The Eye of the Beholder: Violence as a Social Process (2011)

urn:nbn:de:0070-ijcv-2011276 IJCV: Vol. 5 (2) 2011, pp. 345 – 356 The Eye of the Beholder: Violence as a Social Process Teresa Koloma Beck, Willy Brandt School of Public Policy, University of Erfurt, Germany Vol. 5 (2) 2011 Editorial (p. 236) Focus: Youth and Violence Editorial Remarks: Youth at Risk Wilhelm Heitmeyer / Steven F. Messner (pp. 237 – 239) Formations of Violence in Post-Dictatorial Contexts: Logics of Confrontation between the Police and the Young Urban Poor in Contemporary Argentina Alejandro Isla / Daniel Pedro Míguez (pp. 240 – 260) The Central American Fear of Youth Anika Oettler (pp. 261 – 276) The Boys are Coming to Town: Youth, Armed Conflict and Urban Violence in Developing Countries Josjah Kunkeler / Krijn Peters (pp. 277 – 291) Governance, Security and Culture: Assessing Africa’s Youth Bulge Marc Sommers (pp. 292 – 303) Eastern European Transformation and Youth Attitudes Toward Violence Eva M. Groß / Berit Haußmann (pp. 304 – 324) Intergroup Conflict and the Media: An Experimental Study of Greek Students after the 2008 Riots David Hugh-Jones / Alexia Katsnaidou / Gerhard Riener (pp. 325 – 344) Open Section The Eye of the Beholder: Violence as a Social Process Teresa Koloma Beck (pp. 345 – 356) Women without Arms: Gendered Fighter Constructions in Eritrea and Southern Sudan Annette Weber (pp. 357 – 370) Spousal Violence against Women in the Context of Marital Inequality: Perspectives of Pakistani Religious Leaders Rubeena Zakar / Muhammad Zakria Zakar / Alexander Krämer (pp. 371 – 384) This work is licensed under the Creative Commons Attribution-NoDerivatives License. ISSN: 1864–1385 346 IJCV : Vol. 5 (2) 2011, pp. 345 – 356 Beck: Violence as a Social Process The Eye of the Beholder: Violence as a Social Process Teresa Koloma Beck, Willy Brandt School of Public Policy, University of Erfurt, Germany A triangular reconstruction of the social dynamics of violence offers a means to bridge the gap between research on the micro- and meso-level dynamics of violent interaction on the one hand, and theories of power and domination on the other. The origins of this approach are found in the phenomenological programme of social science violence research formulated by German sociologists in the 1990s (Sofsky, von Trotha, Nedelmann, and others). Reconsidering their arguments in the framework of social constructivism, this article reconstructs violence as a triangular process evolving between “performer”, “target” and “observer”. Disentangling the dimensions of the somatic and the social shows, however, that these are not the fixed roles of agents, but changeable modes of experiencing violence. Violent interaction uses the suffering body to stage a positional asymmetry, i.e. a distinction between strength and weakness, between above and below, which can be exploited for the production and reproduction of social order. In the late 1990s and early 2000s, violence research in Germany experienced a renewal, against the background of a rising number of violent conflicts in the post-Cold War world and an ongoing debate about organised violence in Nazi Germany. Two approaches became particularly significant in the social sciences.1 One was the the so-called “Berlin School”, which formed around the works of the Berlin-based anthropologist Georg Elwert and his concept of markets of violence (Elwert 1997, 1999).2 The second approach came to be known as phenomenological violence research and was inspired by the (highly controversial) studies of the sociologist Wolfgang Sofsky (1993, 1996, 1997, 2003).3 At a time when dominant discourses emphasised the barbaric and irrational character of con1 This article and the discussed approaches focus on the social dynamics of violence and the gaps in social science research on the issue. 2 For a broader discussion of the approach including case studies that employ it see Eckert (2004). Elwert’s concept of violence was an integral part of his understanding of anthropology as a discipline, coined “Sceptical Social Anthropology” by Thomas Hüsken (2004). 3 The linguistic style of Sofsky’s work is particularly striking. In an almost literary approach, Sofsky temporary violence,4 the Berlin School and phenomenological violence research set out to systematically analyse its functions in processes of social structure formation. They took different approaches: In the framework of the anthropologically inspired Berlin School, violence was conceived as one possible form of human action; starting from this assumption, research investigated the interrelations between violence and other forms of action, as well as between violence and the formation of social structures.5 The phenomenological approach, by contrast, started from the observation that social science violence research so far had neglected the phenomenology of violent interaction; therefore, little is yet known about the social dynamics of the violent moment itself.6 Against the back- composes “thick descriptions” of idealtypical psychological and social dynamics of different forms of violence. Von Trotha later coined the approach “theoretical ethnography” (1997, 24). For a broader discussion of the approach see Trotha (1997). 4 The leading paradigm at the time was the “new wars” theory (Kaldor 1999; Münkler 2002, 2005). 5 In this logic, Elwert’s thinking, for example, focussed on the interrelations between violent action and exchange and reconstructed the emergence of a particular type of social structure, which he called “markets of violence” (Elwert 1997, 1999). 6 It was only towards the end of the 1960s, against the background of a statistical increase in acts of violence in many industrialised countries, that violence entered research agendas (Imbusch 2002, 26). In a recent seminal work the German social scientist Jan Philipp Reemtsma explores the link between violence and the project of modernity, as well as the role of the social scientist within it (2008). See also Trotha (1997, 10–16). 347 IJCV : Vol. 5 (2) 2011, pp. 345 – 356 Beck: Violence as a Social Process ground of a phenomenologically influenced sociological thinking, scholars argued for a social theory of violence which starts from a reconstruction of the dynamics of violent interaction (Trotha 1997, 9–20). While the Berlin School became influential for empirical research on organised violence, the debate about a phenomenological renewal of violence research remained limited to German sociology and hardly outlasted the academic careers of its founding figures. The key postulate, to combine the phenomenological analysis of violent confrontations with social theory, was never fulfilled.7 This article addresses that gap, seeking to outline a social theory of violence by reconceptualising central arguments of phenomenological violence research in the framework of a particular school of social thought: constructivism. The central question is: how can social constructivist thinking contribute to bridging the conceptual gap between the dynamics of violence on a micro-level and processes of social structure formation? The first section reviews the demands of the phenomenological school of violence research and introduces the etymological distinction between the transitive and intransitive meanings of “violence”. Sections two and three go on to present a theoretical framework for violence research derived from a combination of the phenomenological approach on the one hand and social-constructivist thinking on the other, in which the “observer” plays a key role. I deliberately avoid arguing in the language of any one particular socialconstructivist theory, so as to maintain the general potential of such an approach. Section four outlines some implications of this approach for empirical violence research. The basic argument is that in a social-constructivist framework violence has to be conceived not as a dyadic, but as a triangular dynamic. The social dynamics of violence do not 7 A major argument against such an approach was recently formulated by Slavoj Žižek, who argues that the “overpowering horror” of violent acts and the resulting empathy with the victim prevent us evolve simply as a physical confrontation between a “perpetrator” and a “victim”, but constitutively include a third position, namely the “observer”. I distinguish between “performer”, “target” and “observer” as three modes of experiencing violence. In this perspective, violence is defined as a correlation between inflicting and suffering as observed by a third party. This approach analytically disentangles the dimensions of the somatic and the social, and thus permits a differentiated analysis of the interrelation between the two. It will be argued that violent interaction uses the suffering body to stage a positional asymmetry, i.e. a distinction between strength and weakness, above and below, which might be socially exploited for the production or reproduction of social order. According to this triangular concept of violence the social impact of the somatic processes in violent interaction cannot be ascribed to the intensity of the latter; instead, the social consequences of violent interaction depend on the incident being observed and judged by a public. It will be shown that this perspective on violence permits us to bridge the gap between research on the micro- and meso-level dynamics of violent interaction on one hand, and theories of power, domination and the formation of social order on the other. 1. Violentia and Potestas: Violence in Interaction and Society In an article published in 1997, the sociologist Birgitta Nedelmann discusses the state of the art of violence research and summarises the challenges for consolidating a phenomenologically inspired research agenda. According to her, “new violence research” should develop a conceptual framework which allows the integration of classical sociological theories of social order and domination; it sociologically conceptualises the injuring of bodies and the experience of pain; and, finally, it analyses the subjectively intended meaning (in the Weberian sense) of violent action (Nedelmann 1997, 72–80). The ambition of the phenomenological research programme was hence not simply to reconstruct violent inter- from thinking and from developing “dispassionate” scientific concepts (Žižek 2009, 3). A counter-argument can be formulated in terms of hermeneutic thinking, which emphasises that the researcher and his or her emotions are always involved in social science research; in this regard, violence might be different in degree, but not in kind. 348 IJCV : Vol. 5 (2) 2011, pp. 345 – 356 Beck: Violence as a Social Process actions on a micro-level (as a reading of Sofsky’s works especially might suggest), but to bridge the gap between such micro-level analyses and processes of social structure formation at large. In a broader sense, the phenomenological research programme demanded an exploration of the interrelations between (a particular form of) interaction on the one hand, and the (re-)production of broader social structures on the other. Rephrasing the problem this way draws attention to the correspondence between the phenomenological agenda of violence research and schools of social thought which consider interactions as being decisive for the formation and reproduction of social structures. The constitutive interdependency between interactions and social structures is present in all socialconstructivist theories. It echoes in post-structuralist approaches such as Pierre Bourdieu’s theory of habitus and field (Bourdieu 1989) and Michel Foucault’s discourse theory (Foucault 1977), but also in Niklas Luhmann’s social systems theory (Luhmann 1984, 1995) or Anthony Giddens’s theory of structuration (Giddens 1984); it is particular pronounced in Symbolic Interactionism (Blumer 1977). So far, however, the possible insights to be drawn from a combination of social constructivist theory on the one hand, and a phenomenological approach to violence research on the other have not been systematically explored. In the case of violence, this interdependency between interaction and broader social structures seems to be reflected in the very etymology of the notion, as over the course of linguistic history the meaning oscillates between a transitive and an intransitive pole (Bowman 2001). While in its transitive sense “violence” denotes a relationship between a subject and an object in interaction, it indicates a property or potential of a subject in its intransitive meaning and thus refers to structural aspects. In both the Latin and the Anglo-Saxon languages, the transitive meaning prevailed: “[t]he exercise of physical force so as to inflict injury on, or cause damage to, persons or property; action or conduct 8 The obvious exception is auto-aggressive behaviour, such as suicide or self-mutilation, in which the subject and object of action coincide. characterised by this; treatment or usage tending to cause bodily injury or forcibly interfering with personal freedom” (OED, “violence”). This understanding of violence evokes an image of transgression, of an integral space that is broken in a situation of interaction.8 Historically this transitive meaning was preceded by an intransitive one. The Latin violentia from which “violence” is derived initially signified “vehemence”, “force” or “impetuosity”.9 Instead of referring to a relationship between a subject and an object, the notion of violence originally indicated a property or an enduring potential of a subject only (Bowman 2001, 25–26). Eventually this ambiguity was solved by bifurcation: violentia was reduced to the transitive meaning with which it is associated today, while the intransitive aspects were referred to the notion of potestas. Latin and Anglo-Saxon languages reflect this linguistic distinction between “power”/pouvoir/poder and “violence”/violence/violencia. The German expression Gewalt, by contrast, still echoes the ambiguity of the Latin root (Imbusch 2002, 28–29).10 Efforts to conceive a social theory of “violence” can benefit from this linguistic distinction between transitive and intransitive meaning. From the transitive usage of the word we can infer, firstly, that violence is a social process; the notion refers to a relationship involving a subject and an object. Secondly, the transitive usage suggests that the processes in question typically unfold in interactions and are thus related to the realm of the somatic, to the bodily aspects of human existence. The originally intransitive meaning of the word, however, reminds us, thirdly, that focussing on the violent act alone reduces our understanding of the phenomenon. Like a stone falling into water to create spreading ripples that may change the lines in the sand on a distant shore, violence transcends the moment of inflicted harm and comes to be inscribed into the structure of society itself. In section two I move on to investigate the transitive dimension of violence, reconstructing the social dynamics of violent interaction. Section three then ex- 9 The Latin root of violentia is vis, which means force or bodily strength. 10 The intransitive dimension of the notion is still present in expressions such as Staatsgewalt (“state power”). 349 IJCV : Vol. 5 (2) 2011, pp. 345 – 356 Beck: Violence as a Social Process plores the intransitive dimension, examining how the somatic processes which evolve in violent interaction are related to processes of social structure formation and dynamics of power and domination. 2. The Violent Moment: Three Modes of Experiencing Violent Interaction More than other forms of human action, violence has the potential to produce transformations of the social, because violence is characterised by a juxtaposition of social processes and the somatic aspects of human existence. Violence is a technique of making the body a site of social bargaining processes. Violent action targets the body upon which injuries are inflicted with pain being the result. Pain is always an existential human experience that fundamentally shakes self-awareness. It deprives the individual of the familiar instrumentality of the body and confronts him or her with their bodily existence.11 Moreover, it creates isolation since the experience of pain cannot be shared and can hardly be communicated. Prolonged states of pain therefore erode the sense of time and open up the gates to despair (Trotha 1997, 28–29). Experiencing pain inflicted by violent action is, however, particular. The target shares the loneliness, isolation and despair of all other pain sufferers. But one who suffers in violent interaction does so conscious that the pain is not the result of fate or hazard (as in the case of accidents or illnesses), but has been brought about intentionally by someone else. It is injury and pain inflicted deliberately to enforce the will of one against the resistance of the other (Trotha 1997, 31). Violence can be described as a social process whose function is to negotiate and reconfigure a relationship. Yet this process does not follow a random path; instead it is shaped 11 As Trutz von Trotha remarks in this regard, intense pain can change the experience of positionality in the sense described by Helmuth Plessner. The latter argues that the particularity of being human – in contrast to animals or plants – consists in being able to extend existence beyond the borders of the body, establishing artificial borders and embodying by the structural principle of asymmetry, by the attempt to create a positional difference between the actors involved (Baecker 1996, 99–100; Simon 2000, 109). Violence can therefore be described as a social technique that uses the body to mark and/or (re-)produce an asymmetric constellation in which the inferior position is associated with the experience of suffering, while the superior position is associated with the experience of inflicting suffering. Initially this difference is but situational; the positions of the one who suffers and the one who inflicts suffering can – in principle – be reversed. Yet, depending on the context and dynamics of the situation, the asymmetry might be perpetuated. In this case, violence produces subjects with complementary identities, which are commonly denoted as “perpetrator” and “victim”.12 These two notions are, however, linked to strong value judgments and emotions. Moreover, they evoke the idea of “perpetrator” and “victim” being definite roles of agents. As we will see in the following, contradicting a widespread assumption, the contingency or reversibility of positions is crucial for understanding the social dynamics of violence. Instead of speaking about “victims” and “perpetrators”, the notions of “target” and “performer” will be used here to refer to the two different positions. Rather than roles of agents, “target” and “performer” should be understood as different modes of experiencing violence.13 The “target” is associated with the damage to the body, with states of suffering or passivity, with feelings of fear and pain, with inferiority; it is where the existential character of violence derives from. The position of the “performer”, by contrast, is associated with intentional action aiming at damaging another body, with superiority and the exercise of power. Although these modes might, as mentioned above, lead to the formation of a particular subjectivity (as “victim” or a “perpetrator”, for example) and although the formation of such subjectivities might frequently be a principle moti- them (Plessner 1928). According to this argument, in experiencing pain man (temporarily) drops out of the species, approaching states of animal or even plant existence (Trotha 1997, 29). 12 In the following, quotation marks are used to indicate the non-ontological character of these expressions. 13 A similar idea can be found in Ivana Maček’s anthropological work on the war in Sarajevo (2001). She distinguishes between three “modes” of experiencing war, the “civilian-”, the “soldier-” and the “deserter mode”, understood as three ethically different ways of perceiving the war which introduce different choices of action and different legitimising narratives (Maček, 2001, 218-219). 350 IJCV : Vol. 5 (2) 2011, pp. 345 – 356 Beck: Violence as a Social Process vation for initiating violent interaction, such a development is far from preordained: not every brawl or battle produces the dyad of victimisation and empowerment. At first glance, the complementary modes of experience, “performer” and “target”, seem to comprise what is essential for analysing violence. They permit us to reconstruct the relational dynamics of the situation as well as its somatic aspects. Yet, considering violence as a social process, we have to move beyond the confrontation of “performer” and “target” in the violent act to include a third mode of experience: that of the “observer”. It is only by bringing in this last perspective that the social dynamics of violence can be truly understood. Violence is not simply physical harm intentionally inflicted by one person on another. It is not limited to the asymmetrical dyad of “performer” and a “target”. As a social process it evolves in a triangle where it is not only exercised and suffered, but also observed and judged. The idea of violence evolving in a triangular constellation appears in various approaches. Communicative aspects of violence, with an emphasis on the role of the public, are most prominently treated in the research on terrorism (Schmid and Graaf 1982; Waldmann 2005). With regard to armed group behaviour in civil wars Stathis Kalyvas argues in the same direction, pointing out that the main function of selective violence is to enforce “compliance” to armed rule among the (observing) population (Kalyvas 2006). Under the expression “bystander”, the role of the “observer” has been intensively discussed in genocide research (Vetlesen 2000; Barnett 1999; Grünfeld and Huijboom 2007; Hilberg 1995).14 The major difference between these works and the proposition advanced in this article can be best understood by examining the function of the “observer” in the respective concepts. The above-mentioned approaches conceive the 14 With regard to the individual, the “bystander effect” has been intensely researched in social psychology. Also known as the Genovese syndrome, the notion refers to the individual’s failure to help in an emergency situation. The latter, however, include accidents and similar situations as well as violent interactions(Latane and Darley 1970). “observer” as being crucial for the emergence and dynamics of particular forms of violence (such as terrorism, civil war or genocide); in stark contrast, the triangular concept proposed here introduces the observer as a constitutive third element in the social dynamics of violence.15 Another important difference is that, as pointed out above, “performer”, “target” and “observer” are not conceived as definite roles of particular agents; instead they are modes of experiencing violence, which, in a given violent interaction, might oscillate between different agents. Introducing the “observer” as a third constitutive element in a social theory of violence has a major implication: it permits us to analytically disentangle the dimensions of the somatic and the social. While experiencing violence as a “performer” or as a “target” means to be caught in the physical and emotional dynamics of a violent confrontation, the “observer” mode is in no sense predetermined. The “observer” breaks the hermetic circle of injuring and suffering, and is associated instead with consideration and decision. In this mode violence transcends the immediate confrontation and enters the realms of judgement, debate and memory. It is here that the shift from the transitive meaning of the notion as violentia to its intransitive meaning as potestas takes place. The most important consequence of this approach for empirical research is that “violence” is no longer conceived as an empirically evident phenomenon. At first glance, of course, the commonly assumed obviousness of violence stands to reason because of the somatic character of the processes in question: as violence deals with physical bodies, it is apparent to the senses and should easily be perceived. Taking a closer look, however, this commonsensical understanding is all but unambiguous: everyday experience is full of controversies about whether a certain 15 In his study on trust and violence in modernity, the literary scholar and social scientist Jan Philipp Reemtsma also argues for integrating the third party as a constitutive element in concepts of violence (Reemtsma 2008, 467–82). As a general idea, this thought can already be found in Riches’s Anthropology of Violence (1986). IJCV : Vol. 5 (2) 2011, pp. 345 – 356 Beck: Violence as a Social Process 351 interaction, for example spanking a child, barging into a crowded train, or forcing a kiss on a woman just met in a bar should or should not be considered violence. In spite of the obviousness of the somatic processes is question, the use of the notion of “violence” to describe them is all but undisputed. morrow’s “target”, and so forth. Usually, the agents’ knowledge about this interchangeability of positions is an important factor in the dynamics of violent interactions: where the fear of victimisation is driving the actions of “performers” and “observers”, violent situations quickly gain momentum. The theoretical reconstruction of violence as a triangular rather than dyadic dynamic systematises this contingency, emphasising that the concept of “violence” refers to a particular mode of observation more than to a certain type of interaction. It can be defined as a correlation between inflicting and suffering as observed by a third. Although this attribution might be more compelling in some instances than in others, it is contingent. What is considered violence depends on social norms and individual values, on criteria, thus, which lie in the eye of the beholder. In this sense the “observer” might become decisive for the evolution of a violent interaction. Her or his relative position to the “performer” and the “target” is crucial in defining the potential and limitations of a violent situation: an “observer” sympathising with the “target” potentially limits the options of the “performer”, whereas a timid or even applauding “observer” affirms and encourages the violent assault. Among the three modes of experiencing violence, the “observer” is the most inclusive. The reason for this is that post factum all agents involved (as far as they are still alive) become “observers” of the violent interaction as they remember, reflect, judge, decide. Violent action tends to generate its own public; even if not witnessed in the moment of its occurrence, the spoiled body itself comes to testify to violent action until long after the fact. The “observer” mode is, hence, not limited to witnesses of violence in actu. Instead, it relates to a variety of publics created – intentionally or not – by the violent act: the paralysed and frightened eyewitness, the “performer’s” cheering, goading peer group, global media stridently condemning the deeds, the researcher analysing the situation are all possible manifestations of the “observer” mode of experiencing violence. Considering the social dynamics of violence, the “observer” mode is, finally, the most striking indicator of the relational fragility of violent situations. The commonsensical understanding of violence, which mistakes “performer”, “target” and “observer” for definite roles of particular agents neglects that violent interaction frequently takes place in unsettled situations, in which positions can be reversed quickly: one moment’s “observer” might be next moment’s “performer”; today’s “performer” might be to- The “observer” is hence crucial when it comes to the social effects of the somatic processes evolving in violent interaction; or, to use the distinction introduced above, he or she is pivotal for understanding how the transitive and the intransitive dynamics of “violence” are intertwined. Having explored the former in this section, we will now turn to the latter and discuss the role of violence in the formation of social structures. 3. Beyond the Violent Moment: Violence and the (Re-)Production of Social Order According to the research agenda proposed by Nedelmann (see section one), so-called new violence research should not be limited to the theoretical reconstruction of violent interactions. Instead it should attempt to combine these considerations with theories of power, domination and the formation of social structures at large (Nedelmann 1997, 72–80). This demand was motivated by the prominent role of the notion of violence in classical theories of domination: in Max Weber’s thinking, violence, or rather the credibly institutionalised threat thereof, is conceived as the basis of domination. This very idea is already to be found in the political philosophy of state formation of Jean Bodin (1606) and Thomas Hobbes (1992 [1651]). In this perspective, social order is dependent upon the successful monopolisation of the potential for violent action (Weber 1978, 54). These works thus focus on what had once been the intransitive dimension of the notion of violence. IJCV : Vol. 5 (2) 2011, pp. 345 – 356 Beck: Violence as a Social Process The triangular reconstruction of violence proposed in the preceding section permits us to reframe the propositions of a classical Weberian sociology of domination, linking the intransitive meaning of violence as potestas to the interaction processes associated with the transitive meaning as violentia. A first step in this direction was undertaken by the German sociologist Heinrich Popitz who was a major inspiration to the violence research renewal movement in the 1990s. In Phänomene der Macht (Phenomena of Power) (1986) Popitz sets out to refine the role of violence in a Weberian sociology.16 Introducing the concept of action into Weber’s theoretical framework he proposes conceiving violence as the most direct form of power, as power in action, or, as he put it, “pure action power” (schiere Aktionsmacht) (Popitz 1986, 68, my translation), which is rooted in the general vulnerability (Verletzungsoffenheit) of man (69). On the basis of this discussion of phenomenological aspects of violent interaction, Popitz insists on the systematic role of violence in the formation of broader social structures: “Violence in general and the violence of killing in particular is not just an accident of social relations, not a side issue of social order and not just an extreme case or ultima ratio (about which not much fuss can be made). Violence is actually … an ever-present option of human action. No comprehensive social order is based on the assumption of non-violence. The power to kill and the powerlessness of the victim are latent or manifest determinants of the structure of social coexistence” (translated from Popitz 1986, 83). Unlike Weber, whose writings discuss the monopolisation of violence in terms of a technical problem arising out of the process of state formation, Popitz’s phenomenological work traces how the natural presence of the potential of violence, combined with universal knowledge about the consequences of violence acted out, has an ordering effect on society. He links the structures of society to the dynamics of violent interaction and, in doing so, calls attention to the fact that violence, in the guise of contingency, is also inscribed into social structures characterised by the absence of violent interaction. The German sociologist Dirk Baecker went on to demonstrate that the latter holds true not only for processes 16 Unfortunately, the book has not been translated into English. 352 of socialisation at large in the sense proposed by Bodin, Hobbes or Weber, but also on the meso- and micro-level (Baecker 1996, 94–95). The triangular concept of violence permits us to theoretically refine this relationship between violent interaction and the formation of social structures. As elaborated above, violent interaction follows the structural principle of asymmetry, attempting to stage a positional difference by means of the suffering body, a contrast between strength and weakness, above and below, superiority and inferiority. If the perception of this positional difference can be perpetuated, the latter might be used for the establishment or reproduction of a social order based on domination and subordination. In this case, violent interaction produces particular asymmetric and complementary subjectivities on the side of the “performer” as well as the “target”. In this regard, violence is functionally equivalent to other social techniques of asymmetrisation, such as defamation or derision. The latter, too, stage and/or (re-)produce an asymmetry that is exploitable in a relationship of power. The particularity of violence, however, stems from the fact that the asymmetry is produced by threatening not only the social or mental integrity of the agent, but also the physical conditions of her or his existence. Given the general vulnerability of the human body, the performance of violence needs neither sophisticated equipment nor specialised knowledge to be effective. Violence is, in the words of von Trotha, an “everybody‘s resource” (translated from 1997, 25). Moreover, the somatic character of violence reduces the ambiguity inherent in any communication. Violence is apparent to the senses; and as the memory of violent acts is embodied in wounds or scars, the sensuosity of violence transcends the moment of violent action itself. Violence is easily accessible, easily perceivable and easily understood (Riches 1986, 11) and therefore reduces the contingencies inherent in any communication. Baecker speaks of violence as “deoptionalised communication”, which forces particular attributions and dramatically narrows the range of possible ensuing communications (1996, 101). IJCV : Vol. 5 (2) 2011, pp. 345 – 356 Beck: Violence as a Social Process More than other social techniques of asymmetrisation, violence therefore bears the potential to perpetuate the perception of a positional difference produced in a confrontation and turn it into the social basis of domination building on asymmetric and complementary subjectivities. As any social order is based on the production and reproduction of differences, violence is, hence, a potent instrument for establishing social order and transforming the social environment. The same argument could, of course, be conceived in the framework of a dyadic theory of violence. Yet, the triangular reconstruction of violence proposed here implies one major dissimilarity. Introducing the “observer” as a third constitutive element draws attention to the contingency of the processes in question: in a dyadic concept the social dynamics of violence are conceived as unfolding only in the confrontation between a “target” and a “performer”. The problem of such a representation is that it risks confusing the social and somatic dimensions of the processes in question. The dyadic reconstruction of violence suggests a correspondence between the somatic and the social in the sense that the social impact of “violence” is assumed to be a function of the intensity of the somatic occurrences. Yet the everyday debates mentioned above, which question the adequacy of the notion of “violence” for spanking a child or forcing a kiss , already illustrate the shakiness of this assumption. The triangular reconstruction, by contrast, implies that the social effects of the dyadic, somatic events do depend not on the severity of the latter, but on the perception of the processes in question by a third party. It suggests that, socially, the question of whether or not certain occurrences are “violence” can never be decided by any objective criteria, but depend on the perspective, i.e. the norms, values and objectives, of an “observer”. One major implication of this approach is that it permits us to conceive not only the presence of violence as observed by a third party, but also the absence of this observation. Rigorously conceiving violence as socially constructed implies that different “observers” might judge the same proceedings differently. And it suggests that the formation of such an observation is open to manipulation. The ap- 353 proach therefore draws attention to processes and techniques which either deliberately stage “violence” or attempt to invisibilise it. Contradicting the commonsensical assumption that violence is a process that is by definition initiated by a “performer” causing injury, the triangular reconstruction implies that “violence” might also be enacted on the side of the “target” by staging suffering – independently of the actual intensity of the somatic events in question and even independently of the actual intention of the perceived “performer”. As long as the somatic intensity is low, the social dynamics of “violence” depend chiefly on the victims propensity to display suffering– or not to. In this sense, even a kiss might come to be observed as violence. As pointed out above, the triangular reconstruction draws attention not only to processes of staging violence, but also those of invisibilising it. What might appear as violence to a critical observer can be discursively reframed so as to conceal the coercive character of the measures in question: “torture” or “enhanced interrogation methods”, “forced sterilisation” or “prevention of hereditary disease in offspring”, “massacre” or “mass execution”, “genital mutilation” or “ritual circumcision” – the list of such alternative descriptions of identical occurrences could be extended ad lib. In each of the conceptual pairs, the first description spotlights the coercive character of the action of the performer and the suffering of the target intentionally provoked by it; the second variant, by contrast, clouds these aspects by emphasising the legitimacy of the action deriving from its role in the production or reproduction of social order. Techniques and processes of socially staging and invisibilising “violence” ultimately point to the problem of legitimacy that violence can never escape. Many scholars have pointed out that violent interaction is always a contestable social act, which has to be justified (Riches 1986, 5–8; Schlichte 2009, 85–115). Not by chance, Weber defines the state as the possessor of the monopoly of legitimate violence on a given territory (1978, 54). In a triangular reconstruction of violence, the interrelation between violence and legitimacy can be further refined: 354 IJCV : Vol. 5 (2) 2011, pp. 345 – 356 Beck: Violence as a Social Process rigorously conceiving violence as socially constructed implies that different “observers” might judge the same proceedings differently, which ultimately draws attention to the manipulability of the observation. With regard to the problem of legitimacy, this manipulability can be exploited by the “performer” as well as by the “target”: to prevent delegitimisation “performers” can not only adjust violent action according to the norms and values of relevant “observers”, but also attempt to discursively reframe actions which risk being observed as “violence”. Conversely, delegitimisation can be introduced on the side of the “target” by performing suffering and thus staging violence. These dynamics of legitimisation and delegitimisation through violence are particularly important when it comes to the establishment and reproduction of a monopoly of power. As pointed out by theories of state-building from Hobbes to Tilly, violence plays a central role in the establishment of social order on the large scale (Hobbes 1992; Tilly 1975; 1985). In the medium and long term this central role of violence in the emergence and reproduction of social order risks jeopardising the legitimacy of the latter. Therefore, the monopolisation of the potential for violence is systematically combined with a reinterpretation of the coercive actions of the monopolising power. Yet, as the coercive action persists, the discursive manipulation remains open to contestation: delegitimising the powerful by drawing attention to the suffering caused by coercive actions, by speaking of “torture” and “massacres” instead of “enhanced interrogation methods” and “mass executions” is therefore one of the most threatening instruments in the hands of the less powerful, especially in contexts that cultivate the ideal of non-violence. 4. Implications for Empirical Research The key implication of the triangular concept of violence for empirical research is the shift from conceiving “violence” as being empirically evident to conceiving it as being socially constructed. Accordingly, empirical research has to investigate how this construction takes place. In the triangular framework proposed here, the interdependencies between the somatic dynamics on the one hand, and their social effects on the other, will be crucial. Attention, hence, turns not only to processes in which asymmetries are staged on a somatic level, but simultaneously to the question of how these processes are observed, described and judged by an idealtypical third party. Accordingly, phenomena of violence can be differentiated in terms of two characteristics: (a) the somatic intensity of the processes in question, i.e. the intensity of the bodily transgression, which might be high or low; and (b) the social observability thereof, which, again, might be high or low. The somatic intensity of the events is crucial in determining the scope of possible attributions and interpretations; while in cases of low somatic intensity it might be possible to deny (or stage) the intention to inflict suffering, cases of high somatic intensity narrow the range of possible interpretations. The somatic intensity also affects the agents’ capacity to choose action. The social observability of the same events, by contrast, is critical in determining in how far a particular occurrence can become a subject of debate at all; therefore, the social impact of any struggle over interpretation will be particularly pronounced in cases where the events in question can be widely observed. Table 1 summarises the possible combined expressions of the two characteristics. Table 1: Social observability and somatic intensity Low somatic intensity High somatic intensity Low social observability High social observability 1 3 2 4 In each of the four idealtypical cases the social production (or obscuration) of violence takes place under different conditions and the options of the “performer”, the “target” and the “observer” vary accordingly. In empirical research, the distinction between the somatic and the social dimension might, hence, serve to differentiate dynamics of violence as well as to analyse processes of transition from one form to another. To further develop this approach, research in two directions is needed. First, the figure of the “observer” has to be conceptually refined, in particular in view of its possible empirical manifestations and the related social functions. Empirical and theoretical work is necessary to fulfil this goal. Empirical research can draw on insights from those IJCV : Vol. 5 (2) 2011, pp. 345 – 356 Beck: Violence as a Social Process areas of violence research where the “observer” or the “public” already plays a central role, such as genocide or terrorism research (cf. section two). Starting from there, the social “production” of “targets” and “perpetrators”, of “victims” and “performers” should be explored in a general perspective, investigating diverse settings of organised and unorganised violence. Theoretically, the differentiated elaboration of the “observer” can benefit from a number of recent pieces of research on the conceptual figure of the “third party”.17 Based on this refinement of the third party, a second challenge for further developing this approach can be met: to explore and to conceptualise the interplay between the social and the somatic in the social production of violence. A central question in this regard is in how far particular somatic dynamics limit the possible range of communicative responses and interpretations and how these limitations are dealt with socially. 5. Conclusion Drawing on the phenomenological critique of violence research, to develop a theoretical concept of violence in the framework of social-constructivism, violence can be conceived as a triangular dynamic evolving between a “performer”, a “target” and an “observer”. The latter is pivotal in establishing meaning and judgement, in introducing the political dimension of violent interaction. This analytical perspective permits us to integrate the instrumental and the expressive dimension of violent interactions; it allows us to conceive violence as an act as well as an image, or, to put it differently, as an instrument efficiently serving practical as well as symbolic needs (Riches 1986, 11, 13). Violent action is thus not simply a means to pursue particular ends, but first and foremost a way to create, stage or change asymmetric relationships (cf. Simon 2000, 108–109). 17 See for example Boltanski (1999), Fischer (2006), Werron (2010). The sociologist Gesa Lindemann even argues for a triangular conception of “sociality” in general (2006). 355 Including the “observer” as a third constitutive element in a social theory of violence permits us to conceptually bridge the gap between analyses of violent interaction on the one hand, and discussions of violence in the sense of potestas on the other. This theoretical framework allows to systematically link research on the empirical dynamics of violent interaction with theories of social order, power and domination. IJCV : Vol. 5 (2) 2011, pp. 345 – 356 Beck: Violence as a Social Process References Baecker, D. 1996. Gewalt im System. Soziale Welt 47: 91–109. Barnett, V. J. 1999. Bystanders Conscience and Complicity during the Holocaust. Westport, CT et al.: Greenwood. Blumer, H. 1977. Symbolic Interactionism: Perspective and Method. Englewood Cliffs, NJ: Prentice Hall. Bodin, J. 1606. The Six Bookes of a Common-weale. London: Impensis G. Bishop. Boltanski, L. 1999. Distant Suffering. Morality, Media and Politics. Cambridge: Cambridge University Press. Bourdieu, P. 1989. Outline of a Theory of Practice. Cambridge et al.: Cambridge University Press. Bowman, G. 2001. The Violence in Identity. In Anthropology of Violence and Conflict, ed. B. E. Schmidt and I. W.Schröder, 25–46. London and New York: Routledge. Eckert, J., ed. 2004. Anthropologie der Konflikte: Georg Elwerts konflikttheoretische Thesen in der Diskussion. Bielefeld: Transcript. Elwert, G. 1997. Gewaltmärkte: Beobachtungen zur zweckrationalität von Gewalt. In Soziologie der Gewalt, ed, T. von Trotha, 86–101. Opladen: Westdeutscher Verlag. Elwert, G. 1999. Markets of Violence. In Dynamics of Violence: Processes of Escalation and De-escalation in Violent Group Conflicts, ed. G. Elwert, S. Feuchtwang, and D. Neubert, 85–102. Berlin: Duncker & Humblot. Fischer, J. 2006. Der Dritte/Tertiärität: Zu einer Innovation in den Kultur- und Sozialwissenschaften. In Philosophische Anthropologie im 21. Jahrhundert, ed. H.-P. Krüger and G. Lindemann, 146–63. Berlin: Akademie Verlag. Foucault, M. 1977. Discipline and Punish. London: Allen Lane. Giddens, A. 1984. The Constitution of Society: Outline of the Theory of Structuration. Cambridge: Polity. Grünfeld, F., and A. Huijboom. 2007. The Failure to Prevent Genocide in Rwanda: The Role of Bystanders. Leiden et al.: Nijhoff. Hilberg, R. 1995. Perpetrators, Victims, Bystanders: The Jewish Catastrophe 1933–1945. London: Secker & Warburg. Hobbes, T. 1992 [1651]. Leviathan. Cambridge et al.: Cambridge University Press. Hüsken, T. 2004. Georg Elwert und die Berliner Schule der skeptischen Sozialanthropologie. In Anthropologische Perspektiven auf Konflikt, ed. J. Eckert, 315–30. Bielefeld: Transcript. Imbusch, P. 2002. Der Gewaltbegriff. In Internationales Handbuch der Gewaltforschung, ed. W. Heitmeyer and J. Hagan, 26–57. Wiesbaden: Westdeutscher Verlag. Kaldor, M. 1999. New and Old Wars: Organized Violence in a Global Era. Stanford, CA: Stanford University Press. Kalyvas, S. N. 2006. The Logic of Violence in Civil War. Cambridge and New York: Cambridge University Press. Latane, B., and J. M. Darley. 1970. The Unresponsive Bystander: Why Doesn’t He Help? Englewood Cliffs, NJ: Prentice Hall. Lindemann, G. 2006. Die dritte Person: Das konstitutive Minimum einer Sozialtheorie. In Philosophische Anthropologie im 21. Jahrhundert, ed. H.-P. Krüger and G. Lindemann, 125–45. Berlin: Akademie Verlag. Teresa Koloma Beck [email protected] 356 Luhmann, N. 1984. Soziale Systeme: Grundriss einer allgemeinen Theorie. Frankfurt am Main: Suhrkamp. Luhmann, N. 1995. Social Systems. Stanford, CA: Stanford University Press. Maček, I. 2001. Predicament of War: Sarajevo Experiences and Ethics of War. In Anthropology of Violence and Conflict, ed. B. E. Schmidt and I. W. Schröder, 197–224. London and New York: Routledge. Münkler, H. 2002. Die neuen Kriege. Reinbek bei Hamburg: Rowohlt. Münkler, H. 2005. The New Wars. Cambridge et. al.: Polity. Nedelmann, B. 1997. Gewaltsoziologie am Scheideweg: Die Auseinandersetzung in der gegenwärtingen und Wege der künftigen Gewaltforschung. In Soziologie der Gewalt, T. von Trotha, 59–85. Opladen: Westdeutscher Verlag. OED. Oxford English Dictionary. Online edition. http://www.oed.com. Plessner, H. 1928. Die Stufen des Organischen und der Mensch: Einleitung in die philosophische Anthropologie. Berlin: de Gruyter. Popitz, H. 1986. Phänomene der Macht: Autorität: Herrschaft, Gewalt, Technik. Tübingen: Mohr. Reemtsma, J. P. 2008. Vertrauen und Gewalt: Versuch über eine besondere Konstellation der Moderne. Hamburg: Hamburger Edition. Riches, D. 1986. The Phenomenon of Violence. In The Anthropology of Violence, ed. D. Riches, 1–27. Oxford and New York: Blackwell. Schlichte, K. 2009. In the Shadow of Violence. The Micropolitics of Armed Groups. Frankfurt am Main: Campus. Schmid, A. P., and J. de Graaf. 1982. Violence as Communication. Insurgent Terrorism and the Western News Media. London and Beverly Hills: Sage. Simon, F. B. 2000. Krieg: Systemtheoretische Überlegungen zur Entstehung tödlicher Konflikte. Familiendynamik 25:104–30. Sofsky, W. 1993. Die Ordnung des Terrors: Das Konzentrationslager. Frankfurt am Main: Fischer. Sofsky, W. 1996. Traktat über die Gewalt. Frankfurt am Main: Fischer. Sofsky, W. 1997. The Order of Terror. The Concentration Camp. Princeton, NJ: Princeton University Press. Sofsky, W. 2003. Violence: Terrorism, Genocide, War. London: Granta. Tilly, C. 1975. The Formation of National States in Western Europe. Princeton, NJ: Princeton University Press. Tilly, C. 1985. War Making and State Making as Organized Crime. In Bringing the State Back In, ed. P. Evans, D. Rueschemeyer, and T. Skocpol, 169–87. Cambridge: Cambridge University Press. Trotha, T. von. 1997. Soziologie der Gewalt. In Soziologie der Gewalt, ed. T. von Trotha, 9–56. Opladen: Westdeutscher Verlag. Vetlesen, A. J. 2000. Genocide: A Case for the Responsibility of the Bystander. Journal of Peace Research 37:519–32. Waldmann, P. 2005. Terrorismus: Provokation der Macht. Hamburg: Murmann. Weber, M. 1978. Economy and Society: An Outline Of Interpretive Sociology. Berkeley, Los Angeles, and London: University of California Press. Werron, T. 2010. Direkte Konflikte, indirekte Konkurrenzen: Unterscheidung und Vergleich zweier Formen des Kampfes. Zeitschrift für Soziologie 39:302–18. Žižek, S. 2009. Violence: Six Sideways Reflections. London: Profile.