Reading Journal

guest123

Sanday.pdf

Home >Reading homework help >Reading Journal

... .....

ij)t' ~c) 'J 'SQclAd it \,J 1 l)-i'li(VIe +-LUA leI e , iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiili~~=======----

NOTICE THIS MATERIAL MAY Bi

?ROTECTED BY COPYRIGHT LAW (mLE 17 U.S. CODE)

1 . Cannibalism cross-culturally

In recent years ... cultural anthropologists have ... begun to give the topic [cannibalism] serious analytic attention. This de velopment stems partly from the discovery of new facts and partly from the realization that cannibalism - like incest, aggres sion, the nuclear family, and other phenomena of universal hu man import - is a promising ground on which to exercise certain tpeoretical programs. 1

Anthropological debate on the subject of cannibalism has revolved around three theoretical programs, each of which provides a dis tinctly different lens for viewing the details of cannibalism. Psy chogenic hypotheses explain cannibalism in terms of the satisfac tion of certain psychosexual' needs. The materialist hypothesis presents a utilitarian, adaptive model- people adapt to hunger or protein deficiency by eating one another. The third approach fol lows a hermeneutical path rather than a hypothetico-deductive model in conceptualizing cannibal practice as part of the broader cultural logic of life, death, and reproduction.

In this chapter I show that cannibalism is not a unitary phenom enon but varies with respect to both cultural meaning and cultural content. Cannibalism is never just about eating but is primarily a medium for nongustatory messages - messages having to do with the maintenance, regeneration, and, in some cases, the foundation of the cultural order. In statistical terms, cannibalism can be tied to hunger, but hunger is not necessarily tied to cannibalism (see discussion of Table 5 in this chapter). The job of analysis, I sug gest, requires a synthetic approach, one that examines how mate rial and psychogenic forces are encompassed by cultural systems. We must look, as Geertz says, at how generic potentialities (and, I

Introduction

would add, concerns stemming from material realities) are focused

in specific performances. 2

The complexity of cannibal practice cross-culturally

The discussion that follows is based on an examination of the sample of 156 societies I employed in an earlier study of female power and male dominance. This group otTers scholars a repre sentative sample of the world's known and best-described soci eties. The time period for the sample societies ranges from 1750 B. C. (Babylonians) to the late 1960s. These societies are distrib uted relatively evenly among the six major regions of the world as defined by cross-cultural anthropologists. Additionally, the soci eties represented vary in level of political complexity and type. of subsistence technology. 3 ~~

Of the 156 societies examined, 109 yielded information that I deemed sufficient ~nough to judge whether cannibalism could be classified as present or absent. One-third (34 percent) of this sample yielded information indicating the presence of cannibal ism. Descriptions of cannibalism come from several types of sources: interviews with people who have observed cannibalistic practices in their own society; eyewitness accounts left by mission aries; tribal traditions; and accounts of travelers. Reports of can nibalism are unevenly distributed in various cultural areas of the world. Most come from North America and the PacifIc Islands, with reports from Africa and South America being next in the order of frequency. Only two cases have been reported in the Circum-Mediterranean area and no cases have been reported for the whole of East Eurasia (see Table 1).

The descriptions of cannibalism can be classified according to three general categories: (1) ritual cannibalism is practiced, that is, human flesh is regularly consumed in ritual settings; (2) ritual can nibalism is not reported but institutionalized cannibalism is men tioned in other contexts (i.e., reports of famine, reports of past practice, legend, or hearsay); (3) ritual cannibalism is not reported, but fantasized incidents of cannibalism are feared and take the form of belief in cannibal sorcerers or witches.

A variety of themes appear in reports of cannibalism. The role of hunger is frequently mentioned, and most people believe that cannibalism may occur during times of extreme hunger and fa

Cannibalism cross-culturally

Table 1. Geographical distribution of reports ofcannibalism

Cannibalism

Present Absent Row totals

Geographical area No. % No. % No. %

Sub-Saharan Africa CirClIl1l

Mcditcrrallcall East Eurasia Insular Pacific North Al1lcriCl South and Ccntral

America Column totals

2 a

11 11

6 37 (34%)

IS a

52 4R

11 23 10 12

8 72 (66%)

85 100

48 52

13 23 21 23

14 109 (100%)

100

100 100 100 l()()

100

mine. However, hunger cannibalism is generally treated as revolt ing and reprehensible, the ultimate antisocial act, in some cases punishable by death. Tuzin provides an excellent descri ption of this attitude in his discussion of the Arapesh response to Japanese hunger cannibalism as the ultimate unthinkable act, one that im plied a deranged, anguished- abandonment of humanity.4 Tuzin also mentions, however, that other groups in New Guinea treated hunger cannibalism as commonplace. 5

The food value of human flesh is referred to in many reports from the Pacific. It is not clear, however, whether such reports are the authors' fantasy or actual fact. Quoting from a nineteenth century account, Sahlins notes that Fijian chiefs of the last century did not regard the human victim "in the shape of food," since cannibalism was "a custom intimately connected with the whole fabric of their society." Nevertheless these chiefs told the Europe ans "that they indulged in eating (human flesh) because their coun try furnished nothing but pork, being destitute of beef and all other kinds of meat."6 Reports from the Pacific commonly equate human with animal flesh. The Orokaiva gave as their reason for consuming human flesh their "desire for good food." All victims acquired in an intertribal raid were consumed. Human corpses were handled as if they were animals slain in the hunt. Corpses of grown men were tied by their hands and feet to a pole and carried face downward. Slain children were slung over the warrior's shoul

Introduction Cannibalism cross-culturally der in the manner of a hunter carrying a dead wallaby, with each hand of the body tied to each foot. 7 Lindenbaum reports that the Fore equated pigs and humans and applied the Melanesian pigdin term for meat and small game to the human flesh consumed by women. 8 Despite the reputed equation of human flesh with meat in some cases, the actual consumption in these cases has cultural connotations beyond gustatory considerations. For example, among the Orokaiva the primary reason for acquiring cannibal victims in intertribal raids was to compensate for the spirit of an Orokaiva man killed in such a raid. Fore concepts revolved around the notion that human meat, like pig flesh, helps some humans regenerate.

In many reports, the events associated with cannibalism refer not to hunger but to the physical control of chaos. For example, the victim is cast ~s the living metaphor for animality, chaos, and the powers of darkness - all those~ things people feel must be tamed, destroyed, or assimilated in the interest of an orderly social life. Cannibalism is then associated with a destructive power that must be propitiated or destroyed, and the act of propitiation or destruction is directly tied to social survival. The power is vari ously located. It may be within animals or enemies, or may be harbored as a basic instinct in humans. When projected onto ene mies, cannibalism and torture become the means by which pow erful threats to social life are dissipated. To revenge the loss of one's own, the victim taken in warfare is tortured and reduced to food in the ultimate act of domination. At the same time, by consuming enemy flesh one assimilates the animus of another group's hostile power into one's own.

Other reports tie cannibalism to a basic human instinct that must be controlled for the sake of internal social survival. In these cases cannibalism provides an idiom for deranged and antisocial behavior. For example, in their most secret and supefl1Jturally powerful ritual society, the Bella Coola performed a Cannibal Dance in which they enacted their view of human nature. The Bella Coola believed that during the per.formance of this ritual the cannibal dancer became possessed by an animal force that caused the dancer to want to bite people and filled him or her with an insatiable desire for human flesh. 9 This force was controlled in the dancer with ropes, bathing, and a droning kind of singing. 10 The close connection between the cannibal dancer and the Bella Coola

gods adds a supernatural dimension to the Bella Coola perception of the cannibal instinct of humans. In staging the cannibal ritual, the Bella Coola found ~ way to channel powerful forces into so ciety and to order those forces for social purposes.

Human sacrifice with its associated cannibalism was the means by which the Aztec gained access to the animating forces of the universe. For the Aztec "the flowing of blood [was] eq uivalent to the motion of the world." "Human sacrifice," Sahlins says, "was ... a cosmological necessity in the Aztec scheme, a condition of the continuation of the world." II The Aztec feared that when the gods became hungry their destructive powers would be unleashed against humanity. To keep the mystical forces of the universe in balance and to uphold social equilibrium, the Aztec fed their gods human flesh. By the act of consecration the sacrificial victims were incarnated as gods. Through eating the victim's flesh, men entered into communion with their gods, and divine power was imparted to men.

Exocannibalism (the cannibalism of enemies, slaves, or victims captured in warfare), characterizes the majority of cases. In the few instances of endocannibalism (the cannibalism of relatives) human flesh is a physical channel for communicating social value and pro creative fertility from one generation to the next among a group of humans tied to one ano'ther by virtue of sharing certain sub stances with common ancestors. Endocannibalism recycles and re generates social forces that are believed to be physically constituted in bodily substances or bones at the same time that it binds the living to the dead in perpetuity.

These sketchy descriptions illustrate the diversity in the cultural content of cannibal practice. More recent ethnographic descrip tions of cannibalism reach the same conclusion. Even within the same society, cannibalism may be diversely constituted, as Poole's description of Bimin-Kuskusmin cannibalism illustrates. For Bimin-Kuskusmin,

the idea of cannibalism implicates a complex amalgam of practice and belief, history and myth, and matter-of-fact assertion or elaborate meta phor. The subject enters into crass sexual insults, ribald jokes, and re vered sacred oratory. It is displayed in the plight of famine, the anguish of mourning, and the desperation of insanity. It marks aspects of the social life-cycle from the impulses of the unborn to the ravages of the ancestors. It is projected outward as a feature of the ethnic landscape and

6 7

Cannibalism cross-culturallyIntroduction

inward as an idiom of dreams, possession states, and other personal fan tasy formations. In different contexts it may be seen as an inhuman, ghoulish nightmare or as a sacred, moral duty. But always it is encom passed by the order of ritual and the tenor of ambivalence. The Bimin Kuskusmin have no single term for "cannibalism," for the ideas that are implicated are constructed for particular purposes of discourse that em phasize different dimensions of the phenomenon.

The complexity of cannibalism as a cultural practice means that to reduce it to a dichotomous variable robs it of all cultural con tent.13 Nevertheless I proceed with this exercise as a means for determining whether the kinds of exogenous forces posited by material and psychogenic hypotheses are statistically associated with the practice of consuming human flesh. In doing so I do not intend to suggest £hat culture must conform to material ~con straints, but rather, as Sahlins states, ::that it does so according to a definite symbolic scheme which is never the only one possible." 14 Thus, if hunger is a material force to be reckoned with in societies practicing cannibalism, as Table 5 suggests, I argue that we must look at the effects of hunger and ask how these effects are culturally constituted. The fact that hunger is just as likely to be present in societies that do not practice cannibalism demonstrates Sahlins's point that more than one symbolic order may constitute the effects of a given material force. Thus, hunger is encompassed by a cul tural order that includes cannibal practice in some cases and by some other symbolic scheme, which mayor may not include a physical referent to eating, in others.

The information presented in Tables 1-5 is based solely on re ports of cannibalism falling in the category of institutionalized can nibalism. Reports of cannibalism as fantasy, as a past event, or as a periodic occurrence during times of famine are not included. The reason for limiting the cases to the purported regular consumption of human flesh derives from the stipulations on the data posed by the materialist hypothesis. Since the main causal variable posited by the materialist explanation is the ongo-ing satisfaction of hunger or protein deficiency, obviously the data must reflect actual as op posed to fantasized or infrequent consumption of human flesh. (In subsequent chapters, this restriction on the data will not apply and the discussion will include the fear of cannibalism, whether or not cannibalism is thought to be actually practiced. Additionally, in these chapters I will not be concerned with whether the consump

tion of human flesh actually takes place, because my focus will be on interpreting the rituals in which human flesh is purportedly consumed.) .

The requirement that the data reflect actual instances of canni balism brings to mind Arens's charge that since "no one has ever observed this purported cultural universal," we must be skeptical about its actual existence. 15 A search of the literature convinces me that Arens overstates his case. Although he 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 ob served cannibalism, reliable eyewitness reports do exist. In re sponse to Arens, Sahlins excerpts some of the nineteenth-century eyewitness reports from the journals of Pacific travelers. 16 Addi tionally, eyewitness reports presented in The Jesuit Relations con tradict Arens's assertion that "[ t]he collected documents of the Jesuit missionaries, often referred to as the source for Iroquois cru elty and cannibalism, do not contain an eyewitness description of the latter deed." 17

One of the most compelling eyewitness reports I have encoun tered was penned in 1879 by a native of the Cook Islands who was among the first Polynesian missionaries. Upon learning to write from European missionaries, he kept a log of his travels and wrote many letters, some of which described the consumption of human flesh. One particularly lurid but descriptive example comes from a report of a war that broke out in New Caledonia soon after his arrival there as a missionary.

I followed and watched the battle and saw women taking part in it. They did so in order to carry off the dead. When people were killed, the men tossed the bodies back and the women fetched and carried them. They chopped the bodies up and divided them.... When the battle was over, they all returned home together, the women in front and the men behind. The womenfolk carried the flesh on their backs; the coconut-leaf baskets were full up and the blood oozed over their backs and trickled down their legs. It was a horrible sight to behold. When they reached their homes the earth ovens were lit at each house and they ate the slain. Great was their delight, for they were eating well that day. This was the nature of the food. The fat was yellow and the flesh was dark. It was difficult to separate the flesh from the fat. It was rather like the flesh of sheep.

I looked particularly at our household's share; the flesh was dark like

r .

Introduction

sea-cucumber, the fat was yellow like beef fat, and it smelt like cooked birds, like pigeon or chicken. The share of the chief was the right hand and the right foot. Part of the chief's portion was brought for me, as for the priest, but I returned it. The people were unable to eat it all; the legs and the arms only were consumed, the body itself was left. That was the way of cannibalism in New Caledonia. 18

More recent eyewitness evidence is reported by Poole, who wit nessed acts of 13imin-Kuskusmin mortuary cannibalism ;U1<.1 by Tuzin, who describes eyewitness evidence given him by Arapesh

informants. 19 The f:lct th:lt Arens overst:ltes his C:lse should not be taken to

mean that the thirty-seven cases of cannibalism reported in Table 1 represent undisputed examples of actual cannibalism. The eth nographies upon which I relied are the best available for use in cross-cultural research based on a standard sample. The data on cannibalism, however, are uneven, ranging from lengthy descrip tions of ritual cannibalism reconstructed from informants' recol lections of the past to a few sentences describing the consumption of the hearts of enemies. Keeping in mind the problematic n:lture of the data, the reader is cautioned to look for suggestive trends in the tables rather than irrefutable demonstrations of relationships.

Saga 11 's psyclzogellic hypotheses

I begin by considering the hypotheses in Sagan's study of canni balism that can be examined within a cross-cultural framework. These are not the only dimensions to Sagan's argument. For ex ample, he builds a good case for the role of emotional ambivalence in cannibal practice, an argument I shall return to in Chapter 2, where I suggest that, although Sagan's contribution is important and useful, it is limited by his particular reading of Freud.

Sagan contends that cannibalism "is the elementary form of in stitutionalized aggression."20 Employing the Freudian frustration aggression hypothesis and the idea that "bral incorporation is the elementary psychological response to anger and frustration, Sagan hypothesizes that cannibalism is characteristic of a primitive stage of social development. "The undeveloped imagination of the can nibal," he says, will deal with frustration through oral aggression, because the cannibal "is compelled to take the urge for oral incor-

Cannibalism cross-culturally

poration literally. He eats the person who, by dying, has aban doned him."21 Or, he eats the enemy whose very existence may deny him strength in order to incorporate that strength into his own body. When it occurs in more advanced social systems, Sagan suggests that cannibalism is a regressive response to social disin tegration, for in these cases, he says, "it is inevitable that the sat isfaction of aggressive needs sinks to a more primitive level." This happcned in Nazi Germany, "a society in a statc of psychotic breakdown." The civilizing forces broke down under the strain Germany experienced before the Nazis took power. Although not true cannibalism, Sagan says, the destruction of millions of people, the lamp shades of human skin, and similar practices concentrated on the body, exemplify the reversion to primitive aggression. 22

Citing the work of the Whitings, Sagan hypothesizes that ex tended nursing, a long period of sleeping with the mother, and father absence yield children who are overly dependent on their mothers and hence more prone to frustration and oral aggression. The adult male who carries this unconscious dependence upon infantile and childhood supports and who is also expected to be masculine and brave will need to display his masculinity and his independence of feminine support: "He will eat people, he will kill people, he will make war, he wjll enslave others, and he will dom inate and degrade women."23

Sagan's discussion suggests that as the elementary form of insti tutionalized aggression, cannibalism will occur among the simpler societies, in advanced societies faced with a disintegrating social identity, and in societies in which infant dependence upon the mother is prolonged. We can frame these suggestions in terms of several variables and correlate them with reports of the presence or absence of cannibalism, admitting, however, that this exercise does not do justice to Sagan's more complex ideas.

The first variable measures the level of political complexity. Twenty-five of the thirty-seven societies with reported cannibal ism are politically homogeneous, meaning that the highest level of jural authority is the local community. Thus, cannibalism is more likely to be present in politically homogeneous than heteroge neous societies (see Table 2). However, this information does not support Sagan's hypothesis that cannibalism is a primitive form of aggression because of the fact that more than half (56 percent) of

1110

Introduction

Table 2. Relationship between level ofpolitical sovereignty and cannibalism

Cannibalism

Present Absent Row totals Levels of political

sovereignty No. % No. % No. %

Nothing above local community 25 44 32 56 57 100

One jural level above community 4 23 13 77 17 100

Two jural levels above community 4 44 5 56 9 100

Three or more jural levels above community

Column totals ~ 4 37 (34%)

15 22 72 {66%)

85 26 109 (100%)

• 100

the simpler societies do not practice cannibalism. The most that can be said from the information presented in Table 2 is that can nibalism is more likely to be found in the simpler societies.

From Sagan's discussion of maternal dependency and oral aggression, it is reasonable to assume that cannibalism is associated with such factors as a lengthy postpartum taboo against sexual intercourse and male aggression, including aggression against women. However, these variables are not associated with the cross-cultural incidence of cannibalism in simple societies. There is no statistically significant relationship between the length of the postpartum sex taboo, the variable usually employed as an indi cator of maternal dependency, and the occurrence of cannibalism in politically homogeneous societies. Neither is there any relation ship between the number of indicators of male aggression and the incidence of cannibalism in these societies (see Tables 3 and 4).

However, in politically heterogeneous societies (with at least one jural level above the local community), a significant association between the length of the postpartum sex- taboo and cannibalism emerges. In Sagan's terms, this means that maternal dependency is related to oral aggression (as measured by the presence of can nibalism) in more complex societies. It is also true that in more complex societies there is a significant relationship between male aggression against women and cannibalism (see Tables 3 and 4).

Cannibalism cross-culturally

Table 3. Relationship between length ofpostpartum sex taboo and cannibalism in politically homogeneous and heterogeneous societies

Cannibalism

Length of postpartum sex taboo

Present

No. %

Absent

No. % Row total

Politically homogeneous societies Up to 6 Months From 6 Months to

12 63 16 67 28

more than 2 years Column totals

7 19

37 100

8 24

33 100

15 43

Politically heterogeneous societies Up to 6 months From 6 months to

3 30 22 73 25

more than 2 years Column totals

7 10

70 100

8 30

27 100

15 40

Note: For politically homogeneous societies phi = .04, not significant. For po litically heterogeneous societies phi = .39, P = .007. No information for twenty-six societies.

Elsewhere I have shown that male aggression against women is significantly associated with food stress. I argue that male aggres sion is a reaction to stress as' males seek to dominate controlling material forces by dominating the bodies of women and female reproductive functions. However, I qualify this conclusion by showing that male aggression against women is more likely to be a solution to stress in societies displaying a symbolic orientation to the male creative principle. Thus, adaptation to stress does not always include the subjugation of women and I argue for the ne cessity of examining cultural factors that may shape a people's re action to stress. 24 The same comments apply to the results dis played in Tables 3 and 4. Although male aggression and maternal dependency are related w the presence of cannibalism in politically heterogeneous societies, it is clear from these tables that both of these variables may occur in the absence of cannibalism, suggest ing that we must look beyond the behaviors measured by these variables in order to comprehend the incidence of cannibalism.

A similar argument is called for when examining Table 5, which

Introduction

Relationship between male aggression and cannibalism inTable 4. politically homogeneous and heterogeneous societies

Cannibalism

AbsentPresent Row Male aggression % total % No.No.scale'

Politically homogeneous societies I I0-3 indicators of 42 187 32 11male aggression

4 or 5 indicators of 58 3015 68 15male aggression 100 4822 100 26Column totals Politically heterogeneous s~cieties

. 0-3 indicators of .' 68 172 25 15male aggression 4 or 5 indicators of

7 32 136 75male aggression 3022 1008 100Column totals

Note: For politically homogeneous societies phi = .11, not significant. For polit ically heterogeneous societies phi = .39, P = .02. No information for thirty- one societies . •A Guttman scale formed by five indicators: (1) men's houses, (2) machismo, (3) interpersonal violence, (4) rape, (5) raiding other groups for wives. See Sanday

(1981, Appendix F) for details.

indicates a signifIcant relationship between cannibalism and food stress. Most (29, or 91 percent) of the societies for which there are reports of cannibalism experience occasional hunger or famine or protein defIciency. Although hunger is intimately associated with the practice of cannibalism, we cannot conclude that hunger con stitutes cannibal practice. As Table 5 demonstrates, many societies (43, or 60 percent) that experience food stress show no evidence of cannibalism; thus, here again, we must look to culture to under stand the constitution of cannibal practice.

The data are inconclusive with respect to Sagan's psychogenic hypotheses. Sagan's claims are reductionist and, like the materialist approach, ignore the symbols mediating the experience of oral frustration and the act of oral aggression in cannibalism. Sagan's stress on cannibalism and male aggression as a reaction to oral frustration (as measured by maternal dependency and food stress)

Cannibalism cross-culturally

Table 5. Relationship between food stress and cannibalism in politically homogeneous and heterogeneous societies

Cannibalism

Present Absent Row

Food stress No. % No. % total

Politically homogeneous societies Food is constant 2 9 9 31 11 Occasional hunger or

famine or protein deficiency

Column totals 19 21

91 100

20 29

69 100

39 50

Politically heterogeneous societies Food is constant 1 9 16 41 17 Occasional hunger or

famine or protein defICiency

Column totals 10 11

91 100

23 39

59 100

33 50

Note: For politically homogeneous societies phi = .26, p = .03. For politically heterogeneous societies phi = .28, p = .02. No information for nine societies.

is relevant, as the results shown in Tables 3-5 illustrate. However, I argue that we must examine the underlying ontological struc tures that render maternal dependency, food stress, and associated acts of male aggression relevant to the practice of cannibalism in some cases and not in others since, as Tables 3-5 also indicate, these factors are just as likely to be present in the absence of can nibalism. In Chapter 2, I present the analytic framework that incorporates these considerations. In the remaining part of this chapter I examine the materialist hypothesis, Sahlins's culturalist response, and several other approaches that are useful for compre hending the social and cultural context of cannibal practice.

The materialist approach of Michael Harner and Marvin Harris

The materialist hypotheses proposed by Harris and Harner to ex plain the scale of Aztec human sacrifice focus on hunger and pro tein deficiency. Harner claims that ecological and demographic

I Introduction Cannibalism cross-culturallyfacts explain the scale of Aztec human sacrifice. In the Aztec case Harner sees it is not known what amount of fatty acids is required by the an extreme development, under conditions of environmental circum scription, very high population pressure, and an emphasis on maize ag riculture, of a cultural pattern that grew out of a Circum-Caribbean and Mesoamerican ecological area characterized by substantial wildgamc degradation and the lack of a domesticated herbivore.... IntensifIcation of horticultural practices was possible and occurred widely; but for the necessary satisfaction of essentIal protein requirements, cannibalism was the only possible solution.... From the perspective of cultural ecology and pOpUI:Hioll pressure theory, it is possible to llndcrst:ll1d :ll1d respect the Aztec emphasis on human sacrifice as the natural alld ration:J! re sponse to the material conditions of their existence. 25

Citing an unpublished estimate by a leading authority on. the demography of Central Mexico arourl.d the time of the Conquest, Harner says that 1 percent of the total population, or 250,000, were sacrificed per year in Central Mexico in the fifteenth century. As to what was done with the bodies, Harner relies on accounts writ ten by conquistadores such as Bernal Diaz and Cortes and on the post-Conquest description penned by Sahagun. :2(,

Some reports refer to eJting humJn flesh in a nonsacritlcial con text. Cortes writes that one of his 111en leading a punitive e\:pedi tion came across "loads of maize and roasted children which they [Aztec soldiers 1 had brought as provisions and which they left be hind them when they discovered the Spaniards coming." c7 Simi larly, Sahagun mentions that Aztec merchants discovered traveling in enemy territory were killed and served "up with chili sauce." According to Duran, the flesh of the cIptive caten after sacrifice was not part of the rite itself but "was considered [to be] 'leftovers' and was returned to the captor as a reward for having fed the dei ty." 28

Such rewards were important because captors were recruited from the ranks of commoners who rarely ate meat or poultry. They got their protein from a "floating lOubstance" on the surface of lakes, from amaranth, and from the regular diet of maize and beans. Famines were common and every year people faced the threat of shortage. A prolonged famine in 1450, for example, forced the rulers of the Three-City League to distribute the surplus grain that had been stored for ten years. 29

The scarcity of fats caused another dietary problem. Although

human body, fats <Ire thought to provide a longer-lasting energy SOurce and aSSure the utilization of the essential amino acids for tissue building. In this connection, Harner notes that the Aztecs kept prisoners in wooden cages prior to their sacrifice and ma'y sometimes have fattened them there. 30

In contrast to the commoners, the nobility and the merchant class fed on a rich diet of protein in the form of wild game. Human flesh, too, was reserved for "illustrious and noble people." Thus, during good times human flesh may not have been nutritionally essential for the nobility. Harner suggests, however, that the Con sumption of human flesh probably fluctuated and made its greatest contribution to the diet when protein resources were at their low est ebb. The privilege of eating human flesh provided good insur ance against hunger during times of famine, when the nobility as well as the commoners could suffer significantly. 31

Commoners could partake of human flesh and wild game by taking captives single-handedly in battle. Upon capturing a total of three war prisoners, commoners received the gustatory privi leges of the nobility and were raised to the position of "master of the youths." They also became eligible to host a cannibal feast for their blood relatives and dine at Moctezuma's palace on imported wild game. These were the rewards in an economy of scarce meat. By rewarding successful warriors in this manner, the Aztec rulers 11l0tiv:1ted the poor to participate in offensive military operations. They pumped up an aggressive war machine with the promise of meat. "[U]nderlying the competitive success of that machine," Harner says, "were the ecological extremities of the Valley of Mexico. ".12

Marvin Harris describes preconquest political necessities in the Valley of Mexico along with several other examples to demon strate a more general relationship in human society "between ma terial and spiritual well-being and the cost/benefits ... for increas ing production and controlling population growth."3J In the case of the Aztecs, their material well-being was threatened by Occa sional periods of famine caused by depletion of the Mesoamer ican ecosystem after centuries of intensification and population growth. Their spiritual well-being depended on sacrifice and can nibalism. The severe depletion of animal protein reSOurces in the Valley of Mexico, he claims,

I Introduction Cannibalism cross-culturallymade it uniquely difficult for the Aztec ruling class to prohibit the con sumption of human flesh and to refrain from using it as a reward for transmission was the notion of regeneration and reproduction. loyalty and bravery on the battlefield. It was of greater advantage to the ruling class to sacrifice, redistribute, and eat prisoners of war than to use them as serfs or slaves. Cannibalism therefore remained for the Aztecs an irresistible sacrament, and their state-sponsored ecclesiastical system tipped over to favor an increase rather than a decrease in the ritual butch <:ring o( c.lptivcs :1I1d the redistrihution of hUlllan tlcsh.'4

Sahlins's culturalist rejoinder to Hams and Harner

Sahlins sees the "Western business mentality" at the heart of Har ris's view of Aztec cannibalism. In Harris's utilitarian view, every thing in the social superstructure is governed by its economic func tion so that the me;nings other peop)e give to their lives are nothing more than the material rationalizations we give to our own. "Once we characterize meaningful human practices in these ideological terms," Sahlins says, "we shall have to give up all an thropology, because in the translation everything cultural has been allowed to escape."35

The cultural content Harris ignores is the stupendous system of Aztec sacrifice. Sahlins approaches this content head on: He does not attempt to dodge its complexities. Staying close to his subject matter, he illuminates the logic of sacrifice and shows how canni balism fits within this logic. Aztec cannibalism can only be under stood within the broader system of Aztec sacrifice for by itsel f cannibalism did not exist for the Aztec. It is true that hum:m Acsh was consumed, but neither was it ordinary human flesh nor was it eaten in an ordinary meal. Cannibalism as a cultural category among the Aztec was invented by anthropologists. For the Aztec, the consumption of human flesh was part of a sacrament bringing humans into communion with the gods. The Aztec focused not on the consumption of flesh but on the sacred character of the event. 36 _

Sahlins points out that the logic of Aztec sacrifice is not unique. It is found in many other societies and conforms to Hubert's and Mauss's classic explanation of the nature and function of human sacrifice. Aztec sacrifice brought the sacrificer, "sacrifier," and the victim into union with the divine. The consumption of the con secrated victim transmitted divine power to man. Underlying this

The gods were renewed through the offering, and the sacrifier (the one who has provided the victim but not necessarily the one who sacrifices it) gained divine power by giving up his claim to the victim. The entire process began with mutual adoption between Aztec victim and sacrifier. When the warrior took a prisoner, he declared: "He is as my beloved son." The captive replied: "He is as my beloved f.1ther."J7 Thus, the victim offered up by the Aztec sacrifier was his own child.

The reproductive imagery is manifest in the parallelism drawn between the mother and the warrior. The warrior's job was to nourish the Sun with the blood of adopted captives borne by the warrior to the sacrificial altar. The mother in childbirth was lik ened to the warrior engaged in battle. If she died, she shared the warriors' fate and went to the House of the Sun. When the mother bore a child, the midwife shouted war cries, "which meant that the woman ... had taken a captive."38 Thus, male and female alike Contributed to the physical reproduction of the Aztec universe.

Giving their children to the gods was a cosmological necessity: It was a condition for the continuation of the world. Without proper nourishment the gods could not work on behalf of hu mans. The gods depended on sacrifice for energy. Without it the Sun would not come up, the sky would fall down, and the universe would return to its original state of chaos. The gods depended on humans and humans depended On the gods. The steepness of the Aztec pyramid steps paralleled the course of the sun from dark to light and back to dark. As the victim climbed the steps, he or she was the Sun climbing to its midday zenith. Rolled down the west ern steps of the temple, the victim, like the sun, was going to his or her grave. The Sustenance given to the gods in the offering and to humans in their houses ensured the regeneration of everyone. 39

Sacrifice was also a sociocultural necessity. It was so implicated in the particulars of social relations, politics, and economics, that without sacrifice, the web of human social interactions would come apart. Fundamentally, "Aztec culture was reproduced by hu man sacrifice." Just as the main relations of the Aztec universe were renewed by the blood of captives, so were the relations on the social plane, for in the sacrificial act the logic of both was repre sented. Men were like the gods whose original self-destruction set the sun in motion. According to the principle ofsacrifice, the Aow

Introduction

ing of blood was equivalent to the motion of the world. Without it all would come to an end. 40

Enemies could not be subjugated or exterminated because they supplied the lifeblood of the state. Sahlins agrees that the structure of the empire was conditioned by the system of human sacrifice. But his explanation goes beyond material considerations or cost benefIt analysis. He notes that the high Aztec god Tezcatlipoca has as another name, "Enemy." The figure of this god embodied the power of the enemy. Supernatural power was often conceived as being external to society: "What is beyond society, escaping its order, is precisely what is greater than it." The ritual value of ene mies lay in the greater spiritual power they brought to society. To have annexed and subjugated enemy territory would have meant destroying the lifebl~od of the state. The greater supernatural power of the en em y helps to explain th~-initial ease of the conquest and why the subsequent hostilities were so bloodthirsty. The Spanish were conceived as different, more powerful enemies, and hence more powerful gods. The Spanish were unaware of their own worth as victims. ~1

The physical production and reproduction of cosmological and social categories

Sahlins's analysis of Aztec cannibalism is at once a critique of the idea that human cultures are formulated out of practical activity and utilitarian interests and an example of another approach to the study of culture. Harner and Harris believe that culture is precip itated from the rational activity of individuals pursuing their own best interest. The assumption underlying such utilitarianism is that humans seek to maximize benefits relative to costs. Sahlins's rea soning instead focuses on the symbolic ;1I1d the l11eJningful. The distinctive quality of man is "not that he must live in a material world ... but that he does so according to a meaningful scheme of his own devising." The decisive quality of culture is not that it "must conform to material constraints" but that it constitutes these constraints in a meaningful symbolic order:

[NJature is to culture as the constituted is to the constituting. Culture is not merely nature expressed in another form. Rather the reverse: the action of nature unfolds in the terms of culture; that is, in a form no longer its own but embodied as meaning. Nor is this a mere translation.

Cannibalism cross.eulturally IThe natural fact assumes a new mode of existence as a symbolized fact, I

its cultural deployment and consequence now governed by the relation between its meaningful dimension and other such meanings, rather than the relation between its natural dimension and other such facts.

A striking feature of Sahlins's analysis of Aztec cannibalism is his illumination of the role of the sacrificial complex in the social and cosmological reproduction of the Aztec universe. Men and women contributed to the physical reproduction of the cosmos in a variety of ways: They (along with children) contributed their lifeblood to nourish hungry gods; men conveyed the victim to the sacrificial stone; and women bore new victims in childbirth. The relations of the social order were sustained and regenerated through the idiom of sacrifice and cannibalism. For example, noble titles were conferred on those who contributed sacrificial victims, humans became gods through the sacrificial rites, and the states supplying victims were politically separated from those counted as allies.

More than an idiom for regenerating order and structure, the sacrificial complex was also deeply implicated in the founding of Aztec society (see Chapter 8). The dialectic between submission in sacrifice and dominance in the gruesome rites that followed rit ually marked the development of the Aztec state from its begin ning, when the migrating hunters who were the ancestors of the Aztec first settled in the Valley of Mexico. When the Aztec nobility felt defeated, as they did during the famine of1450, they admitted their submission by increasing the scale of sacrifice and asserted their dominance in arrogantly pretentious cannibal feasting. In myth and history, the Aztec social and political order was consti tuted in terms of struggle. Sacrifice and cannibalism, I suggest, were the primordial metaphors symbolizing dominance and sub rlllSSlOn.

The chartering of a social order and its reproduction are an im portant part of Sahlins's analysis of Fijian cannibalism as well (see Chapter 7). Sahlins presents a myth of the origin of cannibalism that has to do with the origin of Culture. Like Aztec cannibalism, fijian cannibalism is part of the mythical charter for sociery. In practice, Fijian cannibalism could not be separated from the or dered circulation of the principal sources of social reproduction, which established and perpetuated the developed Fijian chiefdom. The chiefdom was organized "by an elaborate cycle of exchange

Introduction

of raw women for cooked men between a basic trio of social cum cosmic categories: foreign warriors, immigrant chiefs, and indig enouS members of the land."" Wives and cooked men are both reproductive. The wives are directly "life-giving"; the cannibal victims are life-giving in that their bodies provide a tangible chan44 nel for the exchange of mana between men and gods. The system of exchange culminating in sacrifIce and cannibalism constituted "an organization of all of nature as well as all society, and of pro duction as well as polity."45 Sahlins concludes that "the historical practice of cannibalism can alternately serve as the concrete refer ent of a mythical theory or its behavioral metaphor." 4(, Thus, Fijian cannibalism, like that of the Aztec, is part of the foundation of the social order. Fijian c~nnibalism also served as a tangible symbol of dominance. The Fijian chief who offeE~d victims to his people le gitimated his chiefly dominance. In the gruesome rites that fol lowed the chiefly.offering, his male and female subjects gave vent

to more lurid displays of dominance. Although Annette Weiner docs not address the issue of canni

balism, her analysis of reproduction is relevant to this discussion because of her emphasis on the speciflc resources that "objectify the general societal process of reproduction, documenting and le gitimizing the fundamental condition whereby ego and 'others' are tied togeth r."4i By reproduction, Weiner means "the way so cieties come to

e terms with the processes whereby individuals give

social identities and things of value to others and the way in which these identities :1nd values come to be replaced by other individuals and regenerated through generations." 4l-\ The specific resources that mark relations acroSS the generations must be material objects with some physical property of durability. possibilities mentioned by Weiner arc substances or objects taken from the corpsc' itsclL4 or material objects used in formal exchange events. ') Weind's comparison of the Bimin-Kuskusmin use of bones as the concrete referent in acts of social reproduction w~th the Trobriand employ ment of bundles of banana leaves raises some interesting hypoth

eses regarding the social concomitants of cannibalism. The fundamental problematic posed by social reproduction.

Weiner says, is "(H1 0w can one draw on the resources and sub stances of others while maintaining and regenerating one's o\\'n resources and substances" without becoming "other" ?50 The Bimin-Kuskus min essentially cut off relations with the other after

Cannibalism cross-culturally

the reproductive potential of the other has been employed to beget children. For them the other always remains essentially suspect, and the substances of the other (namely, affines) are rigidly sepa rated from the substance of the lineage. 51

Poole's analysis of Bimin-Kuskusmin models of procreation, death, and personhood supports Weiner's discussion of reproduc tion and regeneration. Through acts of mortuary cannibalism. the procreative powers of the dead are recycled within the Bimin Kuskusmin lineage and clan, whereas the spirit of the newly dead, provided that it meets the test for proper ancestorhood. takes its place among the clan ancestral spirits that are responsible for nur turing the manifestation of the clan spirit in the bodies of future generations. 52 When a man or woman becomes an ancestor, Poole says, "the mortal individual is substantially dissolved in most re spects, and the wider social bonds founded on eroding substance are significantly sundered." 53 The person who becomes an ances tor leaves a legacy in the form of children, departed ancestral (called .finiik) spirit, bone, bone marrow, and procreative power. This legacy "constitutes the substantial core of the cycle of birth, death, and rebirth, and this cycle turns inward on the clan as the social category that is forever reconstituted in the Bimin-Kuskusmin ideology of societal regeneration JJ (emphasis mine);54 Thus, the clan stands alone in the symbolism of death and rebirth - it is the clan that is per petuated. The symbol of the continuity and perpetuity of the clan "is cast in the substantial symbols of bone in and on living persons, in shrines, cult houses, and ossuaries, and in ritual performance." including ritual anthropophagy. 55 This inward-turning character of Bimin-Kuskusmin acts of social reproduction can be compared - providing that the quite different level of political complexity is t:1kcn into account - with the Aztec state, which, as noted earlier, 3dopted a policy of nonexpansion. The inward-turning nature of the Aztec state was the means by which its hegemony was main t.1ined.

The Trobriand solution does not display the exclusive inward orientation of the Bimin-Kuskusmin. Labor and production of yams and women's wealth are directed within the lineage, but re lationships are not cut off with others, such as affines, fathers, and spouses. Bundles of banana leaves objectify the reproductive sig nificance of women at the same time that they give economic val Idation to relations between individuals of different lineages. Thus,

Introduction

"bundles provide for the linking of networks of relationships that last for three or more generations."56 Trobriand bones, like Bimin Kuskusmin bones, remain within the ritual contexts of ancestors. As significant objects, however, bones "never enter the economic or political domain, for bones do not validate relations external to the ancestral domain."57

This difference between turning inward as opposed to connect ing ties with affines in mortuary ceremonies is one of the social concomitants Strathern relates to the presence of cannibalism in New Guinea Highlands and Fringe Highlands societies. Focusing on marriage exchange and prominence of pig herds, Strathcrn notes that, where cannibalism is present, two factors are also pres ent: (1) "the idea of 'turning back' or of repeating marriage" is accepted just as "the idea of 'turning back' to eat one's own kind is not regarded as wrong"; (2) "herds of domestic pigs, which could be used as substitutes for the exchange and consumption of persons, are less prominent."58

Most Bimin-Kuskusmin marriages are intratribal. Marriages between tribes are usually marriages with women from enemy groups. The fear and antagonism between groups is accentuated because no attempt is made to regenerate these relationships through time, as Weiner notes in the Trobriand case. 59

Bimin-Kuskusmin cannibalism and endogamous structure can be contrasted with the marriage system of the Melpa of the West ern Highlands. The Melpa abhor cannibalism, relegating it to the secret practices of evil witches. The Melpa have elaborate rules against marrying kin, and against repeating marriages between small groups. These prohibitions occur in conjunction with an obvious stress "on proliferating exchange ties, on facing outwards to an expanding network, and on a continuous substitution of wealth items, pork and shell valuables (or nowadays cash), for the person. In this context, cannibalism stands for an unacceptable 'turning back', and is thus symbolically equated with incest."6o

Bimin-Kuskusmin pig herds are tiny by highland standards, ac cording to Poole, and there certainly is not the elaborate network of exchange documented for the Melpa. 61 Nor do pigs figure prominently in Bimin-Kuskusmin mortuary rituals, as they do among the Melpa. Melpa mortuary rites transfer the spirit of the corpse into the world of the ghosts by means of a pig sacrifIce designed to ensure the goodwill of the new ghost and the com

Cannibalism cross-culturally

munity of gr.osts. Eating the pig flesh coincides with the release of the deceased's .soul. The pigs, Strathern concludes, are substi tutes for the person's body: "[T]he pork is eaten instead of the de ceased." The funerary pig sacrifices are presented to the ghosts "in order to persuade them to accept a new ghost, and to the de ceased's maternal kin, in substitution for the flesh which will rot and return to the earth," where it fertilizes and thereby regenerates the soil of the clan territory.62 The bones of the corpse are kept by the paternal kin and placed in special houses. Thus, through the medium of pig flesh, the deceased's spirit is replaced; what the Uimin-Kuskusmin accomplish through mortuary cannibalism the Melpa accomplish through pig sacrifices.

Strathern notes that the practice of cannibalism in the New Guinea Highlands is associated with sparsely popUlated fringe re gions where large herds of domestic pigs are absent. However, he cautions against jumping to the conclusion that protein-hunger is causally related to the practice of cannibalism, because where pigs are absent, alternative Sources of protein are available in wild game, including feral pigs. Furthermore, the Hua, the Gimi, and the Fore are reported to have practiced cannibalism and alI of these groups keep herds of domestic pigs. However, in areas where ag ricultural intensification has proceeded to its greatest lengths, can nibalism is absent. 63

Conclusion

As the most recent ethnographic studies of cannibalism confirm, cannibalism is not a unitary phenomenon but varies both in mean ing and cultural content. The cross-cultural data point to at least six patterns in the practice of cannibalism:

1. Famine cannibalism is frequently mentioned. 2. Cannibalism may be fl?otivated by competition between

groups and the desire to avenge the death of someone lost 111 war.

3. Mortuary cannibalism is part of the physical regeneration of fertile substances required to reproduce future genera tions and maintain ties with the ancestors.

4. Cannibalism is a behavioral referent of a mythical charter for society and, with other social and cosmological cate

Introduction

gories, is a condition for the maintenance and reproduc tion of the social order.

5. Cannibalism is a symbol of evil in the socialization of per sons.

6. Cannibalism is part of the cultural construction of person hood.

As Poole's ethnography of Bimin-Kuskusmin cannibalism shows, several patterns may characterize the expression of cannibalism in one society; or, only one of the patterns described above may be represen ted.

The explanations of cannibalism are also diverse. The data pre sented in this chapter are inconclusive with respect to the claims of psychogenic and materialist hypotheses. However, I do not dis count the role of Psychogenic and matepalist forces and in the' following chapters I examine the interrelationship between mate rial forces and the psy.chological states predicated by rituals of can nibalism. The relationship between food stress and cannibalism leads me to suggest that, like male control of female bodies, can nibalism is part of a hegemonic strategy deVeloped in reaction to a perception of controlling natural or political forces in some cases. This strategy, however, cannot be separated from the system of symbols that predicates a people's understanding of their being in-the-world and formulates their strategies vis-a-vis social regen eration, reproduction, and dominance. More thall just a reaction to external conditions, cannibalism is a tangible symbol that is part of a system of symbols and ritual acts that predicate consciousness in the formulation of the social other and reproduce consciousncss in the ritual domination and control of the social other. Whcre domination and control are subordinate to accommodation and integration, however, cannibalism is absent regardless of the na ture of the food supply.

2 . Analytic framework

[I]nsist as we may upon the distinct character of cultural action, we are invariably forced to the conclusion that the cultural, too, is merely a part of nature. Whatever we do, we do as warm_ blooded, mammalian animals, exemplifying natural effect in all of our actions. In our own terms, then, culture is nature har nessing nature, understanding nature, and coming to know itself. I

This chapter introduces the conceptual framework that guides the analysis and presentation of data in the follOWing chapters. Fifteen case studies provide the material for the more detailed discussion. These case studies are identified in Table 6 by the name and loca tion of the society; and by whether cannibalism is social (that is, under the Control of group decisions) or is antisocial (that is, OUt side group Control). Table 6 also identifies the type of data on which my analysis is based, the type of cannibalism practiced, and the chapters that treat the case studies in more detail.

The fifteen cases were chosen because of the reliability and de tailed nature of the data. Some are drawn from the fieldwork Con ducted by ethnographers in the past decade. In one case, that of the Bimin-Kuskusmin, instances of mortuary cannibalism were observed by the ethnographer. In three Cases (the Hua, Gimi, and Goodenough Islanders), the ethnographer provides detailed infor mation On the practice of cannibalism, as reported by informants, before it Was prohibited by government and missionary officials. Three additional examples of ritual cannibalism are drawn from descriptions reconstructed by anthropologists from the aCCOunts of missionaries and travelers. Although these descriptions are de railed and vivid, the data are not specific on such topics as the theory of procreation and conception, which is essential for under