Appendix: The Debate on the Existence of Cannibalism

In 1979 a heated debate in the field of anthropology on the existence of cannibalism was sparked by the publication of The Man-eating Myth: Anthropology and Anthropophagy by William Arens, an anthropologist at the State University of New York at Stoneybrook. Before turning to Arens’ claims and the responses of his critics, I give some general background on the nature and sources of scholarly studies of cannibalism, and other bodies of literature on the topic. There is far more nonfiction material on the topic than is generally known, and the idea of cannibalism is far more pervasive than we realize as a metaphor, a fear, and a reality among non-human fauna. When considering whether many cultures once had rituals involving the ingestion of human body parts, it is useful to have a sense of the larger universe of sources and references which surrounds the alleged practice. 149

Scholarly Studies of Cannibalism

Anthropology, archaeology, history, and historical documents contain the main body of literature on cannibalism, and virtually all of the literature on socially-sanctioned ritual or customary cannibalism. Histories record both starvation cannibalism and customary or ritual cannibalism. Archaeological studies may occasionally refer to recovered hieroglyphics or other ancient written or pictorial representations of customary or ritual cannibal acts; but they are more likely to involve the study of fossilized bones left from starvation or food cannibalism than the interpretation of material pertaining to customary or ritual cannibalism.

This social scientific literature involves five large and distinct subsets:

• Reports of starvation cannibalism and customary cannibalism in the official and unofficial histories of various Chinese dynasties going back to the second millennium BC [discussed in Chong, Cannibalism in China (1990)].150
• Western historical reports of cannibal practices, from Herodotus to Marco Polo. [Surveys of the oldest written Western sources on cannibalism are given in Peter-Rocher (1994) and Tannahill (1975).]
• Post-Columbian reports by travelers, explorers, missionaries, settlers, colonial governors, and the first amateur and professional anthropologists on cannibal practices among previously unknown simple cultures in Africa, North and South America, Southeast Asia, and Oceania. These start with the first reports from Columbus’ trips and continue through the first half of the twentieth century. One typical and widely-cited example (which I scanned in a nineteenth-century French translation of the original Spanish), written in 1568, is an 850-page diary, The True Story of the Conquest of New Spain, reconstructed from notes by Hernando Díaz del Castillo, who accompanied Hernando Cortes on several long trips to Cuba, Central America, and South America between 1514 and 1547 (1877).
• More carefully compiled ethnographies dating from the late nineteenth century to the present, generally prepared by professional anthropologists, describing and explaining the meaning of cannibal practices among little-known cultures, along with cross-cultural anthropological studies based on such ethnographies. An excellent recent example of an ethnographic study of cannibalism is that by Beth Conklin: “‘Thus Are Our Bodies, Thus Was Our Custom’: Mortuary Cannibalism in an Amazonian Society” (1995). An excellent cross-cultural analysis of exo-cannibal practices associated with war is given in The Comparative Ethnology of South American Indians, Volume V of the classic five-volume Handbook of South American Indians (Steward 1946–1959), in a chapter entitled “Warfare, cannibalism, and human trophies” by Alfred Metraux, the world’s foremost expert on warfare among South American indian tribes. Two other fine cross-cultural studies, which look at both endo- and exo-cannibalism in many parts of the world, are Peggy Reeves Sanday, Divine Hunger (1986) and Maurice Bloch and Jonathan Parry, Death and the Regeneration of Life (1982).
• Archaeological studies of human bones revealing signs of the consumption of flesh as food by other humans. Such studies have become scientifically complex over the past 20 years, with growing reliance on the study of bone breakage and cutmarks with electron microscopes and on radio-isotopic dating. One outstanding example is Timothy D. White’s Prehistoric Cannibalism at Mancos SM TUMR-2346 (1992), a natural-scientific study of the skeletal remains of 29 individuals at a single pueblo site in Colorado, dating from around 1100 AD—a study conducted, photographed, and written up as a model for rigorous assessment of archaeological evidence of cannibalism, and endorsed by reviewers as excellent for this purpose.

In addition to these kinds and contexts of thought and observation relating to cannibalism, there are related journalistic variants, prepared mainly for public entertainment and amusement. Works by Bernheim and Marriner are moderately sensationalistic books of this kind. These books and others like them typically recount a selection of stories drawn from the other kinds of sources described above. In some cases, such books may involve extensive research among primary sources (such as missionaries’ letters and reports) and secondary sources; and they may document instances of cannibalism not previously mentioned in the professional anthropological literature. Bernheim, for example, provides a detailed and well-documented review of cannibalism associated with famine in all parts of the world (Bernheim 1992, “Part II. Chronique de l’extrême faim,” pp 123–239). Popular nonfiction books on cannibalism and human sacrifice by Hogg (1966), Tannahill (1975), and Tierney (1989) are all carefully researched and documented, and all three are cited as sources in subsequent works by professional anthropologists.

Among cultural anthropologists, there are noticeable differences in the national bodies of literature on cannibalism published over the past 50 years. German scholars, more than others, have conducted semi-quantitative cross-cultural studies oriented to producing social-scientific generalizations. See, for example, Frank (1987), Wendt (1989), Volhard (1939), Peter-Rocher (1994), and Menninger (1995). French studies, more than others, tend to stress the psychological and social-psychological sources, meanings, and implications of customary cannibalism. See, for example, Erikson (1986), Pilette (a French Canadian, 1990, 1993), Siran (1989), Thomas (1980), Detienne (1979), Halm-Tisserant (1993), and Hubert and Mauss (1964). English-, Spanish-, and Portuguese-language anthropologists tend to eschew cross-cultural generalizations as well as psychological interpretations, and to focus instead on the particular ethnographic context of cannibal customs: that is, the ritual perceived within the larger framework meaning, importance, action, and need in a given culture. They also tend to look for signs of reaction in the practices of primitive tribes to encounters with the explorers, missionaries, colonialists, and slave traders with whom they were interacting when their practices were being observed and recorded. Good examples of this form of cultural anthropology are the collected articles in The Anthropology of Cannibalism, edited by Paula Brown and Donald Tuzin (1983), and in Warfare, Culture, and Environment, edited by R. Brian Ferguson (1984a). Finally, there are ethnologies and anthropological studies of specific cultures which devote considerable attention to cannibalism. Good examples are those by Abler and Logan (1988), Albert (1988), Balée (1984), Basso (1990), Castro (1992), Clastres (1974), Combès (1992), Dole (1974), Eves (1995), Goldman (1981), Halm-Tisserant (1993), Hassig (1988), MacCormack (1983), McGee (1983), Molet (1956), Saignes (1985), Walens (1981), Whiffen (1915), White (1993).

Given the contentious nature of the subject of cannibalism, it is unfortunate that there seems to have been relatively little cross-referencing and mutual fertilization among what we might call the French, German, and Anglo-American-Spanish traditions of study on the subject. Specifically, there has been no attempt to synthesize the results of quantitatively oriented, global studies of the phenomenon of cannibalism with the rich, psychologically- and ethnographically-oriented material derived from the careful study of specific cultures. This appears to be a result of linguistic limitations as well as intellectual orientation: Many scholars reference only works in English or French, or in English or German; some reference only works that appear (in the original or in translation) in their own language.151

Other Bodies of Literature about Cannibalism

Allusions to cannibalism appear frequently in disparate contexts. Along with reproduction, eating is the foremost activity required for the survival of the species; and the earliest form of eating, breast-feeding, combines the positive experiences of comfort, security, love, and probably sexual arousal with that of satiating hunger through food produced by the body of another person. Moreover, the aphorism ‘eat or be eaten’ expresses the primary relationships between humans and other species and between flora and fauna generally (McNeill 1980). Thus, it is not surprising that eating and the fear of being eaten are pervasive metaphors for many aspects of human existence.

Psychology: In the realm of psychology, cannibalism arises as a metaphor derived from—or an actual extension into adult life of—the ambivalent infantile impulse toward ‘oral incorporation.’ This is the postulated desire of the infant not merely to nurse at the mother’s breast, but to consume, that is, to physically incorporate and possess, the source of food, physical comfort, security, and love, that is, the breast of the mother. Freud’s important work, Three Essays on Sexuality, originally published in 1905 and revised in several subsequent editions through 1925 (Complete Works, Vol. VII, pp 130–243), includes an essay on “Infantile Sexuality,” in which Freud identifies the oral and anal loci of the earliest sexual sensations. About the oral he says (p 198):

We shall give the name of ‘pregenital’ to organizations of sexual life in which the genital zones have not yet taken over the predominant part. We have hitherto identified two such organizations....

The first of these is the oral or, as it might be called, cannibalistic pregenital sexual organization. Here sexual activity has not yet been separated from the ingestion of food; nor are opposite currents within the activity differentiated. The object of both activities is the same; the sexual aim consists in the incorporation of the object—the prototype of a process which, in the form of identification, is later to play such as important psychological part. A relic of this constructed phase of organization...may be seen in thumb-sucking, in which the sexual activity, detached from the nutritive activity, has substituted for the extraneous object one situated in the subject’s own body.

Like the ritual practices of cannibalism discussed later, the impulse toward oral incorporation combines two opposing impulses: loving a person or object—to the point of wanting to integrate that object into oneself; and being prepared to destroy the person or object for the sake of one’s own needs or desires. For infants, the hostile aspect of the desire to incorporate may be associated with frustration that food does not appear promptly on demand, or with the fear that when it does appear, it will never suffice.

A substantial body of literature at the boundary between psychology and anthropology explores several sets of relationships relevant to the role of innate oral-cannibalistic impulses: first, relationships among oedipal, incestuous, and oral-cannibalistic impulses;152 second, the relationship between impulses to oral incorporation and destruction on the one hand, and the development of the ego and the sense of identity on the other;153 and third, relationships between these impulses and their expression in various aspects of culture, including myths, literature, and alleged ritual and customary practices involving cannibalism.154

Fairy tales, myths, and literature: Around the world, myths and children’s stories are replete with tales of the cannibal consumption of children by parents or of human beings by gods or monsters.155 Hansel and Gretel, Jack and the Beanstalk, and Little Red Riding Hood are the most well known fairy tales in which the main element of suspense and drama is the child’s fear of being eaten. In all three cases, the child lacks ordinary adult protection: Hansel and Gretel are lost; Jack has no father and is off seeking to help support his mother; and Little Red Riding Hood has been sent on an errand through what might be expected to be dangerous woods. In all three cases, the child is at risk of being eaten by a less-than-human adult (an old crone, a giant, a wolf).

It is remarkable that in these and other fairy tales, the danger posed to children by villains—goblins, trolls, giants, witches, crones, and wolves—is not a plausible horror, such as being kidnapped, enslaved, sexually abused, beaten, or killed, but the implausible horror of being eaten. The underlying fear for which this is a metaphor is probably the fear of the loss of identity and control entailed in all of the more plausible horrors. However, the combination of the implausible fear of being eaten by a less-than-human adult and the absence of protective parents suggests the worst of all possible dangers: in time of need, parents, far from being protective, will abandon the child and save themselves by performing the monstrous act of killing and eating their own children.

There is a chilling truth which lies behind this nightmarish terror. Histories of famines suggest that eating children is generally the first form of starvation cannibalism, and, thus, the most frequent form. At the same time recognizing that (according to psychoanalytic theory) fears are generally paired with wishes, we can also interpret the child’s fear of being eaten as an inverted expression of the infantile desire not merely to nurse, but to gobble up the breast of the mother.

Several Greek myths involve cannibal acts in which children are eaten by parents or other older relatives (see, for example, Arfouilloux 1993; Halm-Tisserant 1993). The original creator-father, Chronos, eats his children, swallowing them whole before they can murder him to get at his throne. Because in Greek mythology “father-son antagonism is essentially that of eater and eaten,” the genealogical origin of the Greek gods is “a succession of devouring fathers and castrating sons that ends only with Zeus, who takes rather drastic precautions against filial rebellion,” (swallowing his own wife, Metis, so that he can possess her cunning and not be dethroned in turn by the sons he and she might produce) (Kilgour 1990, p 14). Other mythic cannibal events which recur as prominent themes in Greek culture are the consumption of Dionysus by the Titans, an act incorporated in religious rituals by worshipers of Dionysus; and Tereus’ unwitting consumption of his own child, served to him by his wife which is echoed in the Oedipus tragedy (Detienne 1979).

In Homer’s Odyssey, Odysseus, a Zeus-like King, is in danger of being eaten by Polyphemos, the cyclops, who symbolizes chaos; but Odysseus avoids that risk through quick, clever action. Ovid’s later Metamorphoses revolves around the same mythic acts of murder and cannibalism in Latin garb (for an extended analysis of both cases, see Kilgour 1990, pp 20–45).

Ancient written records: In addition to ancient Greek literature and art, ancient records from other parts of the world refer to cannibal threats or acts. The Old Testament alludes to starvation cannibalism as a form of punishment inflicted on parents by God. In Leviticus and in Deuteronomy, God tells Moses and Moses tells the Israelites that if they fail to keep God’s commandments, they will be besieged and reduced to eating their own children: “The Lord shall bring a nation against you from afar, from the ends of the earth, swooping down like a vulture.... They shall besiege you...so that you shall have to eat your own offspring, the flesh of the sons and daughters whom the Lord your God has given you in the stress of the siege.”156 Later, during a famine, a woman admits having killed and eaten her son (Kings II, 6:24–29), an act that is subsequently mourned in Lamentations (2:20) “See, o Lord, and behold; to whom else hast thou done thus: Whether it be women devouring their own offspring, their petted children; Or priest and prophet slain in the sanctuary of the Lord?” and (4:10) “Tender-hearted women with their own hands have cooked their children; They became their food, at the downfall of the daughter of my people.”

This sequence of Old Testament references is discussed by Lasine (1991), who points out (p 30) that “Parental cannibalism is also mentioned in several Assyrian treaties, as well as in Mesopotamian texts as old as the Curse of the Agade (lines 237–238 and Atrahasis (Neo-Assyrian version 2.6.35–37, 48–50). Although not all scholars agree that the curses in Deuteronomy 28 are modeled on Assyrian treaties, it is probable that readers of Deuteronomy would have taken the references to parental cannibalism as a conventional way of epitomizing the devastating results of treaty violation, whether or not the book’s author had intended to emulate Assyrian practice.”

In China the earliest reported reference to cannibalism, cited by Chong (1990, p 47), involves Chou Wang, a Yin ruler whose reign ended in 1122 BC. According to the ancient historian Han Fei Tzu, Chou Wang punished three officials who rebuked him for cannibalism and cruelty by cooking, preserving, and eating their flesh. The first reported case of starvation cannibalism in China occurred during a war-time siege of the capital of Sung in May 594 BC. As Chong describes the incident (p 45), “When the city ran out of provisions, the people sent one of their agents, Hua Yuan, under cover of night into the enemy camp. The agent told the general of the Ch’u army, ‘My master has sent me to inform you of our distress. In the city, we are exchanging our children and eating them and splitting up their bones for fuel.’ Soon afterwards, peace was declared.” Chong comments that the event “is described in many Chinese classics with the words i tzu erh shih (people exchanging one another’s children for food)” (p 45) and he cites five primary historical sources.

In ancient India and Egypt—the other parts of the world for which there are documents dating to 1000 BC—cannibalism appears as a metaphor in creation myths and related religious practices. For example, in the Rig Veda, which dates from the second millennium BC, the earliest the gods, who were children of the primeval man Prajapati, sacrificed Prajapati to himself to create the universe (Flesh and Blood: A History of the Cannibal Complex, Tannahill, pp 22–23):

From his body he made the animals

of air and wood and village....

Thence were born horses,

and all beings with two rows of teeth.

Thence were born cattle,

and thence goats and sheep...,

From his navel came the air,

from his head there came the sky

from his feet the earth, the four quarters from his ear,

thus they fashioned the worlds...157

While this creation story was not explicitly a cannibal event, the language suggests images of body parts being consumed and transformed. Similarly, as the ancient Egyptian myth of the murder, dismemberment, and dispersed burial of Osiris by his brother Set, followed by the reassembling and resurrection of Osiris through the efforts of his wife Isis, is echoed in an indirect form of cannibal sacrifice in which the flesh, bones, and blood of sacrificed individuals are scattered over the fields so that the gods who control the earth and heavens can ‘swallow up’ the sacrificial victim and, in exchange, make the fields fertile (Tannahill, pp 20–21).

Religious practices: Religious practices that involve the sacrifice of an animal often include consumption of parts of the animal by priests; and many comparative studies of the origin and meaning of religious sacrifice include discussion of certain associated cannibal practices.158 In Aztec human sacrifices, priests extracted (but did not eat) the heart and the blood pumping through the heart as the principal offering to the gods, and then pushed the body down the pyramid steps to those who had captured the victim, who cooked and ate the arms and legs (Davies 1981). As discussed later, cannibal rituals in simpler societies, rather than being an adjunct to human sacrifice to gods, tended to be the central religious or spiritual ritual, for which the killing of captives was, in some cases, an integral part.

The Christian ritual of the ‘Eucharist’ offers a contemporary parallel to early religious forms of sacrifice and cannibalism. In this case, the consumption of bread and wine, representing the body and blood of Jesus, is conducted in remembrance of, and to benefit from, the sacrifice of his life, which was made to expiate the sins of humanity. This symbolism resembles that of non-Christian religious practices, in which the consumption (or dispersal throughout fields) of some part of a sacrificed human or animal by priests on behalf of the community, or by members of the community, creates a line of communication that permits or persuades the gods to endow human beings with benefits, such as a good crop or fertility, or, in a later era, grace, forgiveness, and redemption.

The parallel between the Christian ritual of communion and early forms of human sacrifice and cannibalism are discussed less in the theological and religious literature of Christianity than in anthropological and humanistic literary studies. The exception that proves this rule is provided by John Fenton, honorary canon of Christ Church, Oxford, in a 1991 article entitled “Eating People,” published in the venerable quarterly Theology. Fenton argues that the seeming parallels between the Christian Eucharist and primitive practices of human sacrifice-cum-cannibalism must involve a misinterpretation, perhaps fostered by the disciple Paul (believed to be a Greek gentile who had converted to Judaism), because to Jews of Jesus’ era, sacrifice and cannibalism represented an abomination, not an accepted form of religious ritual. Noting that throughout the Bible, cannibalistic metaphors indicate hostility, aggression, and destruction,159 Fenton suggests (p 421) that in the Gospel account of the Last Supper, Jesus “commanded the disciples to take the bread and he passed them the cup to drink, in order to symbolize their responsibility for his death.”

Primate and other non-human biology: In the biological sciences, the ‘conspecific’ consumption of newborn offspring by parents, by each other, or by other adults has been studied in recent years as a counter-intuitive example of how genetic endowment and environment may combine in survival-oriented behavior. A review published in Science (Mock 1992) of the first book-length collection of survey articles on this topic, Cannibalism: Ecology and Evolution among Diverse Taxa, is worth citing at some length:

In the 1960s and ‘70s, refinements of natural selection theory led most biologists to realize that phenotypic traits, including behavior patterns, evolve because of net benefit to the individual’s inclusive fitness. One consequence of this paradigm shift...was that sporadically reported cases of vile or unsavory behaviors performed by animals (such as rape, slavery, infanticide, mate-desertion, and cannibalism) could not simply be assumed to be pathological or aberrant any more.... One could no longer seek comfort in the meager records (many such behaviors are inherently rare and hard to witness) or dismiss them airily as mere by products of captivity....

.... This led to exponential growth on several fronts. Ecological predictions began to emerge, specifying the context in which these behaviors should be found. Eventually, reviews began to appear. The current volume can be viewed, therefore, as the formal rite of passage for the fascinating topic of cannibalism as a very respectable area in evolutionary biology.

Fifteen review chapters by 17 scientists make it abundantly clear that there is nothing particularly astonishing or freakish about the ingestion of conspecific tissue. Such habits have evolved repeatedly as a solution to various problems, often (but not always) involving food shortages.160

While most cases of cannibalism in animals involve insects, fish or amphibians, cannibalism of the young by mammals has been observed. For example, lions which take over a pride may eat the young of their predecessors (Leakey and Lewin 1977, p 220). In recent decades, close observation of chimpanzees (the anthropoid apes which most closely resemble homo sapiens) has revealed cannibal behavior by at least one mother (observed by Jane Goodall) and by a few chief males in family clusters who have eaten a newborn male in cases where there was reason to doubt the paternity (see Hamai et al 1992; Nishida and Kawanaka 1985; Tartabini 1991).

Criminology: The literature of crime and criminology contains occasional references to the law pertaining to cannibalism or to cases of individuals convicted of murder and cannibalism. The latter generally fall into two groups: psychotic or sociopathic serial murderers, and individuals at risk of starving to death after being stranded in a shipwreck or comparable accident. Bernheim and Stavridès (1992) and Marriner (1992) provide brief accounts of the cannibal acts of the following convicted cannibal-murderers (in most cases, multiple murderers) of the nineteenth and twentieth century:

• 1817 the farmer’s wife at Selestat (Bernheim)
• 1824 Leger (Bernheim)
• 1824 necrophage de Saint-Amand (Bernheim)
• 1826 Maria de las Dolores (Bernheim)
• 1858 Comstock (Bernheim)
• 1864 Tirsch (Bernheim)
• 1872 Verzeni (Bernheim)
• 1879 Garayo (Bernheim)
• 1881 Mc T (Bernheim)
• 1891 Eugene L. (Bernheim)
• 1894 Vacher (Bernheim)
• 1897 Luetgert (Marriner)
• 1913–1921 Carl Wilhelm Grossman (Bernheim / Marriner)
• 1918–1924 George Haarmann (Bernheim / Marriner)
• 1921–1924 Karl Denke (Marriner)
• 1928 Albert Fish (Bernheim / Marriner /Heimer)
• 1929 Peter Kurten (Bernheim)
• 1949 John George Haigh (Bernheim/ Marriner)
• 1950 Edward Howard Gein (Bernheim)
• 1955–1976 Kroll (Marriner)
• 1969 Modzieliewski (Bernheim)
• 1970 Frazier (Marriner)
• 1970 Kemper (Marriner)
• 1970 Mullin (Marriner)
• 1970 Stanley Dean Baker (Marriner)
• 1976 Chase (Bernheim)
• 1979 Clement X (Bernheim)
• 1980 Djoumagaliev (Bernheim)
• 1981 Anna (Marriner)
• 1981 Issed Sagawa (Bernheim)
• 1986 Heidnik (Marriner)
• 1986 Weber (Marriner)
• 1989 Rakowitz (Marriner)
• 1991 Dahmer (Bernheim)
• 1992 Chikatilo (Bernheim)

Marriner also recounts in some detail a case mentioned in many sources: In 1611 Countess Elisabeth of Báthory, who ruled a large castle and estate after her husband’s death in 1604, was convicted of having tortured and killed some 600 girls and young women between 1604 and 1610 in order to daily bathe in and drink their blood, which she thought would keep her young. At the time of her arrest on 30 December 1610, her chief means of ‘harvesting’ blood was to put a girl in a narrow iron cage, with nails pointed inward to puncture the skin, and suspend the cage from the ceiling while she sat under it, bathing in a shower of blood (Marriner, pp 129–130).

Stranded, starving travelers: The best known cases of starvation cannibalism by travelers are those of the Donner party members, who were stranded in snowstorms in Nevada while trying to cross the Rocky Mountains in November 1846; by crew members of the Mignonette, which sank in the ocean on 3 July 1884, leaving several officers in a dinghy hundreds of miles from land; and by survivors of the plane crash in the Andes in October 1972, who lived on the flesh of the dead (whose bodies had been frozen) for 70 days. In these cases, the cultures concerned (US, British, and Argentine) tended to judge the cannibal actions as morally warranted and legitimate if the victims were already dead, and as morally wrong and criminal, but only slightly more so, if the victims were on the verge of death or if they were selected to be killed in order to help save a larger group. The British trial of the Mignonette survivors, who had killed to eat, involved the first legal use of the ‘necessity’ defense for cannibal murder. In that case, the defendants, who openly admitted what they had done and argued that it was justified, were convicted of murder and sentenced to death; but Queen Victoria commuted the death sentence to six months in jail (Marriner).

William Arens and His Critics: A Comprehensive Review

As noted in Chapter 5, only one attempt has ever been made to systematically identify, assess, and analyze all reported practices of customary cannibalism in all parts of the world: a 550-page work entitled Kannibalismus by anthropologist Evald Volhard, published in Germany in 1939 and unfortunately never translated into English or French. Volhard uses some 800 sources to identify 914 cultural or linguistic areas (bands, tribes, or larger groupings) for which there are reported practices of cannibalism. Though catalogued in the Harvard library since 1948 and universally cited by German anthropologists writing on the topic, the book was not used as a reference by any of the three English-speaking scholars who have attempted, on a much more modest scale, to provide some useful generalizations about the practice of cannibalism around the world: William Arens, who claims that customary cannibalism did not exist, and Peggy Reeves Sanday (author of Divine Hunger, 1986) and Eli Sagan (author of Cannibalism: Human Aggression and Cultural Form, 1974), who attempt to give global overviews of ritual or customary cannibalism as a cultural practice with examples drawn from diverse cultures.

In this instance, the barriers of language and conflicting scholarly traditions have led to neglect of important sources of a kind that would not be tolerated in the natural sciences because it precludes cumulative learning. Due to the scale of his undertaking, Volhard relied in part on nineteenth-century secondary compilations of primary source material; and nearly all of the many primary sources he cites were published before 1930. No survey article or book on cannibalism reviews the main findings of respected scholars over the past 60–70 years, regardless of language or scholarly tradition. As a result, Arens’ sweeping claims put anthropologists who attempt to critique his work, along with those reviewing the critiques, in the position of providing a less than thorough assessment of his claims, or producing the equivalent of a survey article as the basis of a thorough assessment. Between the material in Chapter 5 and that presented here, I have attempted, in a very brief fashion, to provide a survey article.

When claiming that there is no hard evidence for the existence of food cannibalism or for widespread practice of the consumption of human body parts in customary or ritual cannibalism, Arens argues that early anthropologists’ reports regarding such practices represented projections in which they extrapolated from circumstantial evidence, expressing ethnocentric expectations of ‘the other’ which have been common in all cultures. Moreover, Arens claims, recent anthropological studies of cannibalism have accepted and repeated the earlier reports uncritically, with equally ethnocentric credulity.

To support these claims, Arens reviews some of the original source material for a few of the most widely cited cases of cannibalism. His assessment, presented in three main chapters, covers three groups of cases:

• ‘Classic’ man-eaters discovered by Columbus and others in Central and South America in the sixteenth century: the ‘Caribs’ (from whom the term cannibal derives), the Aztecs of Mexico, and the Tupinamba of Brazil.
• ‘Contemporary’ man-eaters studied by twentieth-century social scientists: the Amahuaca of Brazil, the Fore of New Guinea, and the Azande and several other tribes in Africa
• ‘Prehistoric’ man-eaters of North America studied by contemporary anthropologists and archaeologists: the Iroquois and the Anasazi.

Arens argues that classic cases of cannibalism represented: (1) the wishful thinking or deliberate lies of Spanish slave traders in the Caribbean and Mexico, who were legally forbidden to enslave indigenous people except from among tribes that practiced cannibalism; (2) sensationalism by early travelers and missionaries to Brazil, who wanted to impress their European audience and sell books or impress their denominational financial supporters; or (3) plagiarism of earlier writers by later ones. Regarding contemporary cases, Arens argues that purported eye-witness reports by modern scientists (Dole in the case of the Amahuaca, and Alpers and Gadjusek in the case of the Fore) are faulty because there is reason to believe that they were extrapolations or presumptions based on observed activities that actually stopped short of eating. Finally, Arens argues that the archaeological evidence adduced to support pre-Columbian cannibalism among the Iroquois and the Anasazi is inconclusive and could represent evidence of other processes, such as ‘secondary burial,’ disturbances of graves by wild animals, or accidents.

Because Arens accuses living and dead professional anthropologists of being ethnocentric and sloppy, his book caused a great stir; and it is cited in virtually every subsequent study, occasionally in contexts that support his claims, mainly by scholars who disagree with him.

Several of the original reviewers cast doubt on Arens’ thesis, but do so in such a cautious manner as to leave the reader uncertain about the actual practice of cannibalism. For example, Ivan Brady of SUNY-Oswego, writing in the American Anthropologist, asks, rhetorically, whether cannibalism exists “on the scale and in the manner in which anthropologists (some or all) have assumed in the past?” In reply Brady answers: “Arens does not think so. I agree, but suggest that the discrepancy is neither so wide as he thinks—not everyone is equally reckless with wisdom and facts—nor exists for exactly the same reasons.” Similarly, Vincent Crapanzano of Queens College and the City University of New York Graduate Center, writing in the New York Times Book Review, comments that “Mr. Arens’ book is poorly written, repetitive, snide. His sloppiness is especially regrettable because it lessens the impact of his basic, significant suggestion: that the degree which cannibalism has been practiced has been exaggerated.”

Other reviewers are much more critical, explicitly condemning Arens for his own sloppy scholarship and for egregiously misrepresenting the extent of overstatement and under-documentation in primary sources, based on their knowledge of the literature on which he draws.

Ulla Wagner of the University of Stockholm states: “H]is presentation of the data shows the bias and selectivity that furthers his case. Nowhere does he take up cases where the informants themselves have stated that they practiced cannibalism. For example, when he quotes Hallpike (on p 99), we are led to believe that it is only others that impute cannibalism to certain Papuan groups, when in fact Hallpike gives several quotations which refer to the informant’s own group (Hallpike 1977).” Citing Arens’ sweeping condemnation of the tendency of anthropologists to give credence to reports of cannibalism, “Merely entertaining the possibility of a universal taboo on cannibalism would affect the public’s image and support of the discipline,” Wagner comments that Arens is implying, first, that anthropologists are not to be trusted; second, that they deal only with the exotic; and third, that they have vested interests in maintaining cultural boundaries. To these points she responds that the integrity of anthropologists is probably neither better nor worse than that of scholars in other fields, and that “the other two points are obviously nonsense. Cultural differences are definitely not figments of the anthropological imagination. They are very real, and I cannot see how the endeavor to make understandable that which is strange and alien could ever be construed as being its very opposite.”

In the Anthropological Quarterly, James W. Springer of Northern Illinois University says: “His methods of evaluation are faulty and his critical attitudes amount to little more than a refusal to believe any statement of the existence of cannibalism, combined with a variety of impeachment of the motives of those who report it.” Springer then cites one of Arens’ many sweeping but incorrect claims about specific sources and cultures: “The collected documents of the Jesuit missionaries (Thwaites 1959), often referred to as the source for Iroquois cruelty and cannibalism, do not contain an eyewitness description of the latter deed.” Springer observes that the source in question, a 72-volume work called the Jesuit Relations, contains abundant eyewitness accounts of cannibalism by Indians. For this he cites the index page on which references to these accounts may be found (vol. 72: 124). As specific examples, Springer points out that the “narratives of Father Jogues (vol. 39: 19–221) and donne Regnaut (vol. 34: 25–37) show first-hand knowledge” of the practice.

Thomas Abler of the University of Waterloo, another specialist on the Iroquois, reviewing Arens’ book in Ethnohistory, argues that Arens cannot possibly have looked even at the volume indexes to the Jesuit Relations, which contain references to “Cannibalism—Iroquoi” in 31 volumes of the 72-volume series. Having reviewed all of the indexed references, Abler argues that even if one discounts most of the Indian statements on the grounds that the informants may have tended to boast about their own valor with references to cannibal acts and to vilify enemies with exaggerated claims about their barbarism, there are first-hand accounts by Jesuits, captives, and others which there is no reason to doubt; and he cites, in addition to the two accounts cited by Springer, the following passages: vols. 39: 81, 52:169–171, 53: 139, 62: 75, and 62: 91. Abler also cites four seventeenth-century eyewitness reports from sources other than the Jesuit Relations. In addition, both Springer and Abler review a variety of archaeological sources for claims of cannibalism among the Iroquois in the prehistoric period (roughly 1300–1500), and portions of the evidence supporting the view that cannibalism did occur in that period as well.

The two most damning critiques of Arens work are those of P.G. Rivière of Oxford, a student of the Tupian peoples of Brazil, and Marshall Sahlins, the pre-eminent expert on sixteenth to nineteenth century sources concerning cannibalism on the Fiji and Marquesas Islands and among the Maori of New Zealand. Writing in the journal Man, Rivière details examples of Arens’ inaccuracy, incompleteness, and unwarranted inferences in dealing with the primary sources on Tupi cannibalism. He concludes his review as follows (pp 204–205):

[Arens’ book] has forced me to look again at the sources on Tupi cannibalism and, without doing a complete assessment of the material, I am more than ever confirmed in the opinion that the Tupi-speaking Indians of the Brazilian coast in the sixteenth and seventeenth centuries practiced cannibalism.

Bad books do not usually deserve long reviews, and I have given this one more attention because it is also a dangerous book. With little work and less scholarship, it may well be the origin of a myth.

As indicated the discussion of later anthropological sources below, this fear has been fulfilled to some extent.

In late 1978 in the New York Review of Books, Marshall Sahlins reviewed Cannibals and Kings by Marvin Harris, a book which argues, among other things, that the main reason for cannibalism among simple societies was protein deficiency. Sahlins’ review, which stressed the ritual and symbolic nature of most cannibalism, prompted Arens to write an article-length letter to the editor laying out the main arguments and some of the evidence from his forthcoming book and concluding:

From what I can gather from an extensive review of the literature, every human culture, sub-culture, religion, cult and sect, including our own, has been labeled cannibalistic at one time or another by someone. Yet no one has ever observed this purported cultural universal. This should give pause to consider whether we are dealing with historical reality or an extremely satisfactory myth.

Arens’ letter was published in the March 22, 1979, New York Review of Books, along with a response by Sahlins, under the heading “Cannibalism: An Exchange.” In his response, Sahlins gives long excerpts from the primary sources of eye-witness reports of cannibalism in the seventeenth to nineteenth centuries among the Aztecs, the Maoris, and the Fijians. Noting that he has seen an advance copy of The Man-Eating Myth and is outraged that peer review notwithstanding, Arens is “about to publish a book under the imprint of a famous university press [Oxford] which expounds on the thesis of his letter,” Sahlins concludes with the following scathing attack:

It all follows a familiar American pattern of enterprising social science journalism:

Professor X puts out some outrageous theory, such as the Nazis really didn’t kill the Jews, human civilization comes from another planet, or there is no such thing as cannibalism. Since facts are plainly against him, X’s main argument consists of the expression, in the highest moral tones, of his own disregard for all available evidence to the contrary. He rises instead to the more elevated analytical plane of ad hominem attack on the authors of the primary sources and those credulous enough to believe them. All this provokes Y and Z to issue a rejoinder, such as this one. X now becomes “the controversial Professor X” and his book is respectfully reviewed by non-professionals in Time, Newsweek, and The New Yorker. There follow appearances on radio, TV, and in the columns of the daily newspapers.

The effect is to do away with the usual standards of scholarly value, such as use of evidence or quality of research, as criteria of academic success. Like the marketing of automobiles or toothpaste, academic research is submitted to the one characteristic sense of criticism left to American society: Caveat Emptor [no guarantees unless expressly stated]. So the publishing decisions of academic presses, and ultimately the nature of scholarly research are drawn irresistibly into the orbit of the average common opinion of the consuming public. It’s a scandal.

In my view, the evidence adduced by these reviewers and their more general professional judgment about the integrity and interpretation of the primary sources they cite and others like them provide a sufficient reason to conclude that Arens is wrong: the cultures widely believed to have practiced ritual and customary cannibalism did so. Because of the importance to my own study of not falling prey to ethnocentric exaggeration, I was, however, lef t with nagging doubts about the real extent of ritual cannibalism, as distinct from wanton treatment of body parts which might have been cooked and preserved as trophies but not actually eaten, and which natives might claim to have eaten in order to elicit approval or shocked disapproval from European observers.

To lay these doubts to rest, I surveyed the professional anthropological literature on cannibalism published after Arens’ book had appeared, in order to see whether professional anthropologists had subsequently assembled more comprehensive and carefully re-assessed evidence of the practice of cannibalism.

I found that most professional books and articles on cannibalism published since 1980 list Arens as a source and explain their reasons for disagreeing with him. In some cases, Arens’ claims are dismissed briefly in introductory remarks. The following passage from Sanday illustrates this approach: “Although [Arens] is correct in asserting that the attribution of cannibalism is sometimes a projection of moral superiority, he is incorrect in arguing that cannibalism has never existed. Contrary to his assertion that no one has ever observed cannibalism, reliable eyewitness reports do exist.” Sanday cites five eyewitness reports from diverse periods, including two from the twentieth century.

In other recent studies, new assessments of historical and contemporary material are presented with the purpose, in part, of showing that cannibalism was commonly practiced as a ritual or custom in the cultures in question. Generally speaking, the larger purpose of these studies has been to provide an anthropological description and interpretation of the practice, not just mere confirmation of its existence. The following 20 articles and books published since 1983 all cite Arens’ book, dispute his claims, and offer new evidence concerning actual practices of cannibalism:

Abler (1992) “Scalping, Torture, Cannibalism and Rape: An Ethnohistorical Analysis of Conflicting Cultural Values in War”

Abler and Logan (1988) “Florescence and Demise of Iroquoian Cannibalism: Human Sacrifice and Malinowski’s hypothesis”

Barber (1992) “Archaeology, ethnography and the record of Maori cannibalism before 1815: A critical review”

Bowden (1984) “Maori Cannibalism: An Interpretation”

Brown and Tuzin, eds. (1983) The Ethnography of Cannibalism

Castro, Viveiros de (1992) From the Enemy’s Point of View

Chong (1990) Cannibalism in China

Clunie (1987) “Rokotui Dreketi’s human skull: yaqona cup?”

Combès (1992) La Tragedie Cannibale Chez les Anciens Tupi-Guarani [The cannibal tragedy among the ancient Tupi-Guarani]

Conklin (1995) “‘Thus Are Our Bodies, Thus Was Our Custom’: Mortuary Cannibalism in an Amazonian Society”

Ernandes (1992) “Serotonin Deficiency Hypothesis Explaining the Aztec Human Sacrifice/Cannibalism Complex”

Forsyth (1985) “Three Cheers for Hans Staden: The Case for Brazilian Cannibalism”

Jamieson (1983) “An Examination of Prisoner-Sacrifice and Cannibalism at the St. Lawrence Iroquoian Roebuck Site”

Liep (1987) “Kannibaler og Kulier: Antropofagiske Scener fra en Sydhavso” [Cannibals and coolies: anthropophagic scenes from the South Pacific]

Obeyesekere (1992) “British Cannibals: Contemplation of an Event in the Death and Resurrection of James Cook, Explorer”

Saignes (1985) “La Guerre Contre l’Histoire” [War against history]

Sanday (1986) Divine Hunger: Cannibalism as a Cultural System

Schöppl von Sonnwalden (1992), Kannibalismus bei den noramerikanischen Indianern und Eskimo [Cannibalism among the North American Indians and Eskimos]

Spennemann (1987) “Cannibalism in Fiji: The Analysis of Butchering Marks on Human Bones and the Historical Record with an Appendix on Experimental Butchering with Bamboo Blades”

Whitehead (1984) “Carib Cannibalism: The Historical Evidence”

Reading these studies convinced me, first, that cannibalism not only existed as a practice (not just as a metaphor or symbolic ritual), but that it once occurred in many cultures around the world—mainly though not exclusively in simple cultures, with no more than two levels of political hierarchy (the chief of a given tribe and the chief of a group of tribes). In addition, these studies show that both mortuary cannibalism, showing respect and care for deceased members of one’s own band, and cannibalism consumption of parts of fallen enemies, intended to show disrespect toward or to kill the spirit as well as the body, occurred in diverse, widely separated cultures, with little or no opportunity for “diffusion” of a practice from one culture to the next.

The length of this rejection of Arens’ claim that cannibalism was not widely practiced is largely a function of the disturbance which his unwarranted accusations have created in the anthropological literature. In addition to some of the early book reviews, many of the later sources which confirm the existence of cannibalism equivocate about its extent. Here, too, Sanday’s work is illustrative. While writing a lengthy and important monograph on the diverse meanings of cannibal practices in a dozen different cultures, and while claiming that some cases unquestionably involved the physical ingestion of human flesh, Sanday distances herself from any assertion that cannibal acts were common by treating mythical, symbolic cannibal behavior and literal cannibal behavior as identical for the purposes of her study. Thus, in reviewing the literature on a representative sampling of cultures to assess the incidence of cannibalism, she treated cannibalism as present not only in cultures where ritual or starvation cannibalism was practiced, but also in cultures for which there were “reports of past practice, legend, or hearsay,” and cultures for which “fantasized incidents of cannibalism are feared and take the form of belief in cannibal sorcerers or witches.” This is not to say than Sanday herself was unable to distinguish in the sources between purported real and fantasized cases of cannibalism; but only that for analyzing the meaning of cannibalism, she did not treat the distinction as important.

This is the case in several of the post-Arens studies listed above, in which the authors observe that literal cannibalism undoubtedly occurred in a given culture at some point in the past, but that rituals, myths, and stories involving cannibal behavior are equally good, if not better, for the purpose of exploring and understanding the meaning of the cannibal practice in a given culture.

In addition, there are a few anthropologists who have continued to cite Arens as a credible source; and there have been have been several review studies prompted by his work, which reconsider the primary sources and find many lacking in the degree of detail and credibility one would want. One of the most controversial cases concerns the Fore of New Guinea, among whom women (and some children) transmitted an invariably fatal infectious disease with a 5–25 year incubation period (kuru) either by eating or by handling the decomposing brain of deceased relatives. In an article published in the American Anthropologist in 1982, Steadman and Merbs, anthropologists at Arizona State University who cite Arens and build on his methods, argue that the case for eating was circumstantial at best, and the evidence contained many contradictory and unsubstantiated points. In 1992, however, a research doctor who contributed an article to a book on Human Biology in Papua New Guinea, without citing or showing any evidence of having read the Steadman and Merbs piece, used some of the same evidence to describe the outstanding scientific detective work through which it was found that the disease was transmitted through cannibal consumption of the brains of dead victims. In the meantime, Steadman and Merbs are continuing to be cited by anthropologists as documenting the ease with which fact and the rumor about cannibalism can be confused. The main studies of the usefulness of primary sources prompted by Arens’ work are four German books: Frank (1987), ‘Y se lo comen’: kritische Studie der Schriftquellen zum Kannibalismus der panosprachigen Indianer Ost-Perus und Brasiliens [A critical study of the written sources on cannibalism among the Pano-speaking Indians of East Peru and Brazil]; Menninger (1995) Die Macht der Augenzeugen: Neue Welt und Kannibalen-Mythos, 1492–1600 [The Power of the Eyewitness: The New World and Cannibal Myths 1492–1600]; Peter-Rocher (1994), Kannibalismus in der prähistorischen Forschung: studien zu einer paradigmatischen Deutung und ihren Grundlagen [Cannibalism in prehistoric research: Studies of a paradigmatic interpretation and its foundations]; and Wendt, Kannibalismus in Brasilien: eine Analyse europaischer Reiseberichte und Amerika-Darstellungen für die Zeit zwischen 1500 und 1654 [Cannibalism in Brazil: An analysis of European Travel Reports and Images of America for the Period between 1500 and 1654].

Arens-based doubts also continued to be raised, with damaging effects for cultural anthropology, in the largely unrelated area of the archaeological study of human bones which may show signs of food cannibalism. For example, Trinkhaus (1985), Bullock (1991, 1992), Bahn (1990, 1991, 1992), Pickering (1988), and Russell (1987a, 1987b) argue that archaeological evidence at particular sites is insufficient to warrant a claim of cannibalism, particularly since no reliable evidence of cannibalism in any age exists—a claim for which they cite Arens as the main authority.

Since there is a great deal of undisputed evidence of cannibalism in periods of famine, and this could account for at least some of the archaeological findings pointing to cannibalism, the claim by these authors that there is no proof that any form cannibalism ever occurred is unlikely to have much impact in the field. A much more seriously damaging dissemination of misinformation based on Arens’ book is the analysis of ritual and customary cannibalism presented by White (1992). Since White’s book is intended as a textbook on the archaeological study of cannibalism and since all of the reviewers (except Bahn, who refuses to recognize any form of cannibalism anywhere) agree that the book is excellent for this purpose, White’s treatment is likely to shape the thinking of a generation of archaeologists. After citing Arens and his reviewers briefly and dismissing Sanday because she includes some societies with cannibal myths in her study, White concludes that cultural anthropology cannot be of much use on this subject:

As Arens has suggested, many if not most historical sources on cannibalism are inadequate or inaccurate. “Because ethnographic research no longer seems possible, the study of cannibalism must, of necessity, be accomplished by a historical science. A man in a position to know, Matos Moctezuma, the excavator of a site at which Spanish accounts suggest that human sacrifice took place (the Aztec Templo Mayor in Mexico City) puts it this way (1987: 185): ‘Documentary sources provide us with historical information that is either exaggerated or faithful to observations, depending on the bias of the chronicler and how he has chosen to present his material. Such ethnohistorical information serves as a basis for the hypotheses that are corroborated or invalidated by excavation and archaeological evidence. Archaeology then either validates the written information or demonstrates its unreliability.’ Archaeology seems, therefore, to be the only remaining tool for investigating the existence and extent of cannibalism.”

There are two main problems with this statement: First, even though cannibal practices and, in most cases, the cultures that still had retained them in historical periods have now died out, considerably more research in the cultural anthropology of ritual cannibalism has been conducted since Arens’ book was published—including superb monographs by Combès (who relied entirely on historical sources) and by Castro (who lived with a small band intermittently over a period of years)—and a great deal more remains that can be done with ethnographies and other written sources. Second, unlike cultural anthropology, archaeology offers little if any hope of clarifying the nature, context, and meaning of the great majority of reported cannibal practices, because these practices, unlike food cannibalism or starvation cannibalism, tend not to leave marks on bones; and in most cases (involving the consumption of blood, or of a small bit of soft tissue, or of ash following cremation), they will not have left any detectable archaeological relic.

In conclusion, while ritual and customary endo- and exo-cannibal practices undoubtedly occurred in many simple societies, this fact has not yet been fully rehabilitated in anthropology and archaeology in the wake of Arens’ critique.