In The Devil in the Shape of a Woman, historian Carol F. Karlsen argues that established social attitudes toward women in seventeenth-century New England, and earlier centuries in Europe, explain why women were the primary victims of witch hunts in places like Salem, Fairfield, and elsewhere.[1] Indeed, she posits, women who willingly or inadvertently stepped out of line, who violated expected gender norms, were disproportionately likely to be accused in Puritan society. After establishing that roughly 80% of accused persons in New England from 1620 to 1725 were women, and that men represented both two-thirds of accusers and all of those in positions to decide the fates of the accused, Karlsen observes women’s deviant behaviors or states of affairs that drew Puritan male ire.[2] For instance: “Most witches in New England were middle-aged or old women eligible for inheritances because they had no brothers or sons.”[3] When husbands or fathers had no choice but to leave property to daughters and wives, this violated the favored and common patrilineal succession of the era. Further, women who committed the sins of “discontent, anger, envy, malice, seduction, lying, and pride,” which were strongly associated with their sex, failed to behave as proper Christian women and thus hinted at allegiances to the devil, putting them at risk of accusation.[4] The scholar is careful to note, however, that in the historical record accusers, prosecutors, juries, magistrates, and so on did not explicitly speak of such things as evidence of witchcraft.[5] But the trends suggest that concern over these deviations, whether subliminal or considered, played a role in the trials and executions.
Karlsen’s case is well-crafted. Part of its power is its simplicity: a preexisting ideology about women primed the (male and female) residents of towns like Salem to see witches in female form far more often than male. The fifth chapter could be considered the centerpiece of the work because it most closely examines the question of what a woman was — the view of her nature by the intensely patriarchal societies of Europe and how this view was adopted and modified, or left intact, by the Puritans. Christian Europe saw women as more evil than men.[6] They were of the same nature as Eve, who sought forbidden knowledge, betrayed God, and tempted man. Believed to be “created intellectually, morally, and physically weaker,” women were thought to have “more uncontrollable appetites” for sins like the seven above.[7] It is Karlsen’s exploration of this background that is foundational to the argument. If Christians had long seen women as more evil, a notion of witches-as-women in New England would have been a natural outgrowth (America’s early female religious dissenters, among other developments, added fuel to the fire).[8] The fact that associations between women and witchcraft existed in the European mind before the Puritans set foot in North America reinforces this.[9] Karlsen quotes fifteenth- and sixteenth-century writers: “More women than men are ministers of the devil,” “All witchcraft comes from carnal lust, which in women is insatiable,” “Because Christ did not permit women to administer his sacraments…they are given more authority than men in the administration of the devil’s execrations,” and so on.[10] Another penned that middle aged and older women received no sexual attention from men so they had to seek it from the devil.[11]
Indeed, Karlsen’s use of primary sources is appreciable. She extensively cites witchcraft trials in New England and works by ministers such as Cotton Mather, not only as anecdotal evidence but also, alongside public and family records, to tabulate data, primarily to show that women were special targets of the witch hunts and that most had or might receive property.[12] The author leaves little room for disputing that witch hunt victims were not quite model Puritan women, and that the Puritans believed that those who in any way stepped outside their “place in the social order were the very embodiments of evil,” and therefore had to be destroyed.[13] The work is organized along those lines, which is sensible and engaging. But a stumble occurs during a later dalliance with secondary sources.
One piece of the story — appearing on the last couple pages of the last chapter — stands out as underdeveloped. Karlsen posits that the physical ailments the Puritans blamed on possession, such as convulsions and trances, were psychological breaks, a “physical and emotional response to a set of social conditions,” indeed the social order itself.[14] The gender hierarchy and oppressive religious system were, in other words, too much to bear. Karlsen does cite anthropologists that have studied this phenomenon in other societies, where the minds of oppressed peoples, usually women, split from normalcy and enter states that allow them to disengage from and freely lash out at their oppressors, as, Karlsen argues, possessed New England women did.[15] But the causes of physical manifestations are such a significant part of the story that they deserve far more attention, indeed their own chapter (most of Karlsen’s final chapter explores the questions of who was most likely to be possessed, how they acted, and how the Puritans explained the phenomenon, though it is framed as a culturally-created power struggle early on).[16] This would allow Karlsen room to bring in more sources and better connect the New England story to other anthropological findings, and to flesh out the argument. For instance, she writes that convulsions and other altered states would have been “most common in women raised in particularly religious households,” but does not show that this was true for possessed women in New England.[17] How the ten men who were possessed fit into this hypothesis is unclear.[18] Things also grow interpretive, a perhaps necessary but always perilous endeavor: “…in their inability to eat for days on end, [possessed women] spoke to the depths of their emotional hunger and deprivation, perhaps as well to the denial of their sexual appetites.”[19] This is unsupported. In the dim light of speculation and limited attention, other causes of “possession,” such as historian Linnda Caporael’s ergotism theory (convulsions and hallucinations due to a fungus found in rye), remain enticing.[20] Minds are forced to remain open to causes beyond social pressures, and indeed to multiple, conjoining factors. How physical symptoms arose, of course, does not affect the thesis that prior ideology led to the targeting of women. The concern is whether the anthropological theory fit so well with Karlsen’s thesis — the targeting of women and the physical ailments being the results of a repressive society — that she both gravitated toward the latter and did not grant it the lengthier study it warranted.
Overall, Karlsen’s work is important. As she noted in her introduction, prior historians had given little focus to the role of gender in American witch hunts.[21] Their witch hunts had little to do with the suspicions about women’s nature or the dismay over women pushing against the gender hierarchy and religious order. Written in the late 1980s, The Devil in the Shape of a Woman represented a breakthrough and a turning point. It is a must read for anyone interested in the topic.
For more from the author, subscribe and follow or read his books.
[1] Carol F. Karlsen, The Devil in the Shape of a Woman: Witchcraft in Colonial New England (New York: W.W. Norton & Company, 1998), xiii-xiv. See chapter 5, especially pages 153-162, for European origins.
[2] Ibid., 47-48.
[3] Ibid., 117.
[4] Ibid., 119.
[5] Ibid., 153.
[6] Ibid., 155.
[7] Ibid.
[8] Ibid., 127-128 and chapter 6.
[9] Ibid., chapter 5.
[10] Ibid., 155-156.
[11] Ibid., 157.
[12] See for example ibid., 48-49, 102-103.
[13] Ibid., 181.
[14] Ibid., 248-251.
[15] Ibid., 246-247, with anthropologists cited on footnote 69, page 249, and footnote 71, page 251.
[16] Ibid., 231, 246.
[17] Ibid., 250.
[18] Ibid., 224.
[19] Ibid., 250.
[20] Linnda R. Caporael, “Ergotism: The Satan Loosed in Salem?,” Science 192, no. 4234 (1976): 21–26, http://www.jstor.org/stable/1741715.
[21] Karlsen, The Devil, xii-xiii.
Griffin 