion of the Massachusetts Bay Colony. Therefore, an analysis of its origins and consequences cannot be confined to Salem Village alone, nor can the entire explanation lie (as it does for Miller and many others) solely in the accusations advanced by the “afflicted girls.”

The crisis began in mid-to-late January 1692 when three young Salem Village children suffered from strange fits that contorted their bodies and led them to complain of various ailments, which they eventually—under detailed questioning from concerned adults—attributed to tortures inflicted upon them by three local witches, one of whom was the Native American slave Tituba. Interrogated by the authorities in early March, Tituba confessed to being a witch and implicated the other two named by the children. She also offered vague descriptions of additional witches from other towns. A careful examination of the chronological development of the crisis shows that, at first, the authorities hesitated to arrest anyone other than the three initially accused. But when some married women and twenty-somethings in Salem Village also began to suffer from fits and to offer additional accusations, the crisis escalated. More people, in Salem Village and elsewhere, started to accuse their neighbors of being in league with the devil. Such charges reached an initial peak in May and a second one in July through early September.

Depositions gathered by the authorities frequently, though not always, showed that witnesses had longstanding suspicions of those they accused. Some of the testimony recounted injuries suffered in mysterious incidents that had occurred two or three decades earlier, incidents that the accusers attributed to their neighbors’ practice of witchcraft. Such suspicions then were validated in 1692 by the afflicted people’s claims of torture visited upon them by those very witches in spectral form. It seems clear that, as the crisis developed, the afflicted people (who resided in Salem Village and Andover) heard gossip about suspects from many Essex County towns and incorporated that gossip into their accusations. Scholars disagree about whether the afflicted people acted deliberately and rationally or not, but nearly all historians reject explanations that attribute the crisis to such causes as epilepsy, an outbreak of disease (one author has posited encephalitis), or ergot poisoning, which may lead to hallucinations. Those who have studied the crisis most thoroughly concur that no physical or medical explanation can account for all the evidence.

One might ask: if many of the accusations had roots that were decades old, why were prosecutions not pursued vigorously prior to 1692? Why did that year become so critical? Only a handful of the accused witches in the Salem crisis had been tried previously, and all had been released without formal punishment. Those questions and that observation turn our attention away from the accused and the accusers to the Massachusetts authorities. In 1692, judges took seriously witchcraft claims they had earlier summarily rejected.

The explanation for why the judges acted that way lies in the fact that they were the colony’s councilors and military leaders as well as the justices of the special court established by the governor in May to try the witches. In those other capacities, they were deeply involved in the war with the Wabanaki Indians then enveloping New England. The war had begun in August 1688, and the English colonists had suffered devastating defeats. Once-prosperous settlements on the Maine coast—where fishing and lumbering produced major profits for Salem and Boston merchants and investors—had largely been abandoned after they were overrun by Native warriors.
English settlers had long believed that the Indians of North America were devil-worshippers. Colonists lived in what one historian has termed a “world of wonders,” in which—prior to the adoption of Enlightenment thinking about scientific explanations of natural phenomena—many seemingly inexplicable occurrences were attributed to witchcraft or other supernatural causes. For Puritans, the invisible world was as palpably real as the visible one. Accordingly, Massachusetts residents could readily connect the war in the visible world, in which they were being attacked by Indians, to the war in the invisible world, in which they were being attacked by witches. Both antagonists were acting at the direction of the devil, and if New Englanders could not defeat the devil and his Native allies on the battlefield, they could at least defeat him in the courtroom.

And so it was surely no coincidence that Tituba, a Native American (though not a Wabanaki) was one of the first three accused witches. Nor was it a coincidence that the first large wave of accusations came in mid-April, immediately after one confessing witch revealed that the devil had recruited her into his ranks four years earlier while she was living on the Maine frontier. Once the two challenges to the colony’s existence became linked in New Englanders’ minds, they perceived witches everywhere, just as they similarly feared attacking Indians so much that one panicky Essex County community, Gloucester, mustered its militia for several weeks in July to fight what residents later acknowledged were nothing but spectral Native enemies. And a little-known fact is that among the accused witches were some prominent men long suspected of trading with the Wabanakis.

The special court tried suspects at brief sessions held sporadically from mid-June through late September, 1692. Twenty-seven people were convicted, nineteen of them (fourteen women, five men) hanged; the last executions were on September 22. An additional male suspect, Giles Corey, rejected the court’s authority by refusing to enter a plea, and was pressed to death by heavy stones (a medieval English punishment). In late October, after criticism of the court’s acceptance of the afflicted people’s descriptions of spectral tortures became too loud to ignore, the governor dissolved the special court. But a few accusations and arrests still followed, and trials resumed in January 1693 in regular courts. Three more women, all confessing witches, were convicted, but the governor quickly reprieved them, as he did the eight people who had been convicted by the special court but not yet hanged.

The Salem witchcraft crisis of 1692 was neither a vast conspiracy of local residents, nor, as Miller’s play suggests, an attempt at revenge by a thwarted lover. Rather, it was a product of its own fraught era—the unique confluence of a devastating war in northern New England, rampant gossip, and pre-enlightenment magical thinking. Almost as remarkable is the speed with which the crisis passed. Five years later, as the Indian war ended, one judge and twelve jurors publicly apologized. In 1711, the colony compensated many surviving victims and their families. But not until early this century did the state by law formally declare all the accused to be innocent.

Mary Beth Norton is the Mary Donlan Alger Professor of History at Cornell University. She is the author of Founding Mothers & Fathers: Gendered Power and the Forming of American Society, and and In the Devil's Snare: The Salem Witchcraft Crisis of 1692. For more resources, see here.