Here’s another snippet that will probably be dropped from the chapter. It deals with an issue I’ve discussed earlier, as k-punk has observed: the relation between terror and narrative.

There’s hardly a better example of this (non)relation than Alan Clarke’s film Elephant (1989), simultaneously the most and the least eloquent of statements about terror. The film is almost devoid of dialogue, and consists of a series of assassinations carried out by various un-named, unidentified characters in a depopulated, everyday suburban landscape of shops, factories, parks, gas stations and so on. The camera follows silent and seemingly ordinary figures who make their way determinedly through the city until they come across another equally anonymous figure, and then remorselessly, unfailingly, shoot.

still from Elephant
One assassination follows another, relentlessly, horrifically, without explanation or apparent meaning. More emphatically even than Krzysztof Kieslowski’s Short Film about Killing (1988), Elephant makes no attempt to justify or explain its serial murders. But the film’s silence is unbearable. We inevitably attempt to construct some kind of story, some kind of frame within which to situate the slaughter, and so to relieve (displace) the shock that it causes us. As Richard Kirkland puts it, “the discrete autonomy of Elephant‘s violence is fundamentally compromised by the viewer’s endless and troubled search for narrative” (8).

So once it is “understood” (most likely thanks to its paratexts) that the film is set in Northern Ireland during the troubles, the viewer starts to elaborate a narrative that will give the meaning and logic to the killings that the film resolutely denies us: for instance, we might read them in terms of a “cycle of violence,” “tit for tat killings” performed alternately by Loyalists and Republicans. An elephant never forgets. But in that the film has forced us, its audience, to come up with these clichés, it has also foregrounded the extent to which all discourse about terrorism is imposed upon events and bodies that otherwise stop interpretation short.

Even, in the end, the judgment that such killings are “senseless” (as they are so often described in hackneyed journalistic reports) is itself part of a narrative that aims to give sense to what otherwise subverts the distinction between sense and senselessness.

(At this point let me add a heartfelt recommendation of Robert McLiam Wilson’s Eureka Street, an extraordinary novel that also makes the point that terror is most fundamentally narrative’s interruption.)

Gus Van Sant’s movie Elephant was inspired by Clarke’s film. Unlike its predecessor, Van Sant’s movie does feature dialogue. But what’s interesting here is the way in which language becomes no more than sound, for instance in the scene with the girl in the swimming pool. Van Sant perhaps points to another possible source of Clarke’s title, Bernard MacLaverty’s description of the Troubles as like having an elephant in your living room: so incomprehensible that directly addressing the issue is impossible. All the talk around violence inevitably misses its point(lessness).

One could say much more along these lines about the killing of Jean de Menezes. I don’t buy either the state’s self-justification or the left’s critique of the state. The left seeks explanation by invoking cover-ups, racism, or conspiracy theories. The police and government declare that the shooting in Stockwell was a “tragic mistake.” But the distinction between truth and error, innocent and guilty, intention and accident, is now strictly undecidable in our contemporary control society.


I’m starting to revise another chapter of my book manuscript, and there are a number of short, fairly self-contained, readings that will most likely disappear. One is about Nancy Scheper-Hughes’s Death without Weeping (a title that always reminds me of Nick Cave’s magnificent “The Weeping Song”).

Scheper-Hughes picture of child in coffinScheper-Hughes’s book (which also forms the basis for a special issue of New Internationalist) is a study of women in a Brazilian shantytown and their attitudes in the face of extraordinarily high rates of infant mortality, “the routinization of human suffering in so much of impoverished Northeast Brazil and the ‘normal’ violence of everyday life” (16). As such it is an investigation into affect. Scheper-Hughes aims “to recuperate and politicize the uses of the body and the secret language of the organs” (185). She shows how fear, grief, mourning, joy, anger, and so on, make up a “political economy of the emotions expressed in the somatization of scarcity and deprivation” (326).

Scheper-Hughes writes of “emotions,” but I think it is better to describe her project in terms of affect. Unlike emotion, affect has no clear subject or object. So if I fear the dark, that is a feeling that I have about something. It is an emotion: personalized, individualized, tied to an object. In affect, by contrast, subjects are overwhelmed by and dissolved in a feeling that can seem to encompass the whole world. Emotions are personal; affects are impersonal.

Indeed, Death without Weeping indicates that emotion is secondary to affect. In the shantytown, individualized emotions expressed for a particular object by a particular subject are slow to emerge. Because infant death is so prevalent, feelings are not at first ascribed to persons. Infancy in the Alto do Cruzeiro (site of Scheper-Hughes’s fieldwork) is fundamentally impersonal:

The women of the Alto are slow to “personalize” infants by attributing specific meanings to their whimpers, cries, facial expressions, flailing of arms and leg, kicks and screams. [. . .] Alto women do not scan the infant’s face to note resemblances to other family members. Naming practices follow a similar logic: many Alto infants can remain unnamed and unbaptized until they reach their first birthday. (413)

Alto women avoid (or delay) interpretation and the attribution of meaning. They do not weep for what the death of a child means; the care of these unnamed children is not a cognitive process that would accord particular weight to their personal individuality.

But the lack of emotion does not imply that affect is absent; it is just that “the affection shown the infant and young baby is general and nonspecific. ‘Who doesn’t enjoy a baby?’ people ask” (415). While neither the object (the baby) nor the subject (the mother) are personalized, an affect of enjoyment (“Who doesn’t enjoy?”) encompasses and supersedes individual persons.

See the photo above: the child is almost buried by the flowers that crowd his (or is it her?) coffin. Even his face, marker and sign of individuality, is half obscured. It is not that there is no affect attending the child’s funeral; far from it, as the coffin is overflowing with these tokens of love. But the affect is general, depersonalized.

Depersonalization is not homogenization. Affect traces what Gilles Deleuze terms “singularities,” here the whimpers and cries, kicks and screams, that make childcare anything but monotonous. Deleuze describes infancy precisely in terms of the prevalence of singularities over individuality: “very small children all resemble one another and have hardly any individuality, but they have singularities: a smile, a gesture, a funny face–not subjective qualities” (Pure Immanence 30). In rejecting the logic of representation, the women of the Alto emphasize these singularities of the infants in their care: by refusing to “note resemblances to other family members” (Scheper-Hughes 413), the mothers highlight the infant’s incomparable power to affect and be affected.

Individuation would mean seeing meaning elsewhere: reinforcing filiality, referring infancy to adulthood (this kick or that smile finds its model in a paternal gesture or a shared familial feature), making individuals out of infants at the cost of subordinating them to a transcendent domestic hierarchy. Hence it is individuation and the social rituals of naming and baptism, not the impersonality of a generalized affect, that homogenize and impose social stasis upon the growing child. In the meantime, before this is allowed to happen, in the Alto “small children circulate among relatives and are often reared by more than one mother; on moving into a new household, the child may be given a different name or nickname” (414). Each Alto child is a multiplicity, flexibly adopting a number of different social roles and incarnating multiple identities.

Only later, and “gradually and slowly,” does the Alto infant come “to earn his personal claim to full human status and with it his claim to a personal name and his right to the affections and passionate attachment of his mother” (415). But as multiform affect gives way to the attachments of a subjectified emotion (governed by rights) linking mother and child, the resulting humanization is also a limitation. Multiple and mobile singularities are reduced to a single, fixed identity. One of Scheper-Hughes’s informants states that before this point “the infant is without history. The infant’s story is not yet made up” (437). But we could also say that the child incarnates a pluriform history, or multiple potential stories, yet to be reductively shaped by the single linear narrative that will lay down the law for his or her future.