Skip to navigation | Skip to content

Killing Animals: on the violence of sacrifice, the hunt and the butcher



Humans are killing animals.[1]  Killers of animals, they—many of them—are animals who kill.  Other animals kill, of course, but no others, I suppose, carry out the killing we call sacrifice.  Anthropological literature commonly treats sacrifice as a kind of exchange and an act of communication.  But to grasp the kind of killing that sacrifice involves, consider its place within the set of violent acts against animals that also includes the work of hunters and butchers.

In the closing years of the twentieth century, the eastern Indonesian island of Sumba was at a tipping point as the practitioners of rituals for spirits of ancestors, commonly referred to as “marapu people,” were losing ground to an emerging Protestant majority.[2]  As a young fieldworker open to any conversations going, I was privy to the suspicions, complaints, and braggadocio voiced on all sides.  One of the hottest topics of polemic was animal sacrifice.  Christians liked to quote a local Bible teacher: “What use are the ancestor spirits?  They just use up chickens.”  For their part, marapu people would point out that they only kill animals for ritual purposes, and never without an offering prayer to direct the sacrifice to its goal.  By contrast, they would say, “Christians kill with no words.  They are greedy, slaughtering animals whenever they feel like it, just so they can eat meat.”  It’s a familiar contrast—Detienne and Vernant (1989) report ancient Greek views of meat acquisition similar to those of the marapu people.  But what’s striking to me is the similarity in their moral logic, as well as what both parties do not talk about.  Both focus on an instrumental logic of calculated utilities and leave mostly unmentioned both the religious logic of sacrifice, and, above all, its violence.  The Christian polemic appeals to an ethics of expenditure, treating the wrongfulness of sacrifice as a matter of wasted resources.[3]  For their part, marapu people stress an ethics of obligation.  The ethics of obligation is all the stronger given their passionate interest in the pleasures of meat.  They portray sacrifice as an elevated form of self-denial, and almost envy Christians the illicit liberty they have granted themselves to eat as they wish.[4] Like the Gentiles of the New Testament, Sumbanese Christians are distinguished by their exemption from the onerous constraints of (certain) divine laws.  Indeed, so fundamentally secular is their view of meat (as opposed, say, to wafers and wine) that it could lead to the view that all killing is legal except that undertaken for sacred purposes.

Click to enlarge

Illustration by Ed Linfoot

This secular utilitarianism with regards to killing has become a largely unremarked feature of much contemporary Christianity.  It was, for instance, merely part of the taken for granted background to a famous (and by 1993, unconstitutional) city ordinance in Florida, aimed at Cuban immigrants who practice Santería.  The ordinance made it illegal “to unnecessarily kill, torment, torture, or mutilate an animal in a private or public ceremony not for the primary purpose of food consumption” (Palmié 1996: 184).  (A literal-minded reading of this text might lead us to conclude that it’s fine to unnecessarily torture animals as long as one eats them afterwards.)   Although the city council was clearly motivated by anxieties of class, ethnicity, and nationality, what they chose to focus on was, presumably, that practice most readily to be abominated, the so-called “blood sacrifice.”  The objection is all the more striking for this obvious point (and one that helped lead to the defeat of the measure in the courts as contravening the Constitutionally guaranteed free exercise of religion): assuming most citizens to be meat-eaters, no one was objecting to the killing of animals as such.  The utilitarian logic of the butcher remains the unspoken background against which the religious stands out in marked contrast.

The Sumbanese marapu people’s complaint suggests that one ethical objection to the reasoning offered by Christians is that instrumental meat-making denies its own violence.  By contrast, sacrifice thematizes it.  In this respect, the marapu people may be exemplifying a stance that runs through many traditions of sacrifice.  This stance is summarized in the question posed by Veena Das, in her reflection on India’s long textual traditions of reflection on religious ritual: “can the killing of animals in sacrifice be regarded as a dramatization of everyday acts of violence that we commit simply in order to live?” (2015: 3).  Although they certainly wouldn’t say this in their polemics with Christians, I suspect Sumbanese marapu people might well answer, “of course!”[5]

In the Sumbanese polemical exchange, both parties focus on one aspect of sacrifice, self-denial, whose ethical weight can be measured in the measurement of utilities, the units being desired goods foregone.[6]  This is consistent with ordinary English language usage, in which “sacrifice” can mean giving up something.  The good foregone could be as innocuous as skipping a nap in order to take the children to the zoo.  The Oxford English Dictionary illustrates this aspect of sacrifice with an advertisement that appeared in a Texas newspaper in 1930: “Owner Must Sacrifice Must sell at bargain.”  Cast in more ethically freighted terms, the act of giving up something wanted by the individual, to the extent that person is understood in principle to be self-interested, can be seen as the exchange of an object of personal desire for a greater good of some sort.  Here’s the OED again, quoting from a history of England: Henry VIII “never was known to sacrifice an inclination to the interest or happiness of another.”  The sacrifice that Henry VIII never undertook was the denial of his own desires in favor of others’ ends, most markedly ones of more general value.  A variation of the ethical logic of self-denial underwrites sacrifice in on small society in Indonesia’s Maluku islands, where “Eating without some restriction . . .  is inconceivable for the Huaulu.  There is no such thing as a free lunch, and the currency with which one pays for it is not just sweat.  Sweat does not stink high enough to reach heaven; earning one’s meal must ride on something more substantial—some renunciation, some giving up, some sacrifice, in a word” (Valeri 2000: 302).

But there is another definition of sacrifice, which dwells on the act of killing.  The OED quotes Caxton circa 1473: “The egypciens cryed vnto hercules sacrefice sacrefice hym, whan hercules cam in to the temple he sacrefised hym.”  We speak here of death.  It need not be sacred death.  The production of cadavers for scientific research is conventionally referred to as sacrifice.  As one inquirer asks, on Protocol Online: Your Lab’s Reference Book[7]: “I need to sacrifice adult rats to isolate microglia for primary cell culture purposes. I am reluctant to use an overdose of anaesthesia or CO2, etc. as these may have an affect (sic) on the cells. What do you all think about using guillotines? I’ve done cervical dislocation on mice before but I’m afraid doing the same for adult rats may prove difficult.”  No self-denial here, at least not on the part of the sacrificer (no one’s asking the rats).  To whom are these rats being sacrificed?  Perhaps to the greater good of scientific knowledge; or perhaps we can have a notion of sacrifice without a recipient at all.  The conventional use of religious vocabulary in this context is strikingly euphemistic. As if to anticipate the question “what kinds of death count as killing?” posed by Bhrigupati Singh and Naisargi Dave (2015: 232), it responds “not those carried out in the laboratory.”  In this context, the language of “sacrifice” seems to shift the act’s moral weight from one of “killing” to something closer to what Elizabeth Povinelli (2008) calls “letting die.”  We will shortly see other ways to do this.

By focusing on the ethical logic of sacrifice in terms of quantities of utility expended or forgone, Sumbanese Christians don’t have much to say about the act of killing.  They do not portray killing as such to be central to the logic of sacrifice or to its ethical import.  In this respect, they find good company in the classic theories of sacrifice, which tend, with some exceptions, to play down violence in favor of concept, death in favor of exchange.  Reviewing the major western academic theories of sacrifice, Valerio Valeri summarizes them as gift, communion, efficacious representation, and cathartic act (1985: 62).  The exception, of course, is the last of these, which is exemplified by René Girard’s (1977) speculations about the origins of religion in violence.  Given its lack of ethnographic grounding and reliance on broad psychoanalytical speculation, it’s also the theory Valeri can dismiss most handily, and one to which I too will devote no further attention.  Finding Hubert and Mauss (1964) more useful, Valeri’s own position can be put succinctly: the sacrificial victim “is the subject in an objective form” (1985: 65).  In this respect, then, “sacrifice . . .  requires less a theory of the gift than a theory of representation” (1985: 67).  We can take this to mean representation in both senses of the word, a depiction that resembles something else, or the act of serving in place of something else—both of which can, in due course, bring us back to the gift—a nice rendition of sacrifice’s essence as meaningful and communicative.[8]

But what about the act of killing?  Singh and Dave are interested in the difference between killing that is seen as a moral problem and that which is a routine matter of indifference.  They bring into one line of sight both Vedic sacrifices and the industrial slaughter of chickens for the mass market.  Dave works with people who take their activism on behalf of animal rights to extremes that others would find unbearable.  One of them spends his free time administering medicine to the open wounds of stray dogs in the streets of Bombay (Dave 2014).  Singh and Dave assert that “The mood and the mode of killing matters.  Would we be able to say this if it were humans killing one another?  Probably not” (Singh & Dave 2015: 245).[9]  By stressing the quiddity of violence as such, they push back against a familiar anthropological tradition that draws strong conclusions from the concept of culture.  In this tradition, the focus on cultural particularities requires us to relativize our categories to the specific determinants of a given social world (Keane 2003a).  A remark by Signe Howell exemplifies this position: “there is no identifiable element that in and by itself makes the ritual killing of an animal, or human being for that matter, violent” (1996: 10); Das (2015: 1) makes a similar position the starting premise for any discussion of violence.  Both Howell and Das voice an established ethnographic principle that we should denaturalize our categories of thought, and therefore we shouldn’t assume in advance we know what counts as violence.  This principle plays a crucial role in anthropology’s critical epistemological stance toward ethnocentric ontology, and the political implications it draws from that stance.  At any rate, the category of violence is excessively capacious, encompassing everything from the excited sadism of bear baiter or lynch mob to the indifference of the butcher or the professional hitman. Those who obtain meat themselves rather than from the market, and those who have never formed relations with pets, may not see the killing of animals to be violence at all.  But Singh and Dave seem to suggest there are significant limits to how far you can, or should, relativize the violence of killing.  One way to probe those limits is through the work of comparison.

The pleasure of killing

You might think that hunters are those people least prone to sentimentalize the death of their prey, most likely to take a matter-of-fact attitude like that of the butcher in a big city.  Yet here’s Valeri’s interpretation of the Huaulu prohibition on combing one’s hair in front of dying or freshly killed game: “Such an act of self-ornamentation is disrespectful, because it emphasizes the contrast between the triumphantly alive hunter, pleased with himself for his victory, and the wretchedness of the bloody, defeated animal.  The act implies a refusal to acknowledge that human life is made possible by animal death.” (2000: 323).  Rane Willerslev writes of the Yukaghir hunters of Siberia that “killing and eating animals is inherently problematic.  This is due in large part to the fact that the animals that are most desirable as food—elk, reindeer, and bear—are the ones understood to be most like humans in their moral values and rules of conduct.  As one young hunter once expressed it, ‘When killing an elk or a bear, I sometimes feel that I’ve killed someone human.  But one must banish such thoughts or one would go mad from shame’” (2007: 78).  (Notice, by the way, the echoes of the semiosis of representation that underwrites the Durkheimian tradition: the most problematic animals are those that most resemble me, and therefore in whose place I could stand.  We’ll return to this below.)  Both Huaulu and Yukaghir go to elaborate lengths to displace or dissemble responsibility for the killing that the hunt requires—although it’s not always clear through whose eyes that responsibility is seen and judged.

Sacrifice can stimulate an emotional intensification of the more general experience of the taking of life in the process of staying alive.  In Sumba, the most valuable items of personal property are water buffalo.  Although buffalo are of limited practical use, they are the most important component of marriage exchange owed by the groom’s side.[10]  A man without cattle will be absorbed as a subordinate member of his wife’s patriclan, their children lost to his own people.  Cattle are also the most important sacrificial victims in any major ritual, such as house-building, village purification, the reconciliation of enemies, and funerals.  Sumbanese buffalo killing takes place at the apex of a hierarchy of sacrificial value that includes offerings of betel nut and the killing of chickens, pigs, and horses. It is a hierarchy that reflects a number of things, one of which is the kind of labor, the extent of kinship ties, and the powers of exchange relations that are concretized in the very existence of the animal in the first place (see Keane 1997, chapter 3). True to the Durkheimian models, the act is explicitly communicative.  Prayers are spoken over the sacrificial animal before the killing, in order to alert the spirits and inform the animal of the messages it should carry to the world of the dead.  The killing of the animal is a means of getting the message from this material sphere to that other, immaterial one.  After the killing, the response of the spirits must be sought through divinatory reading of the entrails of the victim.

When I lived on Sumba, chicken sacrifice was far more common than that of larger animals, but the act of killing chickens elicited little interest (apart from the post-mortem divination) and was often considered a stand-in for more substantial sacrifices.  The killing of small and weak animals not only lacks drama, it offers little opportunity for spectatorship.  Buffalo slaughter, on the other hand, is a hugely popular spectacle. It takes place in the village plaza, and everyone who is able to watch does so with great enthusiasm.  A good sacrifice may draw hundreds of enthusiastic spectators, old women and young boys alike craning for a better view.  Each beast is led into the center of the plaza in turn, and held there by several young men grasping a rope strung through its nostrils.  As gongs and drums pound an exhilarating beat, a single young man will approach the animal waving a long machete, often prancing or leaping as he does so.  Dodging the long, sharp horns with a display of bravado he takes aim at the jugular.  Ideally the animal is finished off with a single blow, although it may stagger about for some time, blood gushing in torrents, the crowd cheering, before it falls.  The length of its struggle and the manner of its fall are studied attentively for divinatory import by the crowd.  Once down, the next victim is brought out—at a good event involving rich clans, dozens of corpses may pile up in the plaza before the butchering of the meat begins.  The machismo of the killer, the passion of the spectators, destruction of wealth, and the anticipation of vast quantities of tough, gamey boiled meat make this one of the highpoints of village life, persisting into the era of social media, if Sumbanese posts to Facebook are anything to go by.

What is that enthusiasm about?  To start, there is a certain thrill in the sheer display of wealth and its expenditure.  But many spectators also focus on the bravado of the young men who undertake the killing, and the risk at which this places them.  And people seem to find the fatal blow of the machete and the struggles of the buffalo to be fascinating.  On display in the plaza are physical power, domination, fear, the display of athleticism, identification with or a vast sense of distance from the victim. Sadism or empathy, risk-taking, excitement at the dramatic movements, and amusement at the occasional slapstick may all be involved.[11]  Janet Hoskins (1993) points out that people in Kodi, West Sumba, often laugh at the sight of the slaughter. She says this is a nervous response to their ambivalence and, prompted by a Kodi myth about the first subjugation of cattle, suggests they are identifying with the animal but also ridiculing it for allowing them to humiliate it. In her view, the killing works in part to externalize aspects of themselves that they want to eliminate.

Then, the killing produces meat, which people anticipate with enormous relish.  Sumbanese love to eat meat, but, as noted above, marapu people do so only at ceremonial feasts.  These bodily pleasures are inseparable from the giving and receiving they presuppose, the commensality and reciprocity.  Confronted with evidence of killing, we cannot be sure that violence is the principle focus of attention, or even fear and pain, sadism or empathy.  It may also be mere excitement, in which the spectacle of killing is inseparable from the stimulation of being in a crowd – even collective effervescence – and the anticipation of the feast.

So if Sumbanese objectify themselves in the form of the sacrificial animal, they also absorb that objectified beast into themselves in the form of dead flesh.  The sacrificial dialectic begins with externalization (by identifying themselves with the victim, they project certain aspects of themselves onto the animal) that makes possible, through material means, an internalization of certain aspects of the sacrifice (the flesh but not the spirit of the beast).  That is, they internalize the meat that results from the killing.  This is no mere representation: they are very aware of the pleasures of satiety and renewed vigor this produces. The dead body of the animal becomes part of the revitalized living body of the feasters; the animal rendered an object of human actions contributes to their constitution as subjects both through the agency by which they kill the beast and through the act of consumption by which they appropriate it to themselves.  To this extent, Valeri’s concept-oriented model works well: the buffalo victim is an objectification, an externalization of an aspect of oneself.  We can add that it is at the same time a subjectification of the object world, the transformation of meat into life.  But we cannot eliminate from Sumbanese sacrifice the spectacle of violent killing.  In the latter, however, we should not settle in advance whether the dominant key is fear, pain, prowess, sadism, domination, the thrill of the crowd, or the pleasures of the feast—indeed, all of these are in play. The category of “violence” is both too general and too narrow.

There is yet another aspect of killing to consider as well. As Valeri observed, killing dramatizes the ubiquitous experience of transformation or transition from presence to absence: “sacrificial death and destruction are also images; they represent the passage from the visible to the invisible and thereby make it possible to conceive the transformations the sacrifice is supposed to produce” (1985: 69).  We might say the same of Sumbanese killing, which is meant to extend a bridge to the invisible world.[12]  Sumbanese sacrificial animals convey messages between the manifest world of the living and the invisible world of the dead.  In Sumba, sacrificial killing must, as noted above, always include words, and the reading of entrails.  These foreground the fact that the killing is a transition between perceptible and imperceptible worlds (Keane 2008).  The animal is made to die, in order to bring along with it to the world of the dead words that were spoken by humans in the world of the living, and, through the medium of entrail-reading, to convey a reply.

At any rate, perhaps we do not have to decide which aspect of killing and feasting is the key one. These aspects are all bundled together (Keane 2003b).  All components of this bundle of features (the destruction of wealth, the display of social power, domination over the wild, youthful folly, masculine bravado, plenitude, feelings of meaty satiation, aggression, fear, transition from visible to invisible, from living animal to dead meat, and incorporation of edible object into vital subject) are affordances (Keane 2016), which are in principle available for attention, elaboration, and development.  Different components of this bundle may come into play in different circumstances; some that were only latent in one context may become prominent in another.  But I want to argue that what organizes the relations among these is the act of killing that mediates the sacrificers’ capacity to oscillate between modalities of being subject and object.[13]  The violence of the act of killing matters.  Sacrifice is not merely a particular way of entering into exchange, instigating communications, or deploying representations and embodying concepts.  The act of killing is crucial to the particular forms of identification on which the sacrificial movement between subject and object depends.

Animals as victims and as selves

Nietzsche, of course, famously stresses the violence of sacrifice: “When man thinks it necessary to make for himself a memory, he never accomplishes it without blood, tortures, and sacrifice; the most dreadful sacrifices and forfeitures . . . the most cruel rituals (for all religions are really at bottom systems of cruelty)—all these things originate from that instinct which found in pain its most potent mnemonic” (1918: 45).  To be sure, I would hardly take even Nietzsche, for all his mad genius, to be an authoritative guide to ethnography or cognition, or even to really existing religions, but what should we make of the violence of sacrifice?  Is the violence of killing something we can or should relativize as Howell suggests?  Dave sees the cruelty in even the ordinary work of butchers in urban India.  In one scene, she draws our attention to a boy at the butcher’s workshop.  He picks up a chicken that is not quite dead and begins to play catch with it.  “Occasionally the boy would miss, the bird would fall, and the boy would stomp on it and scream at it and then, laughingly, throw it back again.  The boy knew he was playing with something that had once been alive, and that was part of the pleasure” (Singh and Dave 2015: 239).[14]  I want to suggest a parallel to Sumbanese buffalo sacrifice—both forms of violence only function as they do if humans can enter into the animal’s perspective (but only partly).  The boy’s cruelty and pleasure seem to require some degree of identification with the animate consciousness of the chicken.  If the victim were merely a thing, so much meat, the act of killing might not seem particularly violent, nor would there be much play to it—a rubber ball might provide more pleasure.[15]   

But Dave’s objection seems to lead her elsewhere.  The scene, she writes, manifests “a kind of absolute power over animal life and death”—to this extent, we could say the same of buffalo slaughter—but she goes on to add this: “the power to turn animals into things, as if they were never anything else” (Singh and Dave 2015: 239).  Here we seem to encounter a critical distinction among hunters, sacrificers, and butchers.  We’ll come back to this in a moment, but let’s start by asking the hunters whether animals are things or not.  In his reflections on Huaulu, Valeri remarks “where social relations and social conventions exist against the background of a nonhuman world more powerful than the human one, the concept of total ownership, which wholly submits the owned thing to the owner, which in fact makes it totally a thing in the first place, cannot arise. . . . one confronts not inanimate possessed objects, but living, active beings.  The owned is the vanquished, the tricked, the robbed, the prayed for, the exchanged for” (2000: 308).

The instrumental treatment of animals as meat, in this view, resembles a form of slavery, property that derives from an originary act of subjugation—something apparently alien to Huaulu hunters.  The crucial point here would seem to be the dimension of power.  In contrast to the butcher shop, humans in hunting societies do not see themselves as the dominant power.  The difference may not just be one of strength, speed, or weaponry.  Hunters do indeed confront a power differential, but the power in question is not necessarily coercive or violent in nature; rather, it is a matter of the hunters’ lack of control over outcomes, their dependence on the luck of the chase.  A woman in Canada’s Yukon pungently summarizes the contrast between controlled meat production and Kluane hunting practices in remarking that “We don’t harvest animals, we kill them” (Nadasday 2011).

The limits on human efficacy are quite apparent in the Yukaghir world, where the hunter’s success is at the mercy of the willingness of the prey, or of its spirit master, to give itself up to the human.  What’s involved is not a straightforward matter of power dynamics.  Willerslev tells us that hunters (mostly male) see themselves as seducing their prey through a double act of identification.  The hunter identifies with the animal just enough to be able to act in such a way as to persuade the animal to identify with him.  At that point, the prey voluntarily offers itself to the hunter.  This, of course, helps resolve the potential guilt about killing quasi-humans that we’ve seen above: the original guilt induced by identifying the animal as like a human is resolved in part by reversing roles, and identifying the human as like an animal.  But what occludes the guilt is the displacement of agency, so that animal becomes the responsible party for its fate.  There’s a striking parallel in South Asian sacrificial traditions.  Singh, in conversations with Hindus who had recently abandoned their traditional animal sacrifice, asked whether it had always been wrong: “the answer I most often received was that earlier animal sacrifices were not debased because ‘in earlier times the animal of its own accord banged its head at the altar’. . . .  Nowadays, people claimed, it was harder to make moral arguments for animal sacrifice.” (Singh & Dave 2015: 236; see also Das 2015: 4).[16]  In a sense, then, they seem to imagine that the human act of sacrifice has undergone a historical shift from merely “letting die” to “killing.”

Willerslev points out, however, that in Siberia this identification of human and non-human animal is not total—we’re not in the realm of utterly different and self-contained “ontologies.”  So recall Dave’s claim that the butcher’s violence turns a living being into a thing.  For Yukaghir hunters, it would seem, the case is just the opposite.  Willerslev remarks that “the personhood of animals and things is  . . .  something that emerges in particular contexts of close practical involvement, such as during hunting.  Outside these particular contexts, Yukaghirs do not necessarily see things as persons any more than we do, but instead live in a world of ordinary objects in which the distinction between humans and nonhuman objects is much more readily drawn.” (2007: 8, italics mine).  It is precisely the act of hunting that transforms entities that are (more or less) thing-like into those that are (more or less) person-like. It’s important here to add that the personhood of animals is not first and foremost a matter of Yukaghir holding different ideas, worldviews, concepts, beliefs, ontologies.  It emerges from the practical conditions of hunting as such.  Valeri captures this pragmatics succinctly: “hunting, like war . . . , puts one in confrontation with an enemy, a free and resisting being; it requires knowing it or him, judging it or him by standards similar to those by which one judges oneself and one’s fellow beings.” (Valeri 2000: 302).  In other words, the hunter’s identification with his prey is first and foremost a basic practical requirement for success, not a thought experiment or an ontological postulate.  If you cannot put yourself in the place of your prey, your efforts at hunting are not likely to be very effective.  To be sure, this changing of places is an affordance for all sorts of possible elaborations, but at its base, this is less a case of Lévy-Bruhlian mystical participation than it is a matter of good strategy.

But, to repeat, one must be able to move between these perspectives, and not come to a final resting point in one or the other.  Yukaghir hunters “consider it necessary to assume the identity of their prey in order to kill it.  However, if the hunter loses sight of his own human self in this process and surrenders to the single perspective of the animal, he will undergo an irreversible metamorphosis and transform into the animal imitated.  . . . [leading to] ‘othering’ beyond recovery” (Willerslev 2007: 49).  We might say that the same holds for the identification of sacrificer and sacrificial victim.  There should be enough identification for the death to matter, and to serve its communicative or exchange function, but no more than that—the sacrificer, after all, survives.  And for that survival to matter, there must be death, a point that is dramatized by the violence that brings home the reality of the act of killing.

To return, then, to Dave at the butchers.  Reflecting on the boy who plays with the dying chicken she writes “Surely we have seen or can imagine a cat with a ceased-but-not-killed mouse.  Pleasure and play in the face of a waxing and waning life is part of a shared field of vitality and cessation; and thus, perhaps the recognition of playful cruelty is also part of an ethic of immanent obligation.  At what point does play turn into a different kind of violence?  I have never seen a cat breed mice and load them en masse into a truck toward their death . . . . And so I . . . will reserve the right to consider such a system . . . morally deadening.” (Singh and Dave 2015: 239).  Dave’s distinction between kinds of killing, as play and as instrumental means, parallels the marapu people’s criticism of Christians.

The child’s play at the butchers’ may be extreme, although there’s certainly a kind of ritual playfulness in Sumbanese sacrifice.  But we can see a similar distinction in the case of the Yukaghir.  Living in agriculturally marginal lands, and lacking the economic resources to guarantee imported food, they depend on meat for survival.  So hunters kill in order to live.  Were they not instrumental about it altogether, they wouldn’t last long.  And yet this does not eliminate the problem of identification and violence that links hunting to sacrifice: “The moral dilemma that stems from killing these animals is also observable in all the various phases of the hunting process . . . a hunter may develop dangerous feelings of love for his prey and thus fail to kill it.  . . . . In fact, I encountered people who refused to eat bear altogether on the grounds that bears resemble human beings too closely, and to eat them would therefore be to engage in a form of cannibalism” (Willerslev 2007: 78).  Notice again in each case the focus is on a stance toward the act of killing and the victim.  The distinction turns on the subjecthood of the victim, a quality that flickers in and out of view over the course of its life and death.  In fact, I want to suggest that this flickering in an out of view is part of the point of sacrifice, which thematizes the ability to interact with animals as both subjects and objects, and in doing so, reciprocally, facilitates a motility of stances toward the subjecthood and objecthood of humans as well.

Killers as subjects and as objects

So what are the distinctions among kinds of killing?  We’ve touched on a few.  Ritualized killing is an ethics of sacrifice in which the violence matters as much as the representation.  Killing and eating without ritual is an anti-ethics of utility.  Eating without killing (vegetarianism) is pure utility without violence, an ethics of abstention but also one that lacks an identification of the human with a victim.  And ritualized giving without violence—the substitution of a coconut for an animal, for instance–is an ethics of pure representation.

We might summarize the cases we’ve looked at as follows.  The hunter identifies with the victim, which allows him or her to kill.[17]  The butcher fails to identify with the victim which allows him to kill.  Facing both ways, the sacrificer attempts to control the flow of life.  Although both hunter and butcher are capable of oscillating between stances, identifying or not identifying with the victim by turns, the sacrificer thematizes his capacity to effect this shift.  The sacrificer controls and openly performs the shift of stances.  He treats the animal as both subject and object, person and thing.  Perhaps we might say this too: the sacrificer does not only identify with the victim when it is a person or subject—he also does so when it is an object.  Long ago, Godfrey Lienhardt (1961: 296) argued that Dinka sacrifice demonstrates their power of survival.  But of course it also enacts their death.  To the extent Dinka identify with the animal as it dies, the sacrificers not only endow the victim with subjecthood, but in the same move obtain a point of view on the human as an object. They themselves are potentially so much meat. (This possibility, of course, is made most explicit by some Sumbanese warriors in the past who, it is said, would pretend to eat bits of their victims).[18]

Singh and Dave (2015: 236) describe a poultry dealer who came to suffer chronic nightmares about the chickens he had killed, which led him to join the Humane Society.  He experiences not the studied neutrality of his colleagues but something closer to the sacrificer’s identification with the suffering he inflicts and overcomes.  Here we have a step from sacrifice to vegetarianism, by way of intimacy.  What they share is the lack of indifference.  The Sumbanese Christians (at least as portrayed by their critics) are indifferent: they seek only a meal, like the professional butcher.  The sacrificer, on the other hand, has something more at stake.  So perhaps it’s just a step to the intimacy that Dave finds among the radical animal rights activists: “What interests me about this material on animals and witnessing in urban India is, in part, how it . . .  exfoliates the social skin, because it expands the boundaries of possible relationality . . . . For one, the animal subject is brought into its intimate relation with a human through its unfreedom. Second, by entering into intimacy with an animal in pain, the activist seeks not to be more free, but to render herself even more deeply subject to unequal relations of obligation and responsibility” (Dave 2014: 445).  This depiction of intimacy seems to mirror the vitalistic logic of sacrificial conquest as a form of liberation, based on a kind of social relationship with the victim, to the end of human self affirmation in the face of death.


In 1987, a neighbor of mine on Sumba died suddenly while still in hearty middle age.  Apparently the cause was choking.  Umbu Jon, as I will call him, was wealthy by local standards, and much disliked for his stinginess.  Indeed, many who knew him thought that choking on his food was quite an appropriate way to go.  In Sumba’s very Maussian, gift-based social world, Umbu Jon was an outlier, someone who contravened local norms of performative generosity and openness to exchange.  But he was also a local power broker, and so his funeral was expensive and prolonged.  As the day of the entombment approached, the scene grew uncanny.  Most dramatically, Umbu Jon’s son, a proud man who had escaped village life for a government sinecure in the Provincial capital, a status displayed by the civil service uniform he always wore, went mad.  For days, this man, who was supposed to exemplify civilized and rational modernity, would periodically stagger about the village plaza raving, out of his mind, and had to be restrained by the loincloth-clad country folk he would consider beneath him.  On the day of the entombment, the village plaza had been turned to deep mud from a rainy season downpour.  The negotiations of mortuary exchanges were slowed down by the son’s madness, as well as the logistical challenges posed by the number of parties involved.  It was late afternoon before the slaughter began.  By convention, the first animal to die was Umbu Jon’s prize horse.  Then the buffalo were brought onto the plaza, one by one.  On the now blood-soaked mud, the young men found it hard to keep their footing, slipping awkwardly and sometimes falling as they swung their machetes for the kill.  By nightfall, some two dozen beasts lay heaped in the blood-soaked mud, and the butchering and distribution of meat proceeded by the light of kerosene lanterns.  Exhausted myself, and weighed down by this ominous atmosphere, I left for home before the final pieces of meat were sent to their recipients.  It wasn’t until the next morning that I learned the culminating moment of uncanniness.  Everyone was talking about it.  For when the butchering was complete and they had reached to bottom of the pile of animal corpses, the horse could not be found.  Everyone agreed that Umbu Jon, in a final act of greed, had pulled the horse into the spirit world with him.  It was widely assumed this meant as well that his family would fail to benefit from his riches, which sooner or later would be lost to them.

There are, of course, many ways to tell this story.  One might be to demonstrate that, in Sumba at least, sacrifice must be true killing.  When Umbu Jon takes his horse with him, instead of leaving the corpse with the living as meat (and, presumably, letting its spirit travel to the marapu), he is in effect acting as if to deny the adage “you can’t take it with you.”  He does take it with him.  By keeping what should have been given, he denies the finality of the animal’s death, at least for him.  That is, he denies that sacrifice is loss.  It fails as sacrifice, which is why the event will not yield the benefits of sacrifice.  If death is not death, then killing is not killing, not a true loss, and perhaps not even real violence.  The prospect of failure that the horse’s disappearance evoked for Sumbanese gossip suggests that for a sacrifice to be serious, it must mean really giving up something, and there is no giving up more serious than that of a life.  The sheer violence of sacrificial killing assures that seriousness.  Sacrifice starts with making others die.


Bloch, Maurice  1992  Prey into hunter: The politics of religious experience.  Cambridge: Cambridge University Press.

Das, Veena  2015 (2013)  Violence and nonviolence at the heart of Hindu ethics.  In The Oxford handbook of religion and violence.  Edited by Michael Jerryson, Mark Juergensmeyer, and Margo Kitts.  Online version,, accessed 10 May 2016.

Dave, Naisargi N.  2014  Witness: Humans, animals, and the politics of becoming.  Cultural Anthropology.  29 (3): 433-456.

Detienne, Marcel, and Jean-Pierre Vernant 1989 (1979)  The cuisine of sacrifice among the Greeks, translated by Paula Wissing.  Chicago: University of Chicago Press.

Girard, René 1977 (1972)  Violence and the sacred.  Translated by Patrick Gregory.  Baltimore: The Johns Hopkins University Press.

Hoskins, Janet  1993  Violence, sacrifice, and divination: Giving and taking life in eastern Indonesia.  American Ethnologist.  20 (1): 159-178.

Howell, Signe  1996  Introduction  For the sake of our future: Sacrificing in eastern Indonesia.  Leiden: Research School CNWS, pp. 1-26.

Hubert, Henri, and Marcel Mauss.  1964 (1898)  Sacrifice: Its nature and functions.  Translated by W.D. Hall.  Chicago: University of Chicago Press.

Johnson, Walter.  2018.  Guns in the family.  Boston review: A political and literary forum. March 23, 2018. (accessed April 3, 2018).

Keane, Webb  1997  Signs of recognition: Powers and hazards of representation in an Indonesian society.  Berkeley: University of California Press.

__________  2003a Self-Interpretation, Agency, and the Objects of Anthropology: Reflections on a Genealogy.  Comparative Studies in Society and History.  45 (2): 222-248.

__________  2003b  Semiotics and the social analysis of material things.  Language and Communication.  23 (2-3): 409-425.

__________  2008  The evidence of the senses and the materiality of religion.  Journal of the Royal Anthropological Institute.  14 (1): S110-S127.

__________  2016  Ethical life: Its natural and social histories.  Princeton: Princeton University Press.

Kirksey, S. Eben, and Stefan Helmreich.  2010.  The emergence of multispecies ethnography.  Cultural Anthropology.  25(4): 545-576.

Kohn, Eduardo.  2013.  How forests think: Toward an anthropology beyond the human.   Berkeley: University of California Press.

Lienhardt, Godfrey  1961  Divinity and experience: The religion of the Dinka.  Oxford: University of Oxford Press.

Lowe, Celia.  2006.  Wild profusion: Biodiversity conservation in an Indonesian archipelago.  Princeton: Princeton University Press.

Nadasday, Paul  2007  The gift in the animal: The ontology of hunting and human-animal sociality.  American Ethnologist 34(1): 25-43.

__________.  2011  “We don’t harvest animals; We kill them:” Agricultural metaphors and the politics of wildlife management in the Yukon. In Knowing nature: Conversations at the intersection of political ecology and science studies.  M. Goldman, P. Nadasdy, M. Turner, eds. Pp. 135-151. Chicago: University of Chicago Press.

Nietzsche, Friedrich 1918 (1887)  The genealogy of morals.  Translated by Horace B. Samuel.  New York: Boni and Liveright.

Palmié, Stephan  1996  Which centre, whose margin?  Notes towards an archaeology of US Supreme Court Case 91-948, 1993 (Church of the Lukumi vs. City of Hialeah, South Florida).  In  Inside and outside the law: Anthropological studies of authority and ambiguity,  Edited by Olivia Harris.  London and New York: Routledge, pp. 184-209.

Povinelli, Elizabeth A.  2008.  The child in the broom closet: States of killing and letting die.  South Atlantic Quarterly.  107(3): 509-530.

Singh, Bhrigupati, and Naisargi Dave  2015  On the killing and killability of animals: Nonmoral thoughts for the anthropology of ethics.  Comparative Studies of South Asia, Africa and the Middle East.  35 (2): 232-245.

Valeri, Valerio  1985  Kingship and sacrifice: Ritual and society in ancient Hawaii.  Translated by Paula Wissing.  Chicago: University of Chicago Press.

__________  2000   The forest of taboos: Morality, hunting, and identity among the Huaulu of the Moluccas.  Madison: University of Wisconsin Press.

Willerslev, Rane  2007  Soul hunters: Hunting, animism, and personhood among the Siberian Yukaghirs.  Berkeley: University of California Press.

  1. [1]An earlier version of this article was presented at a workshop on “Vitalism Sacrifice and Utopia” funded by the European Association of Social Anthropologists, the University of Bergen, and HAU, June 2016.  I am grateful to Bjorn Bertelsen, Ruy Blanes, and Giovanni da Col for their invitation, and to all the participants for stimulating discussion.
  2. [2]Marapu refers to ancestor spirits, the most salient of the variety of entities involved in local ritual practice (see Keane 1997).
  3. [3]The theologically weightier accusation would focus on the logic of the fetish, the falsity of the spirits, and the sheer materialism of sacrifice (see Keane 2007).  But these more rarified discussions were not foremost in the everyday conversations of ordinary folks.
  4. [4]Since marapu people do consume the meat of sacrifices, the Christian accusation of wastefulness stands of shaky ground.  Complaints of both church and government (concerned with the putative effects of sacrifice on economic development) dwell on certain styles of competitive feasting that do indeed consume large numbers of costly animals, but the vast majority of marapu sacrifices are modest affairs.
  5. [5]The violence of warfare and the ritualized taking of heads lay just beyond the memory of the oldest people I knew in Sumba, but were still vividly recalled as terrifying and fascinating.  Presenting the head at the harvest altar at the center of the village, fighters ritually linked killing to vitality.  The link recurs in the Sumbanese account of the origins of agriculture, a version of a story widespread in Southeast Asia, in which a woman is killed and dismembered, her dispersed body parts giving rise to the various food crops.
  6. [6]On the role of self-denial in the concept of ethics, see Keane 2016, especially chapters 1 and 2.
  7. [7], accessed May 4, 2016.
  8. [8]Bloch’s theory of “rebounding violence” (1992) is also, ultimately, centered on representation.
  9. [9]Yet even the United States, which ranks near the top of the list in the execution of prisoners, recognizes a distinction between that which is and is not “cruel and unusual punishment,” and takes precautions to neutralize the executioner’s emotional response to the act of killing.  Warfare too is supposed to be governed by certain rules, whatever the realities of the matter may be.
  10. [10]Sacrificed buffalo are sources of meat, but they are not part of the subsistence diet of any Sumbanese.  Aside from that, in those few parts of Sumba with sufficient rainfall to permit wet rice agriculture, buffalo herds are put to work trampling the fields in preparation for planting, but across most of the island this practical application of the animals is not an option.  They are never used as draught animals.
  11. [11]To be strong and strike fear in others is an important masculine (and to a lesser extent, feminine) value for Sumbanese who take pride in their being “fierce people” (tau bani).
  12. [12]Valeri points out this transition works for vegetable offerings too: “decomposition . . . marks the separation from the human and visible world” (1985: 69).
  13. [13]For purposes of this essay, I do not distinguish among roles such as sacrifier and sacrificer (the one carrying out the act) and sacrifier (the one on whose behalf the act is done); see Hubert and Mauss 1964: 10.
  14. [14]We don’t know much more about this boy than this brief description.  But we might glean some sense of the potential mutuality of sadism, machismo, and disgust from historian Walter Johnson’s vivid recollections of his childhood in the hunting culture of the American Midwest, far though that is from an Indian butcher shop (Johnson 2018).  Of the preparation of his first kill, he writes “I will never forget the way that place smelled: humid, oleaginous, nauseating. . . . [The prepared geese] were wrapped in butcher paper, and the light red watercolor stains had already begun to seep through in several places.  The aspects of my psychology and constitution that . . .  still lead me to feel a little queasy when I see chicken legs in the store or cooked up for dinner, or people put to bloody work, had a simpler name that day. I experienced them as personal weakness, as not being quite “grown-up” enough.”  Paired in memory is this: “my father’s friend shot [the turtle] again, and then again and again and again, as it kept surfacing to breathe, fighting for life. ‘Fucking thing won’t die,’ he said, at first bemused, and then finally angry, like it was ruining his whole fucking day.”
  15. [15]This essay was prompted by a call to reconsider the venerable topic of sacrifice.  But in the process of writing, it became apparent to me it should speak as well to less agonistic aspects of human—animal relations (e.g. Kohn 2013, Kirkey and Helmreich 2010, Lowe 2006), a larger context to which I am unable to do more than gesture in this essay.
  16. [16]For an insightful account of the contrast between the violent agency of the killer and the voluntary self-offering of the victim, in a society of hunters and trappers in the Canadian north, see Nadasday 2007.
  17. [17]The hunters in the literature I discuss here are mostly male, although they may require spiritual assistance from women who remain at home.
  18. [18]I was told that head hunters (as recalled by old people remembering even older stories) would take pieces of their victim’s liver and cook them on a skewer, interspersed with pieces of pork.  Then they would demonstrably eat the pork pieces, but on taking the human meat, they would shout “yum! yum!” while feinting to the mouth, then toss it over their shoulders.

Please join our mailing list to receive notification of new issues