In the winter of 1691-92, several people in Salem Village, most of them young women, but eventually including a few men and boys, began behaving in a "strange & unusual manner" (Calef, ii.4), with an affect (1) which was interpreted as illness. The town's minister, Samuel Parris, whose daughter and niece were among those with this odd affect, sought to cure the perceived problem with prayer; others, including a doctor of physick (2) who was called in, felt that the people in question were afflicted with a witch's supernatural curse, and this diagnosis came to be accepted as true. Friends and relatives prompted the "afflicted" people to name their supposed tormentors. On 29 February 1691/92, after over a month of acting oddly, the "afflicted" named three local women as "witches." One of these women, a slave of Mr. Parris named Tituba, said, when questioned, that she was a witch, that the two others arrested were witches, and that there were two other women and a man "from Boston" involved (SWP 747). Shrewdly or luckily, Tituba had realized that the best thing she could do in her situation was to tell the investigators what they wanted to hear.

Thus the diagnosis -- of affliction by witches -- was "proven" to be correct, and at the same time the extent of the perceived problem expanded from three to who knew how many. The strange affect of Parris's children and an increasing number of others continued, and these "afflicted" continued to supply names of supposed "witches." By the end of the year there were "about Fifty persons" with the affect of being afflicted (Hale 423), nineteen people and two dogs had been hung for "witchcraft," another had been tortured to death, five had died in prison from lack of proper food or shelter, and the jails were full with those awaiting trial.

In 1768, Hutchinson published the first history of this witch panic. He considered "whether the afflicted were under bodily distempers, or altogether guilty of fraud and imposture" (ii.19), and decided in favor of fraud. In 1831 Charles Upham agreed: the afflicted had acted with "a malicious disposition to wreak vengeance upon enemies" (Upham, 1831 Lectures 23). In 1867, however, Upham was less certain: it was "almost beyond belief that they were wholly actuated by deliberate and cold-blooded malignancy" and it was hard to say "how much may be attributed to [such 'bodily distempers' as] credulity, hallucination, and the delirium of excitement" (Upham, 1867 Salem Witchcraft 4-5). In 1949, Marion Starkey had no doubts: the afflicted's odd affect was entirely due to psychological 'distempers,' and she offered a pop-Freudian diagnosis of "hysteria" (42). In 1969, Hansen agreed with Starkey that the afflicted had been hysterical, presenting his view with the scholarship and language of the academy.

Starkey's "hysterical bobbysoxers" diagnosis has entered the popular canon and school textbooks, while (with the exception of Godbeer) Hansen's verdict of "hysterical in the scientific sense of that term" (1) has been accepted as true by the majority of scholars (eg, Demos, McMillen, and even Karlsen), who treat the cause of affliction as settled and go on to other projects. While I see the cause as not settled, I will look instead at the way the same descriptions of affect have produced such mutually exclusive interpretations -- fraud and illness -- and suggest why fraud went entirely out of fashion, after being accepted for over a century, while hysteria came into fashion (oddly, only Upham allows a mixture of fraud and illness). I will suggest that these shifts in interpretation are not founded on any new knowledge or new theories of psychology, but grow out of changes in cultural and ideological attitudes, especially toward women, and that they are made possible by the ambiguities of historical documents, by inadequate analyses of the explanations that were available in 1692, and occasionally by poor reasoning on the part of the historians.

According to Calef, afflictions at Salem first appeared as crawling under furniture, using "sundry odd Postures and Antick Gestures" and saying "foolish, ridiculous" things (4). Twelve year old Abigail Williams, for example, charged around the Parris house, flapping her arms like wings and crying "Whish, Whish" (Lawson, 153). She was, in other words, playing. In a society that sought to prevent "physical spontaneity" (on the notion that godly deportment was linked to a godly soul?), such behavior would usually be seen as misbehavior (Godbeer 108 & note 100). However, with a few exceptions (such as John Proctor's disciplining of his servant Mary Warren), the afflicted were treated as ill rather than delinquent. In part this was because their actions were seen as involuntary: Lawson saw Abigail as being "hurryed with Violence" by unseen hands (153). In part it was because the afflicted mostly came from religious households (Godbeer 107), and the parents did not want to believe they had failed to produce properly religious children or servants. In part it was because women were seen as "The Weaker Vessel" and thus more susceptible than men to illnesses (C. Mather, Angel of Bethesda, 233) -- if it had been mostly males who were afflicted, likely they would have been seen as truants.

There were two natural illnesses known at the time that could explain the "fits" or convulsions typical of the afflicted's affect: epilepsy and "the strangulation of the mother," or hysteria. At the time, both were wastebasket categories, less diseases than labels for sets of symptoms (3): epilepsy could refer to just about any kind of convulsive disease (Temkin), while a diagnosis of "the mother" simply meant that the patient was a woman and the diagnoser was baffled or lazy (Slavney, 17). Both were explicitly thought connected to the supernatural: Cotton Mather wrote in his 1724 medical treatise The Angel of Bethesda, that epilepsy can sometimes weaken the mind enough that evil spirits can "Strangely Insinuate themselves into the Malady. . . . Some of the Demoniacks in the Gospels, were Epilepticks" (142-43). In 1664, Thomas Brown wrote that "the mother" could be "heightened to a great excess by the subtlety of the Devil" (Hansen 2). Natural explanations were linked with supernatural ones, but the natural diseases were generally untreatable: supernatural illness could be cured with prayer.

Technically there were two kinds of supernatural illness: possession, in which evil spirits were inside the ill person, and obsession, in which they tormented the person from without. The evil spirits could have been sent directly by the Devil, or they could have been sicced on the afflicted person by a "witch." Practically, "possessed," "obsessed," and "bewitched" were often confused, and "came very near to being synonymous" (Thomas 478). As Abigail Hobbs and Mary Warren found out, obsessed victim and obsessing witch could also become all too synonymous. There was a growing disbelief among conformists in actual demoniac possession (Thomas ch15). Anyway, it was less daunting to deal with a supposed witch than the devil, so the diagnosis in Salem was of a malefic rather than diabolical affliction.

The most detailed descriptions of affliction or possession that we have are not from Salem but from a handful of isolated cases that were written up by ministers and published. One of the earliest of these was the 1672 possession of Elizabeth Knapp, written by Rev. Willard and summarized in Increase Mather's Remarkable Providences. Elizabeth's affect included violent, screaming convulsions. She accused two local women of witching her, but Willard investigated the accusations and found them groundless. He seemed convinced that she had rather herself signed a compact with the devil, and hounded her to confess to it (Demos 99-109). Instead, in the midst of another fit, as Willard thought, the Devil spoke through her in a "grum" voice. Suspecting a "counterfeit," Willard watched her mouth "and observed not any of her [vocal] organs to move" when the voice spoke (Demos 109). Increase Mather elaborated: "Labial Letters [were articulated] without any motion of her Lips," words seemed to "proceed out of her throat, when her Mouth was shut," and also when her mouth was "wide open, without the use of any of the Organs of speech" (22). Short of a conspiracy between Elizabeth and a ventriloquist, this is manifestly impossible unless we adopt 17th century beliefs in the supernatural.

Cotton Mather wrote three accounts of possession. The first, Memorable Providences, came out in 1689 and describes the affliction of the Goodwin children. He emphasizes that "it was perfectly impossible for any Dissimulation" to produce the effects he reports (sect 2). The children's convulsions, he says, were "beyond those that attend an Epilepsy" (sect 3, 13). Yet the children's fits "always" left them able to eat in the evening and sleep well (s5): evidently they never bit through their tongues or otherwise injured themselves, as it was known that epileptics could do (C. M, Angel ch 28). Therefore their fits were not "beyond" those of epilepsy. Mather says the children would sometimes "fly like Geese," with "just their Toes now and then upon the ground"; on one occasion, one "flew the length of the room, about 20 foot" without her "feet all the way touch[ing] the floor" (s13). Mather took the eldest girl into his home for closer observation and a regimen of prayers for a cure. At one point she told him that one of her supernatural persecutors was going to push him down the stairs. Mather thought he was "unaccountably made to step down backward two or three stairs" (s23).

During the 1692 panic, Mercy Short, a war orphan working as a servant in Boston, came down with fits. Cotton Mather again wrote up a report, "A Brand Pluck'd out of the Burning." Mather says that her supernatural tormentors burned her with spectral fire; when this happened, Mather and others smelled brimstone (s 12). On Christmas (a Popish holiday) Mercy said the tormentors "were going to have a Dance" and those present "most plainly Heard and Felt a Dance, as of Barefooted People, upon the Floor" (s 17).

About a year after the trials, Cotton Mather wrote his third and last account of possession, of Margaret Rule (published in Calef's More Wonders as "Another Brand Pluckt out of the Burning"). Margaret's affect included her tormentors forcing potions down her throat and throwing powders into her eyes. All of this was, of course invisible, but when some powder or potion splashed on Margaret's face, Mather says onlookers clearly saw the liquid and the powder. She said the potions burned her throat, and sure enough, people in the room could smell the brimstone (sect 7). Onlookers saw something unseen stir her pillow, and "one present [Mather?] laying his hand there, he to his horror apprehended that he felt tho' none could see it, a living Creature, not altogether unlike a Rat" (sect 8). Others felt this creature as well, Mather insists. Finally, Mather explicitly states that on at least one occasion, "her Tormentors pull'd her up to the Ceiling of the Chamber, and held her there" so that it was "as much as [the spectators] could all do to pull her down again" (sect 8). Calef was extremely skeptical of this last claim, and Mather produced four witnesses who swore to his veracity (More Wonders, ii.68-69). Again, unless we adopt Cotton Mather's world view, all of these things are completely impossible.

When Hutchinson wrote his history, the world view had changed considerably. Inspired by Newton's physics, educated people had begun to assume that all aspects of the universe were explainable in purely rational terms. The invisible world that was so real to Cotton Mather and his contemporaries had begun to vanish. True, there were a "great number of persons" who supposed the accusers had been "under bodily distempers which affected their imaginations" (47), but Hutchinson knows that "a little attention" to the facts will "force conviction" that they were simply frauds (47). If Mather was such a credulous fool to believe that levitation was possible, then how much the easier it must have been to deceive him with a fraudulent affliction, Hutchinson must have reasoned. But if there is no such thing as possession, then the people hung for witchcraft in 1692 and before, were innocent. Someone had to be blamed for shedding that innocent blood. To place blame on the clergy or the judges would be to implicitly question their current authority and trustworthiness. To blame a particular faction in Salem would invite enmity from their descendants. As women and servants, people with no power, the afflicted made a good scapegoat. The afflicted had, furthermore, disappeared -- married and moved away. It was possible to claim that they had gotten their just desserts, becoming "profligate persons, abandoned to all vice," who "passed their days in obscurity and contempt" (47).

That most of the afflicted were women has always been a fact fraught with significance, as is evident by Hutchinson's explicitly sexual idea of their just desserts. But the ideological import of the accused's gender began to increase in the last century, beginning with Upham's later work, Salem Witchcraft (his earlier Lectures is essentially a rehash of Hutchinson's thesis). Upham invents the idea of a secret circle of "girls," evidently from whole cloth. He elaborates on Hale's statements that the entire panic began with experiments in divination (1867 ii.2-3). He also writes character sketches for the members of his imagined circle. While Brattle had noted that the term "afflicted children" was a misnomer, since "there are several young men and women that are afflicted" (185), Upham has obscured not only age but gender, as all his embellishments act to specify the afflicted as "young girls" (Upham 1867 ii.2; my italics). In addition, as already noted, he flip-flops from the fraud theory to the hysteria theory, saying first that they were consummate actors (ii.112), then that they "became intoxicated" with power, "and were swept along by the frenzy they had occasioned" (ii.385). Finally, he claims that "their minds became inflamed and bewildered; and, at the same time, they grew expert in practicing and exhibiting the forms of pretended supernaturalism" (ii.387).

Between the earlier and later Upham, the Seneca Falls Conference had launched a feminist movement -- demanding suffrage, education, and property rights for women -- that was becoming increasingly powerful in 1867. The culture's treatment of women was beginning to alter, and the world-view of sexist discourse had to change with it. In addition, it was more possible to see the panic of 1692 as a blameless tragedy, in light of the recently ended Civil War and its attendant slaughter.

In the eighty year gap between Upham and Starkey, feminism continued to gain power, and the social position of women changed profoundly. Novels on Salem (previously in complete agreement with Hutchinson) began to follow Upham's lead and to flip-flop, portraying the "afflicted girls" as both faking and mentally ill. Grinstead, for example, makes the historical afflicted fakers, but makes his fictitious afflicted protagonist, Elva, fall in love with George Burroughs: she becomes guilt ridden over this adulterous passion and falls ill. Combined with her confused hallucinations (from a brew of Tituba's) of flying off on a broom and signing a pact with the Devil, by the time she testifies against Burroughs she no longer has all her marbles, and cannot tell what she remembers of him and what she imagines (Grinstead, Elva).

Starkey wrote The Devil in Massachusetts in 1949, in the aftermath of World War II and the Holocaust, and in the midst of the Red scare. While Huthcinson and Upham needed to fix blame for a tragedy, Starkey wants to emphasize that tragedy will emerge whenever "irrationality" and "hysteria" triumph over reason and good sense (15). As such, a "pure" hysteria theory is ideal. In addition, however, she is writing at a time when the concept of "teenager" is being invented, as adults began to be aware of and alarmed about adolescent culture and sexuality, especially the sexuality of adolescent and college age women. Thus, she writes that there were several "girls" who had reached the age of "sixteen, seventeen, and even twenty still manless and unprovided for, and these girls were instinct with repressed vitality, with all manner of cravings and urges" (32). Ann Putnam, on the other hand, is "on the verge of adolescence," another term new to 1949, and is "already subject to the preliminary strains of that difficult period, to unexplained pains and heaviness in the limbs, to dizziness and flashes of imagination so vivid that they sometimes resembled hallucinations" (38). Starkey expands on Upham's expansion of Hale's account of fortune telling, and portrays the conflict between church values and the illicit, inherently evil fortune telling as the source of the "girls'" hysteria, hysteria that seems built into adolescent women who don't get married right away.

While Starkey wants to absolve blame, Hansen wants to place it. However, he doesn't find the old fraud thesis compelling: hysteria fits in much better with 20th century ideas of women's passive and neurotic nature. In order to place blame and have hysteria, he decides that some of the accused were in fact real witches, and the assumed hysteria of the afflicted is a psychosomatic illness rising from their belief that they have been cursed.

While Hutchinson and Upham looked at the accounts of someone like Cotton Mather, saw a credulous and superstitious belief in the supernatural, and concluded that none of that person's account was trustworthy -- that the author was an unreliable narrator -- Hansen sees narratives amounting to a point-by-point account of Charcotian hysteria and concludes that the author was completely reliable. Hansen is then put in the position of the credulous fool, forced to explain away descriptions of things that could not possibly have happened. Some, such as smells of brimstone, he puts down to suggestion (173). Others, such as Margaret Rule's levitation to the ceiling, he cannot explain (184-185); Mather is a "scrupulously accurate observer" (176), and Mather has explicitly said that Margaret was antigravitied up there. Ergo, it is inexplicable.

In brief, Hansen (and Starkey) has a pin problem. If the afflicted were hysterical, that is, "not responsible for their actions" (43) and acting on unconscious impulses, then while they might have sought to injure themselves by biting their arms or sticking pins in their bodies (eg, SWP 80-81), they could not, by definition, have had the presence of mind to do so covertly, using sleight of hand so that judges, parents, and others would not see them doing it. His argument destroys itself.

Knowing that fortune telling and so on was constantly being condemned from the pulpit (and therefore constantly there), knowing that folk magic was a common place deed (Godbeer), Starkey and Hansen's explanations seem to imply that witch scares should have been happening all the time.

Part of the reason why multiple and mutually exclusive views can come out of the same source material is that our knowledge of Salem is limited in a peculiar way. The time and area concerned are quite limited, and it is possible for us to encapsulate it and make it into a dramatic set-piece. Our knowledge of that set-piece is remarkably detailed for the period: Salem is possibly the best-documented witch panic in all of European history. We can not only know who accused whom of what when, but can often find out the genealogies of the actors, the locations of their houses and farms, and how they willed that property on dying. At the same time, however, for all the available material, diaries, personal letters, newspapers, contemporary plays or stories interpreting what happened -- all of these are missing. We know many of the facts, but none of the motivations, none of the personal insights into the mood and self-concept of the time that are so much a part of, say, Elizabethan history. True, we have numerous sermons which reveal the religious world view of the era's ministers (not necessarily that of the average churchgoer) (4). But there is no guide to the parts of daily life beyond religion, no way to know the someones that Betsy Parris or Abigail Williams were. History is story-telling, and here the facts can form elaborate novel-like stories quite unlike most other history of the period. The lack of insight, however, gives us enormous freedom in imposing our own interpretations and prejudices on that story.

As a concise and clear-cut example of the kind of ahistorical history that I have been trying to show has been practiced in the complex and ambiguous case of the afflicted, take the origin of the Tituba myth. In 1831, early in the slavery debate, well before the Civil War or the Emancipation Proclamation, Charles Upham mentions Tituba only as "an Indian woman attached to Mr. Parris' family" who "by using severe treatment towards her, she was made to confess that the charge was true" (Lectures 22; italics original) In 1867, after the Civil War and in the middle of Reconstruction, Upham writes an elaborate fantasy in which "in all probability" Tituba (now named) taught voodoo to those who later became afflicted, and suggests that without this teaching, none of them would have ever begun to misbehave (Salem Witchcraft ii 2-5). In a world where the discourse of racism could be contained and defined by slavery, Tituba was a simple victim. With the abolition of slavery, racial discourse had to be rewritten, and for Upham, or perhaps for his folklore informants, the presence of a black woman at the origin point of the panic became fraught with new and sinister meanings.

In his study of 16th century demoniac possession in England and France, Walker notes that while there were three accepted explanations for possession -- supernatural, disease, or fraud -- a combination of factors was for some reason rarely conceived of (15). As human beings we tend to prefer simple, clear cut explanations. That some of the afflicted may have indeed been hearing or seeing things that weren't there, but shrewdly used that illness to their advantage in order to get as much attention and nursing as they wished -- such explanations don't fit into either our notions of good and evil people or our ideas of female passivity and sickliness. They may, however, be the best explanation for what happened in 1692.

Bernheimer, Charles and Claire Kahane, eds. In Dora's Case, 2nd ed. New York: Columbia U. P, 1990.

Boyer, Paul and Stephen Nissenbaum. Salem Possessed. Cambridge: Harvard U. P, 1974.

Boyer, Paul and Stephen Nissenbaum. The Salem Witchcraft Papers. New York: Da Capo, 1977. 3 vols.

Burr, George. Narratives of the Witchcraft Cases. New York(?): Charles Scribner's Sons, 1914.

Calef, Robert. More Wonders of the Invisible World. ed. Samuel Drake. New York: Burt Franklin, 1970. Reprints the 1866 edition. 2 vols numbered 2 and 3.

Evans, Martha Noel. Fits and Starts. Ithaca: Cornell U. P, 1991.

Demos, John. Entertaining Satan. New York: Oxford U.P, 1982.

Godbeer, Richard. The Devil's Dominion. Cambridge: Cambridge U.P, 1992.

Hansen, Chadwick. Witchcraft At Salem. New York: George Braziller, 1969.

Karlsen, Carol. The Devil in the Shape of a Woman. New York: W.W. Norton, 1987.

Hutchinson, Thomas. The History of the Colony and Province of Massachusetts-Bay. ed. Lawrence Mayo. Cambridge: Harvard U.P, 1936. vol. 2 of 3.

Mather, Cotton. The Angel of Bethesda. Barre, Mass: American Antiquarian Society, 1972.

Mather, Cotton. Wonders of the Invisible World. Samuel Drake, ed. New York: Burt Franklin, 1970. Reprints the 1866 edition.

Roy, Alec, ed. Hysteria. New York: John Wiley & Sons, 1982.

Slavney, Phillip. Perspectives on "Hysteria". Baltimore: Johns Hopkins U. P, 1990.

Starkey, Marion. The Devil in Massachusetts. New York: Anchor Books, 1949.

Temkin, Owsei. The Falling Sickness, 2nd ed. Baltimore: Johns Hopkins, 1971.

Thomas, Keith. Religion and the Decline of Magic. London: Weidenfeld and Nicolson, 1971.

Upham, Charles. Salem Witchcraft. Boston: Wiggin & Lunt, 1867.

Upham, Charles. Lectures on Witchcraft. Boston: Carter, Hendee, & Babcock, 1831.

Veith, Ilza. "Four Thousand Years of Hysteria." in Hysterical Personality. ed. Horowitz, Mardi. New York: Jason Aronson, 1977.

Walker, D.P. Unclean Spirits. Philidelphia: Pennsylvania U. P, 1981.

Weisman, Richard. Witchcraft, Magic, and Religion in 17th-Century Massachusetts. Amherst: Massachusetts U. P, 1984.

1 "Affect" is a psychiatric term used to describe inclusive behavior: words, body language, actions, and manner. It implies the intricate interaction of observer and observed, the "patient's affect" affecting the psychiatrist. While psychiatrists may describe a "patent's affect" in language implying that it has an objective reality, "affect" is in fact subjective -- the affect a person's behavior has on them. This last sense is the most important for us: "affect" is a matter of seeming. Back

2 I use this archaic term rather than the anachronistic and misleading word "physician." My underlying assumption throughout this essay is that modern historians have all too often failed to remember such vital distinctions, causing them to ascribe a modern world view on a distinctly premodern culture. Back

3 A wastebasket category contains the leftovers in a classification system. Hysteria and epilepsy are not diseases but labels for a number of disparate and unrelated ills with superficial similarities. Back

4 Godbeer suggests that the nuanced division in highly learned sermons between the devil's temptation and the sinner's free choice to sin was lost on a primarily unlettered layfolk (ch3). Back