Elaine Scarry’s (1985) argument regarding the relation between torture and pain has served as an explicit or implicit theoretical inspiration in many of the ethnographic studies of people’s experiences of war and violence and loss of epistemological framework (Feldman et al. 1993; Green 1995; Jackson 2002; Maček 2000a, 2000b; Nordstrom 1997; Povrzanović 1993, 1997; Scheper-Huges 1992; Van de Port 1998). Her basic contention is that the feeling of physical pain is an irreducible bodily experience that cannot be objectified by anything outside the body including language. This is why torture is such an effective instrument of power, because the inexpressible and non-communicative feeling of pain, the ‘unmaking of the world’, as she calls it, that takes control of the victim can only be alleviated through a new version of the world − another version of the Truth (see also Löfving and Maček 2000: 5; Frykman 2002). Based on ethnographic fieldwork in war-torn Mozambique, Nordstrom (1997) expands Scarry’s idea of torture’s ‘unmaking of the world’ to the general wartime situation experienced by civilians. Nordstrom sees the structure of this experience as bearing clear parallels to Bakhtin’s idea of the grotesque:
When the familiar and everyday are turned into implements of torture and murder, the familiar everyday world is rendered grotesque – not merely by the fact of the present terror and repression, but by the enduring nature of association. (p. 168).
She offers an example of a man who had his throat cut in the family’s mortar by intruding soldiers, the whole village being forced to watch, and she asks if the onlookers will ever be able to use a mortar without having the drama run at the back of their minds. The strength of terror, she argues, exists in its ability to break down the everyday world and destroy the webs of significance into which people’s self-identities are woven. In this way terror was used intentionally by all parties in the war in Mozambique to eliminate political will and military resistance among civilians. But besides unmaking the world, terror also incites different kinds of resistance: retelling the grotesque practices and linking them with the perpetrators, rebuilding broken social relations and institutions of the community, and imagining a world beyond that could function as a source of creativity in everyday life. When listening to the horrible atrocities to which people had been subjected, Nordstrom was furthermore impressed by the concurrent life-generating creativity of her informants. In sum this made her see how in Mozambique peace and solid social relations were built from the bottom up. In a related but less optimistic study of the siege of Sarajevo, Ivana Maček (2000a, 2000b) invents the concept ‘the deserter’s mode of perception’ in an attempt to communicate people’s war experiences. The deserter’s mode is a process “… in which all individuals, faced with unexplainable violent disruptions of their lives, constantly engage in making sense of their situation” (Maček 2000a: 240). It is a sense-making process whereby people, in their attempt to categorise and understand the situation, easily fell victim to the nationalistic discourse that swept the country.
A problem with some of the ethnographies focusing on experiences of war and violence is that bare survival becomes labelled as resistance. This either romanticises people’s coping strategies or expands and thereby weakens the concept of resistance. However, it does focus on a crucial discussion: whether the creative component is violence (as the structural perspective has it) or people’s strategies and reactions toward violence. Does violence create identity, or is it people’s reactions to violence that create or destroy positions of identity?5 The focus on the disintegration of everyday life furthermore questions the concept of violence. Many scholars have asked: Is violence only physical? (See for instance Riches 1986; Nordstrom and Martin 1992; Nordstrom and Robben 1995; Simons 1999). I only want to add that violence does not need to physically attack the body to produce an ‘unmaking of the world.’ A constant threat of violence can disrupt the fabric of everyday life (see, for instance, Gilliland 1995; Ross 2001).
Narratives
It is difficult to uphold a division between, on the one hand, the experience of loss of meaning and, on the other, the narratives people use to explain and make sense of this loss, especially if both matters are only revealed to the ethnographer in communication. That is to say, what part of communication expresses the pure and unmediated experience of violence and what part is a narrative representation of it? In medical anthropology, the idea of pure existential experience has flourished (Kleinman and Kleinman 1995, Good 1994); the same can be considered true in experientially oriented war anthropology. The critical philosophical problem, however, is that experience must always pass through language to be visible/audible (Mead 1967). Some studies have nevertheless focused more on the narratives used to make sense of the violence experienced and the destroyed everyday world than on the ‘pure experience’ that the unmaking of the world can be said to represent − which may be due to the ethnographer’s arrival after the violence has stopped (Zur 1998; Warren 1993; Malkki 1998; Mehta 2001; Jansen 2000a, 2000b; Das 1990, 2001; Perera 2001; Kolind 2006, 2007; Jackson 2002). In these studies, the concept of narrative6 (or similar concepts) is often used to systematise informants’ more or less ordered and identifiable versions of events and relate them to the overarching discourses in which they are cast. In a study of Hutu refugees’ memory of violence, Malkki (1998), for instance, found what she calls a ‘mythico-history’, or cosmological ordering of the past, a ‘world-making.’ The Hutus construct themselves in their attempt to seize history, condemn the Tutsis’ atrocities, and legitimise their own position in Rwandan society. Malkki (1998) does not try to sort out true facts from invented ones in these mythico-histories:
Different regimes of truth exist for different historical actors, and particularly historical events support any number of different narrative elaborations. (p. 104).
What her analysis misses, though, is a consideration of other possible ‘regimes of truth’ or discourses that affect the Hutus’ narratives. In a more phenomenologically inspired narrative approach, Jackson (2002) focuses on what he calls ‘the politics of storytelling.’ In the face of violent events and other disempowering circumstances, storytelling (narratives) – understood as “a result of ongoing dialogue and redaction within fields of intersubjectivity” (ibid. 22) – can be seen as a way to transform private meanings into public ones and to sustain a sense of agency, both vital aspects for human existence. In other words, the very act of putting experience into words (and thereby sharing it, making it part of the social) is theorised as a necessary human capacity to overcome the ‘unmaking of the world’ and recreate a meaningful life.
However, it is also possible to think about narratives and people’s categorisations as less consistent and static than the prevalent use of narratives seems to indicate. When primarily focusing on the ability of narratives to order the world cosmologically and morally, one may overlook the more chaotic, fragmented and contradictory aspect of social life and its strategic uses, especially in post-war societies. The branch of anthropology inspired by Mary Douglas, which basically sees human activity as cosmological ordering of the world, may overlook another feature of social interaction: the persistent situational and strategic mixing of categories. In a study of post-war Croatia, Jansen (2000b, 2006), for instance, shows how on the surface people reproduce the nationalistic rhetoric that has poisoned the whole area, but on closer inspection what looks like conformism is rather a manifestation of agents’ situational and concrete strategies. For instance, a Croat family (from Bosnia) living in a Serbian house in Croatia evoked the authoritative nationalist discourse of ‘language right’ (even though the languages are nearly identical and even though before the war the family lived peacefully with their Bosnian Serb and Muslim neighbours), which allowed them “to resist subjection to another, possible threatening discourse, that of rights of property and return” (Jansen 2000b: 14). This example shows the intricate relationship between discourse and agency, or the way in which people’s narratives of the war are both situational and strategic.
For some scholars the recent shift in the anthropology of war and violence
