THE CANNIBALISM ISSUE : STRUCTURALIST AND MATERIALISTINTERPRETATIONS AND OTHER CONCERNS

Jeffrey P. BlickDepartment of Anthropology

University of Pittsburgh

"Unfortunately, the custom of consuming human flesh, like exoticsexual practices, polygamy, and other alien habits, raisesviolent, unintellectual passions in the Western scholars whostudy them."

- Brian Fagan, The Aztecs, p. 233

Cannibalism as a topic of interest, both general andanthropological, has seen a revival of concern in the last sevenyears or so. Primarily responsible for this revival is thecontroversial stance taken by William Arens in his book, The Man­Eating Myth: Anthropology and Anthropophagy (1979). Followingthe publication of this work has been a deluge of articles andbooks, popular and specialized, dealing with this topic in eithera pro or con fashion regarding the existence or non-existence ofcannibalism as an actual cultural practice in past (or even morerecent) history (Riviere 1980; Parry 1982; Strather 1982; Brownand Tuzin 1983; Harris 1985; Kolata 1986; Sanday 1986). A notemust be made here that Arens specifically states that he is onlyreferring to "cannibalism as a custom" (Arens 1979:21), or ritualcannibalism, as opposed to the other forms of cannibalism such assurvival cannibalism, dietary cannibalism, or the other varioustypes of cannibalism that have been proposed which will be dealtwith in greater detail in this paper.

The resultant furor due to Arens· book has caused the re­evaluation of cannibalism as a cultural phenomenon byanthropologists from varying schools of interpretation. MarvinHarris, the materialist spokesman, recently dealt withcannibalism in a chapter called "People Eating" in his 1985 book,Good to Eat: Riddles of Food and Culture. More recently, Peggy~Sanday, representative-Df the symbolist/interpretive school,has published her book on the sUbject, Divine Hunger: Cannibalismas ~ Cultural System (1986). So, it is quite evident thatcannibalism as an issue is still very much alive in the presentday; its appetite is being fed by the inevitable conflict arousedby the different research strategies involved.

At this point it is warranted that we define cannibalism andlist the various types of cannibalism that have been devised,most of which will be discussed in this paper. Cannibalism isthe act of eating human flesh or human parts by human beings orfeeding on the flesh or parts of one·s own kind. A typology ofcannibalism can be derived from the suggestions of other authorsin this field. Typologies that have been developed include:

1

CeremonialMortuary/FuneraryMagicalFertilityPietisticCustomary

dietary, ceremonial, and obligatory or emergency rationcannibalism (Nickens 1975: 291); gastronomic, ritual or magical,and survival cannibalism (Arens 1979:18); famine, competition orrevenge, mortuary, cannibalism as a "behavioral referent of amythical charter for society and, with other social andcosmological categories, is a condition for the maintenance andreproduction of the social order" (Sanday 1986:25-26),cannibalism as "part of the cultural construction of personhood"(Sanday, 1986:26); survival, ritual, and preferential cannibalism(Sagan 1974); endophagy, exophagy, ritual, and famine cannibalism(Tannahill 1975:9, 106); dietetic, famine, fertility, gluttony,magic, pietistic, punishment, revenge, shipwreck, and siegecannibalism (Hogg 1958). Other types of cannibalism mentioned inthe literature are: cannibal warfare (Harrisson 1948:97-109),survival and ritualistic cannibalism (Kolata 1986:1497), warfareand mortuary cannibalism (Harris 1985:204), and ritualcannibalism (Lindenbaum 1979:22).

From this multitudinous list, I propose .that a moredefinitive typology be derived using the above as a pool fromwhich to draw types and subtypes. This typology would be definedas:

I. Survival Cannibalism - the eating of human flesh orhuman parts by humans under adverse, stressful conditions as alast resort to survive.

A. FamineB. Obligatory or Emergency RationC. ShipwreckD. Siege/War

II. Ritual Cannibalism - the eating of human flesh or humanparts by humans in a ceremonial or religious context repetitivelyover.time.

A.B.C.D.E.F.

III. Warfare Cannibalism - the eating of human flesh orhuman parts by humans of victims captured especially forcannibalistic purposes, revenge, punishment, or out ofcompetition.

A. RevengeB. PunishmentC. Competition

IV. Dietary Cannibalism - the eating of human flesh orhuman parts by humans as a gustatory preference.

A. Preferential

2

B. GastronomicC. DieteticD. GluttonyE. Nutritional

v. Symbolic Cannibalism - the eating of human flesh orhuman parts by humans either in actuality or by substitution (ofhuman for other types of matter), which has as its purposefunctional, symbolic or psychological definitions of a group orperson.

A. Cannibalism as a "behavioral referent of a mythicalcharter ••. for the maintenance and reproduction of the socialorder" (Sanday 1986:25-26)

B. Cannibalism as a ·symbol of evil in thesocialization of persons" (Sanday 1986:26)

. C. Cannibalism as ·part of the cultural constructionof personhood" (Sanday 1986:26)

It must be noted here that, as with any other typology, thisone is by its nature arbitrary, and theoretically, overlappingmay occur among any of the types, or perhaps among all of them.

Other concepts of cannibalism are the notions ofendocannibalism, exocannibalism (endophagy and exophagy have beenproposed by Tannahill 1975), as well as Sagan's notions ofaffectionate and aggressive cannibalism (1974). Arens remarkshumorously on this note in citing Dole's (1962:567) category,"gastronomic endocannibalism" which "transforms the act of eatinghuman beings into something almost unpronounceable by the humanmouth" (Arens 1979:18).

Just as the typological considerations concerningcannibalism are difficult to untangle, so are the variousresearch strategies from which it is possible to interpretcannibalism. A few of these include the cultural materLalist,structuralist, functionalist, psychological, and symbolistapproaches. The cultural materialist tends to view cannibalismas a response to environmental stresses of protein deficiency(see Harris 1985, 1979, 1977: Harner 1977). The structuralistview is more complex. Cannibalism is viewed through themythology of various cultures to explain the similarities ofthose cultures through the likenesses of unconscious thoughtprocesses. Functionalists would consider the function or purposeof cannibalism in a society together as a vehicle of groupsolidarity, identity, and cohesion. Furthermore, functionalistswould state that cannibalism provides the ultimate in-group/out­group distinction: We eat others, not ourselves, or vice versa.The psychological approach would tend to focus on cannibalism asa form of oral sadism accompanied with feelings of ambivalence(love/hate) toward the victim. Finally, the symbolist wouldinterpret cannibalism as a metaphor for the simultaneous creation

3

and destruction of life tied up with the fertility andregeneration of the practicing culture.

The main issue involved in this paper is the character ofthe difference between the structuralist and cultural materialistschools concerning the nature and interpretation of cannibalism.The structuralist school views cannibalism as a phenomenonprimarily through the mythology of various cultures as arepresentation of the collective consciousness of human beingswithout regard to whether cannibalism actually exists in reality,rather focusing on analyzing the myths in which cannibalismoccurs. From this angle, cannibalism can be conceived of as amodel or a social form which serves some purpose in the societyin question. This concept is best elucidated by Levi-Strauss:"A structural model may be conscious or unconscious without thisdifference affecting its nature. It can only be said that whenthe structure of a certain type of phenomena does not lie at agreat depth, it is more likely that some kind of model, standingas a screen to hide it, will exist in the collectiveconsciousness. For conscious models, which are usually known as'norms' are by definition very poor ones, since they are notintended to explain the phenomena but to perpetuate them"(1963:281). Thus, cannibalism serves as a norm, or rather as adevice exerting normative controls on a group through themaintenance of the social order while perpetuating thecannibalism myths in a cyclic fashion.

One of the basic tools of the structuralist is that of thedichotomy, The tendency for mankind to think in terms of binaryoppositions has long been recognized by the structuralist andprovides a technique by which to discern the nature of howcultures divide their world, while providing insight into how acertain group thinks. Basic to the mode of structuralism is thenotion that

Uniformities in cultures arise from thestructures of the human brain and consequentsimilarities of unconscious thoughtprocesses. The most important structuralfeature of the human mind is the tendency todichotomize, or to think in terms ofbinary oppositions, and then to attempt tomediate this opposition by a third concept,which may serve as the basis for yet anotheropposition. A recurrent opposition present inmany myths, for example, is cUlture:nature(Harris 1980:522-523).

Use of the dichotomy allows an analysis of the components of mythinto meaningful categories to be subjected to emic and/or eticinterpretations. Applied to the study of cannibalism, the use ofbinary oppositions allows the anthropologist to dissect the

4

native world in terms of tame:wild, safety:danger, human: inhuman,culture:nature.

In the case of the Yanomamo, this duality is very explicit.According to Chagnon, the Yanomamo are "endocannibalisticanthropophagers"(1983:105) with a lively cannibal complex both inthe realm of myth and ritual mortuary cannibalism. The nature ofthe Yanomamo dichotomy is best exemplified by an observation byChagnon:

The Yanomamo pit Culture againstNature, ••. and see the cosmos in many contextsas an 'either/or' moiety. Thus, and animal,captured in the wild, is 'of the forest,' butonce brought into the village, it is 'of thevillage' and somehow different, for it isthen a part of Culture. For this reason,they do not otherwise eat edible pets ••• forto them, it is similar to cannibalism: eatingsomething 'cultural' and therefore'humanlike' (1983:96).

Through the above example, it is quite evident that the useof dichotomies to analyze the native world view can be of greatuse; it enables a clear cut division of emic categories to beundertaken, in this case revealing the Yanomamo conceptions 'ofthe village' and 'of the forest'. Unfortunately, clear cutdichotomies do not always exist, and while providing problematicanalytical stumbling blocks, these vague dichotomies can be veryrevealing as well. Once again using the Yanomamo, their view ofthe jaguar indicates an ambivalence toward an animal that is atthe same time both almost human and wild, dangerous jungle beast.This ambivalence is reflected in the liminal status granted thejaguar.

Jaguar is peculiar to them (the Yanomamo),for he combines several human capacitieswhile at the same time he is 'natural'. Itis almost as if they bothe respect and fearjaguars, and the 'dual' nature of Jaguarmight be viewed as a kind of contradiction ofthe sort that Levi-Strauss had in mind whenh~ viewed myths as the intellectual vehiclethrough which such contradictions arereconciled (Chagnon 1983:96).

Using other myth material from South America, Levi-Straussfound a widespread cannibal complex among numerous "tribesr..Mythic instances of South American cannibalism frequently occurin origin myths, giving credence to Sanday's notion that"Cannibalism is a behavioral referent of a mythical charter forsociety" (1986:25). Among the myths featuring cannibalism

5

collected by Levi-Strauss are M22, "The origin of the jaguar"(Mataco); M23, "The origin of tobacco" (Toba-Pilaga); M24 "Theorigin of tobacco" (Tereno) i M65, "The origin of fire" (Mbya)(see Levi-Strauss 1964:99-100 140). In each of these myths,cannibalism is the causal el~ment for the origin of importantitems in the cosmological, ritual and survival inventory of eachof these "tribes". '

As Chagnon found cannibalism of "cultural" animals horrificamong the Yanomamo, Levi-Strauss found cannibalism of "cultural"plants horrific among the Timbira. "A Timbira myth (M7~)con~irms that a vague feeling of guilt attached to, th~sagr~cult~ral,process (slash and burn), which made a cert~~n fo~mof cann~bal~sm the precondition of civilized nourishment' (Lev

7­

Strauss .1964:15~). The seeming paradoxical nature of th~sobservat~on ~nd~cates that cannibalism is indeed a very complexconcept, not necessarily restricted to the eating of humans, butextended also,to the realm of the vegetable kingdom. Among somegroups ~f n:t7v~ S~uth Americans (Mundurucu), there was a tabooon fell~ng l~v~ng trees for firewood (Levi-Strauss 1964:151).

It is evident that there is a connection between the treesand h~mans.in t~e minds of these aborigines. A type of,mysticalrelat~onsh~p ex~sts between the two: mankind must term~nate thel~fe of tr~es to cUlti~ate vegetation in order to co~tinue thel~fe of mank~nd. ,In t~~s cyclical arrangement, mank~nd.can ~eviewed as a potent~al k~ller of the very substance that g~ve~ h

7m

life, indirectly threa~ening himself. This confusing scenar~o ~sbest related by Lev~-Strauss who notes that

It is now clear that, for natives whopractice the 'slash and burn' technique, evenvegetarian COoking cannot be dissociated froma kind, of 'cannibalism', which is also~egetar~an. M~n's mortality manifests itself~n two ways: e~ther through a natural death _old age or disease _ as is the case withtrees when they 'die' and become firewood~ orby a violent death at the hands of an enemy,who may be a cannibal _ therefore an ogreand who ~s always a cannibal, at least in themetaphor~c sense, eVen in the guise of thetree feller attacking the living tree(1964:152) •

From the above obserVations it can be seen that these .swiddenhorticulturalists (slaSh and burneri) live in a mutuallysymbiotic relationship with their environment; they make room forthe food crops to grow a~d the crops in turn flourish and,prov~denourishment. In recogn1tion of this complementary relat~onsh~p,the natives have incorpOrated the concept of alternatingcooperation into their World view. As Sanday has so aptly

6

phrased it, "the elementary structureson the establishment of a reciprocalindividual and his or her physical(1986:121) .

of consciousness are basedrelationship between the

and social environment"

Typical of the structuralist interpretation is the notionthat cannibalism need not actually be practiced, but the presenceof reputed cannibalism in myth or ritual is of interest since itinstills fear, acts as the focus ~f ritual drama, or otherwisecreates social forms which can then be studied by theanthropologist. In his research on the Hewa of the WesternHighlands of Papua New Guinea, Steadman has taken the stance that

I have no evidence which suggests thatcannibalism is, or was in the past practicedin Hewa society. The Hew claim that it wasand is - but by witches. But the questionwhether Hewa witches do in fact eat theirfellow man is irrelevant to my explanation ofthe Hewa use of the witchcraft idiom, sincethe idiom is based on the fear that theymight (1975:118-119).

So it is really not necessary for actual cannibalism to be or tohave been practiced in a certain culture for structural analysisto proceed. The meager existence of cannibalism in mythic formor as a normative control can be seen as a mental constructsimultaneously having strong concrete manifestations in a societyin which cannibalism as a fact does not even exist. Poole hasreached the same conclusion in his studies of the Bimin-Kuskusminof the West Sepik Province of New Guinea. While Poole hasobserved real acts of mortuary cannibalism in his fieldwork, hestill feels that, "whether or not Bimin-Kuskusmin anthropophagyis portrayed in belief or in practice, all acts of eating humanflesh are elaborate metaphoric, ritual and symbolicconstructions" (1983:30).

This aspect of the symbolic nature of cannibalism presentsanother side to an already multifaceted phenomenon. To someanthopologists the structural, interpretive, and symbolic natureof cannibalism is obvious. Yet to others, the mental constructsassociated with the cannibalism complex are disregarded in favorof materialist and environmental concerns. Marshalll Sahlins hastied these loose ends together neatly by concluding that, "Theproblem, of course, is that cannibalism is always 'symbolic',even when it is 'real'" (1983:88). Other anthropologists seem toagree that the symbolic interpretation of cannibal myths andrituals is extremely important to solving the puzzle of how thecannibalism complex fits into a certain society. ShirleyLindenbaum, through her research on kuru sorcery and cannibalismamong the Fore of the Eastern Highlands of Papua New Guinea, hasdeveloped a keen awareness of the need for symbolic

7

interpretation of cannibal behavior. She notes that "an adequateanthropological analysis goes beyond the matter of whether or notcannibalism actually occurs, and requires an investigation of thesymbolic or ideological dimensions of reported acts orbeliefs ...Thus, we are confronted with the double, or perhapssimultaneous, tasks of eliciting for each culture a coherent'world view' as well as evaluating those beliefs andrepresentations together with what we observe to be the socialbehaviors and social realities" (Lindenbaum 1983:96). (Thepoints pronounced by Sahlins and Lindenbaum above are seen as thejustifications in the creation of the type "Symbolic Cannibalism"suggested earlier in this paper.)

It is imperative that cannibalism be explained in all itsmanifestations among the diverse cultures where it has beenrecorded. Whether or not the particular "brand" of cannibalismpracticed is ideologically or materially based is secondary tothe anthropological duty of recording, reporting, andinterpreting the act(s) accurately. If the anthropologistsinvolved in the past (and probably future) debates on the natureof cannibalism would stick to trying to faithfully interpretreality instead of collecting data which supports their ownpersonal interpretations, the discipline generally would bebetter off and cannibalism would probably be better understood.With these thoughts in mind, the words of Sanday becomeparticularly relevant to structuralist aims:

Since the main causal variable posited by thematerialist explanation is the ongoingsatisfaction of hunger or protein deficiency,obviously the data must reflect actual asopposed to fantasized or infrequentconsumption of flesh ••• the discussion willinclude the fear of cannibalism, whether ornot cannibalism is thought to be actuallypracticed ••• l will not be concerned withwhether the consumption of human fleshactually takes place, because my focus willbe on interpreting the rituals in which humanflesh is purportedly consumed (1986:8-9).

The cultural materialist school views cannibalism as aphenomenon primarily as a result of environmental constraintsplaced on human populations. These constraints would be the needfor food and a lack of certain types of proteins. Marvin Harrisprobably sums up the goals of cultural materialism best: "theprim~ry . task of anthr9pology is to give causal explanations forthe differences and similarities in thought and behavior foundamong human groups •.• cultural materialism makes the assumptionthat this task can best be carried out be studying the materialconstraints to which human existence is subjected. Theseconstraints arise from the need to produce food, shelter, tools

8

and machines, and to reproduce human populations within limitsset by biology and the environment •.• For cultural materialists,the most likely causes of variation in the mental or spiritualaspects of human life are the variations in the materialconstraints affecting the way people cope with the problems ofsatisfying basic needs in a particular habitat" (1980:5).

If we look at some particular cases it can definitely benoted that such environmental stresses on human populations canbe cited as causes for cannibalism. In his research on nativesof Colombia, Thomas S. Schorr has explained various types ofinternecine hostilities, including cannibalism, as a means ofcoping with restricted or fluctuating environmental resources.Two of these "tribes" were the Gorrons and Pijaos of the WesternCordillera of Colombia, both notorious warriors and cannibals asreported by the early Spanish chroniclers. Schorr saw theirbellicose activities in this light: "Raids, plunder,cannibalism, and capture were adaptive responses,institutionalized and practiced reciprocally when it becamenecessary to alleviate subsistence risks of a given group causedby deficiencies in production as a result of droughts, plagues,floods, or other seasonal variations" (1968:458). Rather thanbeing destructive, wasteful actions of wanton barbarity, theintertribal warfare of Western Cordilleran Colombia can be seenas a "positive" adaptive response to variable environmentalconditions, As the produce fluctuated, the human population alsoneeded to fluctuate to maintain equilibrium with the carryingcapacity of the local surroundings. As a result, the institutionof cannibal warfare developed "stably structured, symbioticpractices of predatory fighting and killing through periodicwarfare and cannibalism, (which) far from being disruptive ormaladaptive, served to maintain each group at a level of optimumpopulation numbers in the face of periodic threats to survival fl

(Schorr 1970:19).

In perhaps the most famous, or rather infamous, case ofinstitutionalized people eating, the Aztecs of the Central Valleyof Mexico practiced a large-scale system ofwarfare/sacrifice/cannibalism during precolumbian and (briefly)postcontact eras. As proposed by Michael J. Harner, the Aztecspracticed this lavish system of sacrifice/cannibalism in order tosupplement their protein deficient diets which were heavilyreliant on the starchy staples of maize, beans, and gourds (seeHarner, 1977). Harris has gone much further, using Harner'sbasic thesis as a springboard for interpreting the Aztec systemas an elaborate meat redistribution chain for the nobility of theempire.

fromhavefatIf

redistribution of meatvictims actually

improved the protein anddiet of the Aztec nation.

Could thesacrificialsignificantlycontent in the

9

the population of the Valley of Mexico was 2million and the number of prisoners availablefor redistribution per annum was only 15,000,the answer is no ... The point should be nothow much these cannibal redistributionscontributed to the health and vigor of theaverage citizen but how much thecost/benefits of political control underwenta favorable shift as a result of having humanflesh to reward selected groups at crucialperiods ... if the meat was supplied inconcentrated packages to the nobility,soldiers, and their retainers, and if thesupply was synchronized to compensate fordeficits in the agricultural cycle, thepayoff for Moctezuma and the ruling classmight have been significant enough to staveoff political collapse (Harris 1977:165­166) .

Marvin Harris unfortunately relies on too many "if" statements inhis theory, making it all the more tenuous without ever testingthe basic hypothesis of his proposal: that it was in fact anactual meat redistribution network. Harris' theory of the Aztecmeat redistribution system is probably the most well-knowninstance of cannibalism, having been featured in the popularbook, Cannibals and Kings (1977). Perhaps in recognition of thenumerous flaws in his Aztec scheme, Harris has reworked some ofthe details of his theory. Now, according to Harris, there wasan added impetus to the development of Aztec cannibalism, that ofthe lack of a large domesticated meatbearing animal.

As in other state societies, the Aztec elitehad to strike a balance between thenutritional benefits provided by human fleshand the political and economic costs ofdestroying the wealth-producing potential ofhuman labor power ••• The reason they (theAztecs) made this unique choice was thattheir system of food production was uniquelydevoid of efficient sources of animal foods.The Aztec had never succeeded indomesticating a single large herbivore oromnivore" (Harris 1985:228). The argumentspresented by Harris make use of facts andclaim to be satisfactory explanations of theAztec ~it~~t~on, however, there is a body ofliterature that presents evidence to thecontrary (see Sahlins 1979a, 1979b, 1979).

Although Harris' explanation may be weak, upon thoroughscrutiny, other evidence exists for cannibalism as a result of

10

marginal environmental conditions, In one particular case, thistime an archeological site, the Richards site . (Fort AncientPhase) in Muskingum County, Ohio, the human osteological remainsprovide indications of cannibalism having taken place. Thesystematic butchering of humans based on the "butcher marks" lefton the bones, similar to marks left on animal bones by aboriginaland early Anglo-European populations, suggests the activities ofcannibalism. Also, the spatial distribution of the refuse pitsin which these remains were found indicates limited temporalityto the duration of cannibal practices.

The refuse pits containing the bulk of thesample were concentrated in Features 16through 23, adjacent features, which issuggestive that the act of cannibalism at theRichards site was restricted to a limitedperiod of time and not practiced on a regularbasis. This evidence might imply that thetype of cannibalism ,may not have beenpreferential, but may well indicate asituation of survival cannibalism (Patterson1977:138) •

Patterson goes on to make a logical deduction from this evidencethat, "An irregular practice of cannibalism could well haveoccurred as the result of food shortage, perhaps during a severe~inter period" (1977:138).

Other evidence for cannibalism as a result of environmentalstress can be found in the American Southwest among Pueblo IIIand Anasazi cultures. At a site called 5MTUMR2346 in the MancosCanyon of southwestern Colorado, human skeletal remains werefound exhibiting splintering and cut marks indicating marrowremoval and butchery practices. Paul R. Nickens summarizes themeaning of this find quite well:

While the presence of cannibalism in theSouthwest is established in the archeologicalrecord, its significance remains to bedetermined. Half of the totaloccurrences ••. report evidence of violentdeath in association with cannibalizedremains, however, no definite magico­religious significance is apparent at any ofthe sites. This information coupled with thevagaries of food supply during Pueblo timesprobably indicates that obligatory oremergency ration cannibalism was practicedduring hard times although this hypothesiscannot be considered conclusive at this time(Nickens 1975:292).

11

There is some evidence for drought conditions in the Southwestbased on dendrochronological evidence for ca. 1222-1227 A.D.(James M. Adovasio: Personal Communication) and also for the lastquarter of the thirteenth century A.D. "There are strongindications that a severe drought extended over the plateau from1276 to 1299, and quite possibly the Anasazi found agriculture asthey had come to depend on it impossible" (Muench 1974:19).

In fact, it may be possible to correlate the climaticdeterioration in the American Southwest to marginal climaticconditions in the American Northeast. This period of poorerclimates is known as the Little Ice Age and may have been aworldwide phenomenon from ca, 1200-1800 A.D. This cooling periodwas probably responsible for a decrease in the number of frost­free growing days, particularly in the northern climes such asthe American Northeast, where cannibalism may have been practicedas a response to harsher climates among such groups as theIroquois, Huron, and Monongahela (William C. Johnson: PersonalCommunication) • In numerous cases it can be seen thatcannibalism exists as a response to environmental stress. Thisis the type of evidence the cultural materialist should look forin trying to prove theories of cannibalism as an adaptation toecological constraints. As we can see, there is a very goodcorrelation of environmentally marginal situations andcannibalism. Interestingly, the kind of cannibalism that seemsto predominate in this situation is of the survival type. Thismakes sense in that what were not usually considered food sourceswere resorted to in times of dire need (except in the cases ofthe Iroquians, where a cannibalism complex already existed).

Switching over to an area far removed from the above, wefind more evidence to support the cultural materialist argument,In his extensive research in New Guinea among the Melpa of theWestern Highlands Province and other groups, Strathern hasnoticed an interesting relationship between cannibalism and thelack of domesticated pigs. Strathern explains,

If we look at the Highlands area ingeographical terms, it is evident that thepractice of cannibalism is (according toreports) associated most clearly withsparsely-populated fringe regions where largeherds of domestic pigs are absent. Althoughwe cannot argue from this that protein-hungrypeople become anthropophagous, there isobviously a material correlation of someinterest here (1982:111).

New Guinea cannibalism chiefly is- practiced in themortuary/funerary mode of ritual cannibalism, therefore making itan exception to the correlation of marginal areas and instancesof survival cannibalism noted earlier. This last discrepancy

12

only adds fuel to the fire of the argument that the materialistexplanation is inherently unsatisfactory as a means to interpretcannibalism as a cultural phenomenon.

Marshall Sahlins is probably the leading adversary of thematerialist school, with strong support from Sanday, Parry, andChagnon. Sahlins' attack principally revolves around the Aztecsacrifice/cannibalism complex as presented by Harris. The mainargument posited by Sahlins is that the ceremonially intricateand symbolically abundant ritual of Aztec sacrifice/cannibalismis too strongly based on an ideology and world view to be evenpartially explained by materialist doctrine.

Clearly the cultural content at issue, thisstupendous system of sacrifice, is too rich,logically as well as practically, to beexplained by the natural need for protein bywhich Harris proposes to account for it. Toaccept his view, we have to make some kind ofbargain with the ethnographic reality,trading away what we know about it in orderto understand it; or at least, it takes aheroic act of utilitarian faith to concludethat this sacrificial system was a way theAztecs had of getting some meat (Sahlins1978:45).

Overall, the materialist explanation is presented by Sahlins asincomplete in its attempt to illustrate that Aztecsacrifice/cannibalism was in fact an ideological system thatsprung from the Aztec's response to a protein-poor environment,In a further attempt to demonstrate Harris' failure to adequatelyexplain the Aztec system, Sahlins goes on to point out thatHarris has neglected a major view of sacrifice that makes theAztec routine all the more logical through a symbolicinterpretation. "What is important here is that in a culinarytheory of cannibalism the ritual as such is without motivation.Nor could it be of interest to Harris that the Logic of Aztecsacrifice conforms to that in the classic texts of Frenchsociology on the SUbject (notably Hubert's and Mauss Sacrifice)."Offered as food to the god, the victim takes on the nature of agod. Consumed then by man, the offering transmits this divinepower to man. Death is thus turned into life: what happens tothe sacrificial victim may be made to happen in reverse to theone who sacrifices" (Sahlins 1978:46).

It is true that the Aztec sacrifice/cannibalism wassymbolically charged. The sacrifice of victims was necessary toplacate the destructive gods of war (Huitzilopochtli) and rain(Tlaloc), both of whom were capable of destroying the nation ifimproperly worshipped. Also, the sacrificial, cannibalisticrituals were performed in a world renewal rite in which every 52

13

years the cosmological cycle had to be restored or else the endof the world was inevitable. "The Aztecs saw war and sacrificeas an eternal order, as a means to an end - the end ofperpetuating a doomed world" (Fagan 1984:236). Numerousanthropologists have voiced their opinions about the failure ofcultural materialism to explain what are clearly ideologicallybased instances of cannibalism. In fact, Sanday has suggestedthat in the Aztec case, the ideological ritual complex existedfirst and perhaps only then was it adapted to changingenvironmental conditions. She has said, "The Aztec response tofamine will, no doubt, be taken by some as good evidence for thematerialist hypothesis ... l suggest, however, that the sacrificialcode was expanded, not created, in response to famine anddrought" (Sanday 1986:183). The Aztec case is probably the mosthotly debated of the issues in the recent cannibalismcontroversy.

Augmenting the argument that the cultural materialistexplanation is insufficient, . there exists some intersting datafrom Benares, India as related by Jonathan Parry, about apeculiar Hindu sect called the Aghoris. These pleasure shunningdisciples of Siva inhabit the cremation grounds and are reputedto be responsible for ritual sacrifices and acts of cannibalism.There is a strong local oral tradition perpetuating the Aghoriacts of cannibalism and similar reports have reached thenewspapers and magazines within the last decade. According toParry, however, the cannibalistic acts of the Aghori are based onritual rather than material concerns. "I am convinced that ifnecrophagy is indeed practiced by any of the Aghoris Iencountered, it has nothing whatever to do with the requirementsof a balanced diet (as Harris 1977, has somewhat implausiblyclaimed for the Aztecs); but is an irregular - perhaps even aonce-off-affair, performed in a ritualized manner at night duringcertain phases of the moon (associated with Siva)" (Parry1982:90).

Perhaps the clearest and best illustrated account of theinadequacy of cultural materialism as an interpretation ofcannibalism is Chagnon's concise example, albeit a hypotheticalone, it is no less relevant.

For example, q valiant warrior overcomes hisequally valiant enemy in a mortal hand-to­hand contest. In celebration of his victory,he rips his enemy's heart out and rituallydevours a portion of it to both honor hisdead enemy - and perhaps acquire some of hisenemy's valor to add to his own. Is thevaliant warrior short of protein, or is hisact more sensibly interpreted as a symbolicgesture? Perhaps the best way to drive homethe argument is to ask whether the taking of

14

Holy Communion is evidence of caloricshortage or a symbolic, mystical - andritually cannibalistic - act of eating the'body and blood' of man called Jesus Christ?Holy Communion falls into a special categoryof cannibalism called theophagy, the Eatingof Gods, and would make little sense ifexplained in terms of calorie or proteinshortages. Most anthropophagy - the eatingof man - is also like this (Chagnon1983:105) .

In a current work on the subject, Sanday has espousedassertions that in some cases the cultural materialistexplanation may be applicable. Evidence presented earlier inthis paper seems to give credence to this conclusion,particularly in cases of survival cannibalism where theenvironmental situation was marginal. However, Sanday goes on tofurther explain that

The significant association I establish•.. between the regular experience of hunger,famine, or protein deficiency and cannibalpractice suggests that the material forcescannot be ignored. However, the fact thathunger is just as likely to characterizesocieties that do not practice cannibalism•.• suggests that the materialist explanationis insufficient (Sanday 1986:x-xi).

In the classic model of Marxist analysis, the infrastructure (thematerial aspects of society) is seen as determining thesuperstructure (the ideological aspects of society). Sanday, infact, turns this concept around in order to explain not onlycannibalism, but implicitly other ideological systems as well.Her explanation is in direct conflict with cultural materialistviews and represents a major premise of structuralism and otherschools. Sanday iterates, "1 suggest that during the formulationof a people's relationship to the world and to other beings,ontological considerations take precedence over the utilitarianconcerns given priority in the materialist point of view .•• andthat these considerations frame a people's response to stress"(1986:229) •

At this point, it becomes necessary to address the cause ofthe recent debate on the nature of cannibalism, namely Arens'book The Man"":'eating Myth ••• (1979). Arens' summary statementof his position reduces the essence of his book to a singlesentence, "although the theoretical possiblility of customarycannibalism cannot be dismissed, the available evidence does notpermit the facile assumption that the act was or ever has been aprevalent cultural feature" (1979:182). It is for this

15

statement, I am afraid, that Arens has paid with his professionalreputation. The ethnographic, archeaological, and ethnohistoricevidence from widely distributed geographical areasoverwhelmingly nullifies Arens' claim. Strong support for theexistence of cannibalism comes from the New World areas of theAmerican Northeast, Southwest, Mesoamerica and South America.Old World areas where cannibalism has existed include India,Africa (not treated in this paper), and the Pacific Islands.Riviere has noted

Arens's catastrophic failure to take accountof well-authenticated cases of cannibalism .•.Arens considers only four other sources onTupi cannibalism. It is not possible to tellwhether he is ignorant of all the others orchose to ignore them ... Bad books do notusually deserve long reviews, and I havegiven this one more attention because it isalso a dangerous book. With little work andless scholarship, it may well be the originof a myth (1980:205,205).

Donald Forsyth has perhaps mounted the most detailed criticism ofArens, including a list of all the sources that verify HansStaden's account of his visit among the Tupinamba which Arensfailed to cite. Forsyth's critique of Arens' work is directedfrom an ethnohistoric viewpoint, and serves as a fine example ofehtnohistoric research in its own right. Forsyth has claimedthat

Arens's treatment of Tupinamba cannibalism isan example of poor ethnohistoric scholarship... because his methods do not meet the basicstandards of historical analysis ••• Arens'swork, at least with respect to the Tupinamba,is marred by inadequate control ofchronology, selective use of data to make hiscase (while ignoring data which contradicthis thesis), misleading and improbableinterpretations of the sources' narratives,and a totally deficient treatment of thecorroborative data available on the matter ofcannibalism•.. For these reasons Arens'snegation of the occurence of customary

cannibalism among the Tupinamba ••• cannot betaken seriously (1985:31).

With these final words on this subject,time to dismiss Arens' conclusionmisinformation that it is.

16

it i~ efficient aton cannibalism as

thisthe

The purpose of the typology suggested earlier in this paperis twofold: first, to summarize and combine previously suggestedtypes and in the process extract redundancies; second, to presenta clearer, more definitive system based on previously pUblishedterminology with categories that will hopefully be satisfactoryto anthropologists from varying (if not conflicting) researchstrategies. The need for a definitive typology is evident from areview of the literature. While many typologies of cannibalismalready exist (Sanday's is particularly impressive), none of theminclude all of the categories suggested in this paper. The mainneed this paper meets is to have a typology of cannibalism thatis concentrated, comprehensive, and significantly consolidated inone place.

Cannibalism, as viewed by the structuralists, is seen as aphenomenon primarily through the mythology of various cultures asa representation of the collective consciousness of the group inquestion. The structuralist interpretation is not reliant on theactual existence of cannibalism, rather it focuses on theanalysis of myths in which cannibalism occurs. As a device ofnormative control, cannibalism can be very effective in a cultureconcerning the social order, while serving at the same time toperpetuate itself as a myth through the reinforcement of normswhich in turn regulate group of individual behavior. A basictool for structural analysis of myth and culture is thedichotomy. Since man universally thinks in terms of binaryoppositions, these oppositions are useful to the structuralistanthropologist in analyzing native myth, culture, and world view.Significantly, cannibalism need not exist in reality, yet theconcrete manifestations of cannibalism in the form of normativecontrols are seen as the key devices by which to understand thenative society.

The cultural materialist views cannibal~sm as an actualityin which the constraints of the environment, namely in the formof hunger and protein needs, cause an adaptive response in thehuman population. The materialist interpretation of cannibalismas formulated by Harner (1977) and expounded by Harris (1985,1979, 1977) uses as its case study the Aztec system ofsacrifice/cannibalism. In this example, the Aztec system is seenas a complex meat redistribution network as a response to aprotein shortage due to a lack of domesticated meat-bearinganimals. Other materialist examples of cannibalism come from theethnohistoric and archaeological records in which instances ofwarfare and survival cannibalism, respectively, are seen asresponses to marginal cannibalism, respectively.

Some anthropologists feel that the materialist explanationis an insufficient approach to cannibalism in many cases. Yet itis obvious that environmental constraints have played a largerole in the development of cannibalism as an adaptive response inthe American Northeast, Southwest, and the Western Cordillera of

17

Colombia. This brings us to a very important conclusion aboutthe nature of cannibalism and, perhaps more importantly, aboutthe nature of anthropology itself. To be successful,specifically in the case of cannibalism (and perhaps in othercases as well), the approach must be multidisciplinary, or atleast multifaceted, as far as concerns the application ofresearch strategies which in turn may yield a more appropriateexplanation in combination than when divided. Marshall Sahlinshas made the critical observation that "The problem, of course,is that cannibalism is always 'symbolic', even when it is 'real'"(1983:88). - This is the most concise statement that bridges thetwo interpretations, structuralism and cultural materialism.Although it is perhaps painfully obvious, what is even morepainful is the fact that in some cases research strategies areadhered to as if they were religious commandments at the cost ofan objective interpretation of an anthropological reality.Shirley Lindenbaum has voiced her opinion on this matter quiteappropriately, "we are confronted with the double, or perhapssimultaneous, tasks of eliciting for each culture a coherent'world view" as well as evaluating those beliefs andrepresentations together with what we observe to be the socialbehaviors and social realities" (1983:96). It is more thanevident that ideological constructs have concrete manifestationsin the real world, sometimes determined by the constraints of theenvironment, and other times determined by the basic human needto understand or control the environment. Using an integratedresearch strategy, allowing the strengths of the one tocompensate for the weaknesses of the other, the anthropologistwould be more able to interpret cultural phenonema, such ascannibalism, in a more accurate, unbiased manner. Once again,Lindenbaum has penetrated to the heart of the issue:

••• myth, ritual, and historic action givemeaning each to the other ... Sahlins (1983)unique dialectic vision •.. does away with theso-called myth/reality problem ..• The strengthof the method lies in the revelation ofsystemic and transforming culturallogics ••. With this approach, the way in whichsymbols reflect and/or mask reality is notconsidered problematic (1983:97).

The time has come for anthropology to mature as a disciplinebased on a "dialectic approach" in which contradictions arerevealed and then synthetically resolved. This end can beachieved by the means of adopting flexible or multifacetedresearch strategies which in turn would allow anthropologistsmore comprehensive analyses of social forms, such as cannibalism.In the end, the individual anthropologist, the discipline itselfand, perhaps most significantly, the public would benefit bybeing relieved of inaccurate and biased interpretations which inturn lead only to the perpetuation of misinformation.

18

Dr. James L. Watson

Dr. William c. Johnson

Dr. Thomas S. Schorr

Dr. Arthur Tuden

Dr. Richard Scaglion

Dr. Jeremy A. Sabloff

Thanks to:

for their ideas and references

19

References Cited

Adovasio, James M.Personal Communication.

Arens, William1979 The Man-Eating Myth: Anthropology and Anthropophagy.

New York: Oxford University Press.

Brown, Paula and Donald Tuzin (eds.)1983 Cannibalism. Washington, D.C.: Society for Psycho­

logical Anthropology.

Chagnon, Napoleon A.1983 Yanomamo: The Fierce people.

Holt, Rinehart and Winston.(3rd ed.) New York:

Dole, Gertrude1962 Endocannibalism Among the Amahuaca Indians.

Transactions of the New York Academy of Science(Series II) 24:567-573.

Fagan, Brian M.1984 The Aztecs. New York: W.H. Freeman and Company.

Forsyth, Donald W.1985 Three Cheers for Hans Staden: The Case for Brazilian

Cannibalism. Ethnohistory 32(1) :17-36.

Harner, Michael J.1977 The Ecological Basis for Aztec Sacrifice.

Ethnologist 4:117-135.

Harris, Marvin1985 Good to Eat. New York: Simon and Schuster.

American

1980 Culture, People, Nature, New York: Harper and Row.

1979 Cannibals and Kings: An Exchange. New York Review ofBooks 28 June:52-53.

1977 Cannibals and Kings. New York: Random House.

Harrison, Thomas H.1948 Living Among Cannibals.

Co. Ltd.London: George Harrap and

Hogg, Garry1966 Cannibalism and Human Sacrifice.

Citadel Press.

20

New York: The

Johnson, William C.Personal Communication.

Kolata, Gina1986 Anthropologists Suggest

Science 2(29) 1497-1500.Cannibalism is a Myth.

Levi-Strauss, Claude1964 The Raw and the Cooked. New York: Harper and Row.

1963 Structural Anthropology. New York: Basic books, Inc.

Lindenbaum, Shirley1983 Cannibalism: Symbolic Production and Consumption.

The Ethnography of Cannibalism. Paula Brown andDonald Tuzin (eds.). Washington, D.C.: Society forPsychological Anthropology 94-106.

1979 Kuru Sorcery: Disease and Danger in the New GuineaHighlands. Palo Alto, California: MayfieldPublishing Company.

Muench, David and Donald G. Pike1974 Anasazi: Ancient People of the Rock.

Harmony.New York:

Nickens, Paul R.1975 Prehistoric Cannibalism in the Mancos Canyon,

Southwestern Colorado. The Kiva 40(4):283-293.

Parry, Jonathon1982 Sacrificial Death and the

and the Regeneration ofJonathan Parry (eds.).University Press 74-110.

Necrophagous Ascetic. DeathLife. Maurice Bloch and

New York: Cambridge

Patterson, Richard P.1977 Prelimanary Report of Fragmented Human Skeletal

Remains from the Richards Site: Evidence ofCannibalism. The Richards Site and the Philo Phase ofthe Fort Ancient Tradition. Jeff Carskadden and JamesMorton (eds.) Occasional Papers Nos. 1-9 of theMuskingum Valley Archaeological Survey 134-143.

Poole, Fitz John Porter1983 Cannibals, Tricksters, and Witches: Anthropophagic

Images Among Bimin-Kuskusmin. The Ethnography ofCannibalism. Paula Brown and Donald Tuzin (eds.)Washington, D.C.: Society for PsychologicalAnthropology 6-32.

21

Riviere, P.G.1980 Review of William Arens' The Man-Eating Myth. Man

15 (1) : 203-205.

New

of thePaulaD. C. :

Sagan, Eli1974 Cannibalism: Human Aggression and Cultural Form.

York: Harper and Row.Sahlins, Marshall

1983 Raw Women, Cooked Men, and Other "Great Things"Fiji Islands. The Ethnography of Cannibalism.Brown and Donald Tuzin (eds.) Washington,Society for Psychological Anthropology 72-93.

1979a Cannibalism: An Exchange. New York Review of Books.22 March:46-47.

1979b Cannibals and Kings: An Exchange. New York Review ofBooks. 28 June:52-53.

1978 CultureBooks.

as Protein and Profit.23 November:45-53.

New York Review of

Sanday, Peggy R.1986 Divine Hunger: Cannibalism as a Cultural System. New

York: Cambridge University Press.

Schorr, Thomas S.1970 Fighting and

Populations:Controversies.Americanistas.

Killing Behavior Between HumanA Reflection of the Current

XXXIX Congreso Internacional deLima, Peru. August 5.

1968 Cuaca Valley Settlements, A Culture EcologicalInterpretation. XXXVII Congreso InternacionalAmericanistas. Republica Argentina 1966. ActasMemorias 1:449-466, Figs. 1-7, Buenos Aires. LibrartS.R.L.

Steadman, Lyle1975 Cannibal Witches

XLVI (2) : 114-121.in the Hewa. Oceania

Strathern,1982

Andrew J.witchcraft, Greed, Cannibalism and Some Related Themesfrom the New Guinea Highlands. Death and theRegeneration of Life. Maurice Bloch and JonathanParry (eds.) New York: Cambridge University Press111-133.

Tannahill, Reay1975 Flesh and Blood: A History of the Cannibal Complex.

New York: Stein and Day.

22