SOURCE: Ramsey, Colin. “Cannibalism and Infant Killing: A System of ‘Demonizing’ Motifs in Indian Captivity Narratives.” Clio 24, no. 1 (fall 1994): 55-68.

[In the essay below, Ramsey links captivity narratives with the demonizing of Native Americans during the Puritan era.]

The Puritan phase of the Indian captivity narrative, both in its binary “good vs. evil” oppositions and in its imagery, established the paradigm for much of the subsequent development of the Indian captivity narrative form—helping to fix particular (and ethnocentric) views of the Indian in the American imagination, and thereby making those same images and motifs readily available for political and ideological manipulation. This paper will examine two such captivity narrative motifs, ubiquitous in Puritan captivities but by no means limited to them, namely, the motifs of Indian cannibalism and infanticide. I will discuss these motifs within Indian captivity narratives as a “demonology,” defined by Phillips Stevens, Jr., as

an elaborate body of belief about an evil force that is inexorably undermining the society's most cherished values and institutions. The evil it describes may be embodied in and perpetrated by a specific group, a minority which becomes the scapegoat. … Examples of such persecuted minorities in Western history are the Jews, over and over again; or Bolsheviks; or Japanese-Americans. … [And also] late medieval heresies, including witchcraft. …¹

I will add to those “demonized” groups the Native Americans, and consider the corresponding cultural results of such “demonology.” To do so, I will examine the history of the above motifs, cannibalism and infanticide, positing them within a tradition of demonizing imagery and folklore which pre-dates the Puritans considerably, and, finally, I will identify the same “demonizing” motifs within contemporary folklore, offering through the comparison a partial explanation for the pervasiveness of the motifs in captivity narratives, and more broadly, American culture, despite strong evidence that Indians are not and never were “cannibals” or “baby killers,” at least not in any systematic and culturally sanctioned way. Indeed, reports of cannibalism and infanticide, without basis in significant rational or empirical evidence, have regularly demonized many groups of people throughout Western history; as this paper will demonstrate, the motifs can be seen as a standardized means for characterizing any group which the dominant culture finds threatening as animalistic barbarians, as dehumanized and “other.”

Among the most important factors in the standardization of the motifs of cannibalism and infanticide within the Puritan version of the captivity narrative was the particular relationship of those narratives to the Puritan world view. From its beginnings, the Puritan captivity narrative worked as a myth which, according to Richard Slotkin, “reduced the Puritan state of mind … along with the events of colonization and settlement, into archetypal drama,” and which demanded that the Puritans reject the Indian “cannibal Eucharist.”² Most scholars echo Slotkin's assessment of the relation of the “Puritan mind” to Indian captivity narratives. William S. Simmons, for instance, argues that the Puritans ultimately saw the world as the scene of a continuing battle between the forces of light and darkness, saints and devils, and that this mental framework provided Puritans with a ready-made theory for interpreting cultural differences between themselves and the Native Americans—that the Indians were “cannibals” who “worshipped devils” and who themselves were “bewitched” or “witches.” Moreover, these beliefs were so pervasive that they became “matter of fact assumptions in the vocabulary of all the New English who wrote about Indian culture.”³

Robert Berkhofer further argues that the Puritan clerical and intellectual elite “picked up this method [the captivity narrative] for impressing the power of the Lord and the sinfulness of His people.”⁴ Moreover, Berkhofer argues that the early New England captivity narrative's best-seller status led to the “retention of its basic premise of the horror Whites suffered under Indian ‘enslavement’” (84-85). Thus the “horrors” of Indian captivity, as represented in Puritan captivity accounts, became the standard “horrors” of subsequent Indian captivity narratives. Puritan clerical authorities, such as the Mathers, seized upon the Indian captivity narrative as an instrument of manipulation; Indian captivity was cast as a trial of the spirit. Such narratives were designed to highlight God's great protecting providence. To help demonstrate the important role of the divine, these narratives portrayed the Indians as incalculably evil, cannibals and baby killers, creatures so evil that only by God's help could a Puritan survive captivity in their hands. In addition, the sensationalism of such narratives ensured a steady readership, and the Puritan clerics encouraged and contributed to their circulation. As long as the Indians were portrayed as barbarians, the narratives were a perfect template for religious instruction according to Puritan doctrine.

Mary Rowlandson's captivity narrative⁵ was the first in a long succession of Puritan captivity accounts that painted Indians as Satanic cannibalistic infant-killers. Rowlandson's language conveys this message implicitly: she describes the Indians as “a company of hellhounds” (2), who smash out the brains of some children and shoot others. “Thus we were butchered,” she writes, and all the while the Indians were “roaring, singing, ranting and insulting,”—the scene looked to Rowlandson like “a company of sheep torn by wolves” (3). Later in the narrative, the motifs of cannibalism and infanticide are blended together when Rowlandson inquires after her son. She writes:

I had not seen my son a pretty while, and here [sic] was an Indian of whom I made inquiry after him, and asked him when he saw him: he answered me, that such a time his master roasted him and that himself did eat a piece of him, as big as his two fingers, and that he was very good meat.

(15)⁶

Although Rowlandson later expresses disbelief at this story, attributing it to the Indian's “horrible addictedness to lying,”⁷ the motifs stuck, ready as the Puritans were to believe any story about the “devilish” Indians. Not surprisingly, given the Puritan's particular world view, the threat of cannibalism and the more overt accounts of the murder of children are picked up by, and amplified in, subsequent Puritan narratives, maximizing their de-humanizing effect. Cotton Mather, for example, in Decennium Luctuosum writes that during the march to Indian villages on the Kennebec following the so called “Salmon Falls Massacre,” a woman named Mary Plaisted complained that she could not march quickly because of the infant she carried in her arms. An obliging brave dashed out the child's brains against a tree (an extremely popular way of killing babies in the narratives), and told her to “walk faster than she did before.”⁸ Mather further relates that this was customary treatment of infants in Indian hands. Later in Mather's Decennium Luctuosum, the cannibalistic Indians are seen working in concert with the Quakers—who are after souls. The Indians eat the left-overs:

the Quakers have chosen the very same Frontiers and Outskirts, of the province [as the Indians] for their more Spiritual Assaults; … and have been Labouring [sic] incessantly … to Poison the Souls of poor people, in the very places, where the Bodies and Estates of the people have presently after been devoured by the Salvages.

(162)

As the political context shifted, so did the target of the demonization in the Puritan captivity narratives. In Reverend John Norton's The Redeemed Captive, published during the French and Indian War, the French are portrayed as cannibals along with the Indians:

After some time the Indians seemed to be in a Russle; and presently rushed up into the Watch-Box, bro't down the dead Corpse, carried it out of the Fort, scalpt it, and cut off the Head and Arms: A young Frenchman took one of the Arms and Flay'd it, roasted the Flesh, and offere'd some of it … to eat.⁹

During the Revolutionary War, captivity narratives paint the British along with, as always, the Indians as brutal infant killers. In A Narrative of the Capture and Treatment of John Dodge, Dodge reports that “the British governor of Detroit, Henry Hamilton, gave Indians a bounty for American Scalps, ordering both Indians and British soldiers alike ‘not to spare man, woman or child.’”¹⁰ Importantly, though Indian cannibalism and infanticide in the above narratives are depicted in association with various and shifting other groups, the Indian role remains the same. Thus with the help of a politically and religiously motivated clergy and a public “primed,” so-to-speak, by their own cultural baggage, the motifs of Indian cannibalism and infanticide so prevalent in Puritan captivities became standardized in captivity accounts and in American culture.

Nevertheless, one is moved to consider, if for no other reason than the very regularity with which the motifs appear, whether Native Americans ever did systematically kill children and/or eat people. Though it would be impossible to argue that no Native American ever killed a child or consumed human flesh, a brief review of current events shows that individual aberrations of that kind exist even today in virtually all cultures. Furthermore, when one takes into account both the social contexts from which captivity narratives arose and the dubious nature of their authorship, and compares the motifs in the narratives to the long tradition of inaccurate “demonizing” characterizations of “out” groups—based at best on misunderstandings and at worst on conscious political manipulation—one has considerable reason to doubt the accuracy of the reported incidents of cannibalism and infanticide in Indian captivity narratives.

For instance, all the narratives cited above were published either during, or right after, wars, and were propagandistic in nature. Mary Rowlandson's narrative was published during King Philip's War, which coincided with the rise of Increase Mather, and later, of his son Cotton: both were enormously influential in the formation and mytho-political use of the captivity narrative and had an interest in demonizing Native Americans. Indeed, the Rowlandson narrative underwent several subsequent surges of popularity during later wars. Greg Sieminski argues that the renewed popularity of Rowlandson's narrative during the 1770s was due to the narrative's very great effectiveness as propaganda. The colonists named the riots in Boston in March of 1770 the “Boston Massacre”; the Rowlandson narrative had supplied the images “they knew best: an Indian raid on a frontier settlement.”¹¹ The British forces involved in the riots became, by implication, cannibals and baby killers—they were effectively demonized. During the French and Indian War, as has been demonstrated above, the French were similarly “demonized,” and, as European expansion brought Whites into increasing conflict with Native Americans, the established demonology continued to offer ready excuses for the destruction of Native Americans and their cultures. Cannibals and infant killers deserved no better than to be “exterminated”—which would also conveniently free the land for White settlement. Clearly, the captivity narrative was an effective way of making an opponent in war, or later, a competitor for land, into a non-human, particularly through painting him as an infant killer and cannibal. As Levernier and Cohen argue, many captivities were designed to “horrify audiences into hating” the Indian, who “is painted as so irredeemably brutish that he deserves to be deprived of his lands.”¹²

In addition, and despite much recent revisionist scholarship, the historical record remains colonial in nature. Mainstream American history is still a history written by the conqueror; thus, negative stereotyping of the conquered remains largely unchallenged. Indeed, many early historians of colonial North America advanced, even expanded, such demonizing racial stereotypes. For example, Cotton Mather, who situated himself as the principal chronicler and historian of the Puritan experience in America, “elaborated this myth [the captivity narrative] as the historical framework for summarizing Indian-White relationships throughout the seventeenth century.”¹³ David Stannard argues, even more strongly, that all American history is “colonialist history,” written from a Eurocentric perspective which implies there was no history before Columbus. It sees the 30,000-year, or more, history of Native Americans prior to Columbus' arrival as inconsequential. These notions amount to a “political mythology” which reinforces the views held by the dominant culture—that, among other things, Native Americans were “primitive savages”¹⁴ living in darkness who were infant-killers and cannibals.

And not surprisingly, the genesis of the demonization of Native Americans as infant-killers and cannibals lies with Columbus. He reported that on an island called “Carib” there were people who were “very fierce” and who ate “human flesh.” Thus began the “line of savage images of the Indian as not only hostile but depraved.”¹⁵ Significantly, Columbus was informed of these inhuman natives through the Arawaks, who describe the people on the islands to the south as having “one eye and dogs' noses, who ate men, and when they took a man, they cut off his head and drank his blood and castrated him.”¹⁶ Columbus probably had little trouble believing this sort of thing; it reinforced many of the cultural notions already existent in Europe for hundreds of years—not just as a familiar demonology but as part of broader cultural notions of the Wild Man of folklore. Thus, “the depraved nature of man eaters captured the imagination of Europeans. … they were truly wild men of the lowest order, clearly beyond the pale of civilization” (45). Just so, reports of cannibalism and infanticide in the native inhabitants of the “New World” insured a readership and popularity back in Spain, and “served as a pretext for their [Indians'] enslavement to swell the labor supply in the new Spanish colonies” (46).

Though W. Arens' study The Man Eating Myth is controversial, if valid, it further discredits Columbus' accounts of cannibalism, and points to the kind of political motivation which greatly contributed to the stubborn recurrence of the motifs of Indian cannibalism and infanticide in captivity narratives. According to Arens, the Arawaks had every reason to provide a believing Columbus with demonizing stories about the Caribs. The Arawaks and the Caribs, the native inhabitants first encountered by Columbus, were two distinct groups with separate world views. Included in the world view of the Arawaks was fear and hatred of the more aggressive people on the southern islands in the Caribbean, and thus they were “eager to fill him [Columbus] in on the gossip about their enemies to the south.”¹⁷ In reality, when Columbus landed the following year to colonize the southern islands, the Caribs “ran from their villages at the sight of the Spaniards.” As Arens wryly puts it, “perhaps they too had heard of the existence of man eaters on distant islands” (46). One can only assume from Columbus' silence on the subject that the fleeing Caribs had human noses and two eyes apiece.

As with the Puritans, the political convenience of such a demonology, combined with the sometimes dubious nature of the sources for such stories and the colonizing cultures' predisposition to believe them, throws doubt on whether the Native Americans encountered by Columbus were really cannibals. What is not in doubt is that Europeans, and the West in general, have a long tradition of demonizing groups of people through accounts of cannibalism and baby-killing. That these characterizations often stem from misunderstandings of the groups being demonized is of little importance to the groups doing the demonizing since, if the demonization does not serve some overt political end, it seems always to, at least, reinforce the cultural assumptions of the dominant group.

That cannibalism and infanticide are an effective means to demonization is supported by the regularity with which the motifs have appeared in association with various scapegoat minority groups. For instance, while the origins of the cannibalism motif may be impossible to trace, Herodotus in the fifth century bc wrote of a people across the desert who lived lawlessly and “devoured human flesh” (10). In the second century after Christ, the Christian communities in the Roman Empire “were the object of strange suspicions and accusations”—Minucius Felix related the then popular belief about Christian religious rites: a child, covered in dough, would be stabbed by new initiates, who were unaware of what they were doing; the rest would then drink the child's blood and compete for his limbs.¹⁸ There are two plausible explanations for these presumably erroneous beliefs. First, the demonization served the political and economic ends of the Roman Senators. Marcus Aurelius and the Senate made it possible for local notables to purchase condemned criminals, such as the cannibalistic baby-killing Christians, for use in the gladiatorial games, at a much lower cost than to hire a gladiator. Even more expediently, the demonization of the Christians such that they could be executed in the games would reduce “an alien and potentially troublesome group.” Second, the Christian Eucharist is a kind of symbolic cannibalism and thus could easily be misinterpreted as the real thing by a society already suspicious of this strange new group in their midst (7-10). Though the above pattern proves nothing directly about reports of cannibalism and infanticide on the part of Native Americans, the motif patterns are strikingly similar. Certainly, the motifs have a long history as part of European “demonologies” and appear quite regularly throughout European history. The list of those demonized in this manner reads like an honor roll of European out groups: of so-called medieval heretics it was said “They cut their [babies', what else?] tender flesh all over with sharp knives and catch the stream of blood in basins”; of the Knights of the Templars when they began to threaten the power of King Philip of France, “[the Templars'] idol was anointed with the fat of roasted infants”; of the so-called “witches” of the fifteenth, sixteenth, and seventeenth centuries, “according to some writers of the time, to kill, cook, and eat a baby which had not yet been baptized was a witch's greatest pleasure” (19, 88, 100). Each of the above demonologies follows largely the same motif pattern and arises out of similar social contexts. As in the Indian captivity narrative, the groups being demonized are seen as threatening to the society doing the demonizing; and with each successive demonology, European culture became more bound to those same motifs. Is it any wonder that the Puritans decided the Native Americans, who were quite different and seemingly strange, were cannibals and baby killers?

The above demonology surrounding “witches” is especially pertinent to Puritan captivity accounts. In fact, what is essentially the demonology of the captivity narratives had become such a part of the Puritan consciousness at the end of King Philip's War that the Witch Trials were the direct result: “the strain and anxiety of revolutionary times culminated in the witchcraft delusion of 1692-93, in which Cotton Mather and his reworkings of the captivity mythology played a conspicuous part.”¹⁹ Mercy Short, a returned captive from the Salmon Falls “Massacre,” who was reportedly “possessed by devils,” helped Cotton Mather, who was working on his “history” of New England, “discover the common pattern” in each of the “assaults on pious New England—Indians and paganism, ministerial frauds and heretics, the assaults of the Quakers, the assaults of the royal governor on colonial prerogatives, and the final assault of the witches and the Invisible Kingdom in 1692” (129). As demonstrated earlier in this paper, all of the above groups were accused of cannibalism and infanticide at one time or another, and, as might be expected, Mather saw in all of the above groups a threat to “pious Puritan society.” The fears associated with Indians, notably, cannibalism and infant killing, were carried forward into the Witch Trials.

Mather, then, used Mercy Short as a symbol; the witch trials, as Slotkin notes, became none other than a Puritan attempt to exorcise from within themselves what they feared most: that they, as they imagined the Indians, were cannibalistic, baby-killing, primitive devils. In this instance, the tropes of demonology became not just damaging cultural stereotypes, but community wide, even cultural, psychotic delusions; the Puritans, under stress, framed the world in terms of the traditional motifs of demonology with which they were all too familiar; thus they saw cannibals and baby killers in their own midst, as they had “seen” the same qualities in the Indian, and as the Romans had “seen” them in the early Christians.

A further testament to the peculiar power of the motifs of cannibalism and infanticide, many folklore narratives, particularly those loosely categorized under the heading Contemporary Legend, often contain many of the same demonizing motif patterns as the Indian captivity narratives. Jeffrey Victor, for instance, has outlined the growth of satanic cult rumors in Jamestown, New York. He reports that the rumors of a “satanic cult” in the vicinity accumulated over time, gaining wider and wider circulation and force. The main rumor story, though there were countless variations, was that the “Satanists” were planning to “kidnap and sacrifice a blond, blue-eyed virgin.”²⁰ The community's responses to such rumor stories were massive: at school board meetings parents, worried over rumors of “ritual sacrifice” and fearing their children would be kidnapped for use in such sacrifices, demanded the speedy apprehension of the “Satanists” at all costs. Victor argues that a “symbolic interactionist” interpretation is the best explanation for the eruption of such rumor-panics:

The key insight of symbolic interactionism is that people interact with their environment through symbolic cultural meanings. The stories of rumor-panics embody cultural symbols, which a group of people can use to give meaning to their social reality. The rumor stories become “real” through the interaction process of the consensual validation of reality. … [It] is best understood as … a group's fearful reaction to collectively shared stories about immediately threatening circumstances.

(60)

There are some interesting similarities: the Puritans, reinforced by their clergy, saw the very possibility of Indian captivity as both extremely fearful and immediately threatening. It thus might be reasonably argued that the Puritans' culture, under increasing economic and cultural stress as time moved on, created the “reality” of cannibalistic infant-killing Indians through the “consensual validation of reality.” In other words, a similar mechanism of symbolic interaction could easily have produced narratives about cannibalistic Indians within the Puritan communities of the seventeenth and eighteenth centuries, just as such a mechanism produced the Satanic rumor-panic narratives in the modern communities of upstate New York. Moreover, the role of the Puritan clergy in the development, dissemination, and validation of such narratives resembles the role of the secular police “authorities” to Satanism rumor narratives in the modern context; both the Puritan clerics and the police positioned themselves as experts, and their interpretations of events were accepted with little or no empirical scrutiny. Such “experts” helped to bolster the believability of both sets of narratives, and contributed to their circulation.

And in fact, the symbolic outlines of the Satanist rumor-panic stories closely match those of the Puritan captivity narratives and other earlier demonologies, that children are “kidnapped and murdered” and their blood and body parts are used in cannibalistic rituals. Victor argues that these very motifs are the symbolic way that communities express anxieties about their future (52). Predictably, the “evil strangers” in these stories are usually people from some widely despised group in society.

The comparisons between these “urban belief tales” and the Western historical tradition of demonizing imagery are tantalizing. Bill Ellis notes that the current “urban-legend” of “The Castrated Boy,” a particularly nasty infanticide narrative, was widely believed in ancient Rome, “Non nova sed ingenita, … what we see today as our modern folklore may in fact be only universal human hates and anxieties in a contemporary cloak.”²¹ Jan Harold Brunvand cites in The Vanishing Hitchhiker a narrative called “The Hippie Baby Sitter.” In this legend, a young couple (evocative of the same scene of domestic happiness which is likewise shattered by the Indian raid in captivity narratives) hire an unknown “hippie” baby sitter. When they call to check up on the girl, she is almost incoherent (“on drugs” like all “wild-haired” hippies) but assures them all is well as she has placed their turkey in the oven. The couple rushes home “to find their baby roasting in the oven.”²² Compare this to a scene from the Fanny Kelly captivity Narrative of My Captivity Among the Sioux Indians:

One day, the Indians went into a house where they found a woman making bread. Her infant child lay in the cradle, unconscious of its fate. Snatching it from its little bed they thrust it into the heated oven, its screams torturing the wretched mother.²³

The similarities are striking: the “hippie” being demonized in the baby sitter narrative symbolizes the same fears that the “Indian” does in the Kelly narrative, i.e., “wildness,” nature (hippies are often “tree-huggers” and the environmental movement suffers under the same demonizing imagery), licentiousness (the “free-love” movement associated with “hippies”), in a word, everything which the Indian has come to symbolize in American culture. Even the form of the story in the Kelly narrative conforms to traditional folklore narratives: the “one day” beginning is ubiquitous in them. Again, the similarities are suggestive; the cultural tradition of demonization motifs in the captivity narratives, themselves influenced by cultural antecedents, might very easily have influenced the “Hippie Baby Sitter” narrative.

Nor are more scholarly works free from these culturally and mythologically driven motif patterns. Indeed, even the work of some contemporary scholars remains influenced by them; reports of Indian cannibalism and infanticide are still often taken at face value. Richard VanDerBeets writes that “cannibalism … [is] a practice more widespread among American Indians than is commonly understood.”²⁴ He cites as evidence for this generalization the same Fanny Kelly narrative as cited above, which is highly propagandistic in nature and which contains folkloristic accounts of babies being roasted. He approaches the Kelly narrative as utterly factual, assuming the barbarian nature of the Indian as a given. Yet VanDerBeets' assumptions are in direct conflict with recent studies, like that of W. Arens, which suggest that cannibalism and murder as ritual customs—including such “common knowledge” notions as that the Aztecs committed systematic and widespread human sacrifice or that South Pacific Islanders were cannibals—are extremely rare in any part of the world. All this points to, as Victor puts it, “how internalized legends easily create believability.”²⁵

And it is the internalized nature of these motifs which has made them such powerful tools for the creation of demonologies, not just in reference to the “scapegoating” of Native Americans via the Indian captivity narrative, but for a vast number of marginalized groups over the centuries. Though it is problematic to argue that cannibalism and infanticide motifs are human universals, univocal in interpretation as demonizing stereotypes, certainly they appear over and over again in very much the same form: the less-than-human and evil “other” (fill in any number of marginalized groups) kidnap and thoughtlessly torture and kill our children, and then proceed to dismember and eat their body parts and/or drink their blood. These motifs are not limited to Western sources; as Stevens points out, such motif patterns have been found in the Islamic world, in Asia, and in the traditional narratives of tribal cultures around the globe (129). Nor are these motifs limited to conquering or dominant cultures. As Arens reports in his study, Western anthropologists have been demonized by many of the cultures they were studying, according to the same motif pattern, as have colonial governments by the colonized peoples.²⁶ In any case, regardless of the groups involved, such demonologies seem always to develop within “situations of prolonged, unrelenting and often unexplainable social stress.”²⁷ Such socio/political stresses have been commonplace throughout world history, and the recurrence of the motif pattern within such historical contexts points to a great need for further research into their deeper symbolic significance. The political impact of such demonologies seems clear—they place whatever group is being demonized “beyond the pale of civilization”—but important questions remain: why do human cultures respond to particular kinds of stress in this way? Why are such motifs so easily internalized, and why do they remain so stubbornly pervasive, even into contemporary times, as a narrative pattern?

Although this does not absolutely prove that no Native American ever killed a child, we must remember that examples of such aberrant behavior exist, on an individual basis, in White culture. The point is, the belief that such things are sanctioned by other cultures less “civilized” than our own is what is truly erroneous. Furthermore, that belief is a mistake with terrible consequences. Thus, culturally determined misunderstandings and the political and ideological exploitation of such misunderstandings, in concert with an ever-growing cultural tradition that picks up on the motifs produced and transforms them into symbols—thereby forming the unconscious assumptions of a culture over the centuries—create a willingness to believe in the notion that there are “barbarous cannibalistic baby killers” at the gates. This is a powerful demonology, and the history and current status of the European culture's interaction with Native American cultures shows how devastating such a demonology can be. Further research and the continual re-examination of assumptions will help to reduce the commonalty of such “demonizing” motifs and the need to see others as barbarous cannibals—so that we might deal more rationally with “inner demons,” the “wild men within.”

Notes

1. Phillip Stevens, Jr., “‘New’ Legends: Some Perspectives from Anthropology,” Western Folklore 49 (1990):128.

2. Richard Slotkin, Regeneration through Violence: The Mythology of the American Frontier, 1600-1860 (Middletown, CT: Wesleyan UP, 1973), 94.

3. William S. Simmons, “Cultural Bias in the New England Puritans' Perception of Indians,” William and Mary Quarterly 42 (1981):56.

4. Robert F. Berkhofer, The White Man's Indian (New York: Knopf, 1978), 81.

5. All quotations are from Mary Rowlandson, A True History (London: J. Poole, 1682), the fourth edition of her narrative. This edition is generally considered to be the closest to the now lost first edition. Mary White Rowlandson, “A True History of the Captivity and Restoration of Mrs. Mary Rowlandson,” (rpt. of 1682 London ed.) in Narratives of Indian Captives, vol. 1 (New York: Garland, 1977).

6. The “roasting” children motif is a stable element in many types of folklore, from “Hansel and Gretel” to “The Hippie Baby Sitter” in Jan Harold Brunvand's, The Vanishing Hitchhiker: American Urban Legends and Their Meanings (New York: Norton, 1981).

7. Rowlandson, 15.

8. Cotton Mather, “Decennium Luctuosum,” (rpt. of 1699 ed.) in Narratives of Indian Captives, vol. 3 (New York: Garland, 1978), 56-57.

9. John Norton, The Redeemed Captive, (rpt. of 1748 ed.) in Narratives of Indian Captives, vol. 6 (New York: Garland, 1978), 10.

10. James A. Levernier and Hennig Cohen, eds., The Indians and Their Captives (Westport, CT: Greenwood, 1977), 50.

11. Greg Sieminski, “The Puritan Captivity Narrative and the Politics of the American Revolution,” American Quarterly 42 (1990):36-37.

12. Levernier and Cohen, xxii.

13. Slotkin, 71.

14. David E. Stannard, “Recounting the Fables of Infanticide and the Functions of Political Myth,” Journal of American Studies 25 (1991):381-83.

15. Berkhofer, 7.

16. Stanley L. Robe, “The Wild Men and Spain's Brave New World,” in The Wild Man Within: An Image in Western Thought from the Renaissance to Romanticism, ed. Edward Dudley and Maximilliam E. Novak (Pittsburgh: U of Pittsburgh P, 1973), 44.

17. W. Arens, The Man Eating Myth (New York: Oxford UP, 1979), 45. According to Arens, in a further irony of cultural misunderstandings, the Spanish mispronounced “Carib”—making it into “canib” and thus the word for man-eater is now “cannibal” and not “arawakibal” because it was the Arawaks whom Columbus encountered first (45-46).

18. Norman Cohn, Europe's Inner Demons (New York: Basic Books, 1975), 10.

19. Slotkin, 117.

20. Jeffrey Victor, “Satanic Cult Rumors as Contemporary Legend,” Western Folklore 49 (1990):52.

21. Bill Ellis, “De Legendis Urbis: Modern Legends in Ancient Rome,” Journal of American Folklore 96 (1983):206.

22. Brunvand, 65.

23. Fanny Kelly, Narrative of My Captivity Among the Sioux Indians (Hartford: Mutual Publishing Company, 1871), 117.

24. Richard VanDerBeets, “The Indian Captivity Narrative as Ritual,” American Literature 43 (1972):550.

25. Victor, 70.

26. See Arens, (1979).

27. Stevens, 129.