Violence and publicity: constructions of political responsibility after 9/11

Abstract

Focussing upon recent political philosophical reflections on the War on Terror, this paper asks whether violence can be understood without undermining the empirical and normative potential of public action to curtail it. Explanations of political violence are often either dismissed as mere exoneration, or end up reducing all forms of power to violence. Negotiating a path between these poles requires developing a double sense of responsibility that affirms that while violence can be comprehended as a rational phenomenon this does not in principle undermine the potential of public action to democratize violence. Theoretical approaches that ontologize violence as an ineradicable feature of ‘the political’ are criticized for leaving little space for thinking about the possibility of legitimate public action in curtailing political violence. The public dimensions of the relationships between justification, legitimacy and political violence are considered by relating contemporary left‐liberal arguments about terrorism and legitimate violence to recent feminist conceptualizations of the practice of taking responsibility.

Keywords:

• violence
• publicity
• responsibility
• justification
• legitimation

Explaining, excusing, and exculpating

Recent political philosophical reflections on the War on Terror address a long‐standing interdiction on treating violence as a normal feature of political processes. This paper considers the question of whether violence can be made intelligible as a feature of human affairs without undermining the empirical and normative potential of public action to curtail violence.

Social scientists suggest that ‘terror is a recurrent political strategy adopted by a wide variety of actors rather than a creed, a separate variety of politics, or the work of a distinctive class of people’ (Tilly 2003, p. xii). On this relational view of political violence as a form of interaction and claims‐making, terrorist violence is best understood as one aspect of a broader repertoire of contentious politics. However, analysing terrorist violence from this perspective challenges the emphasis on identifying culpable actors who can be held responsible for acts of violence found in ‘blame‐centred theories’ of political violence (Mann 2005, p. 8). For this reason, social scientific efforts at explaining and contextualizing ‘terrorism’ often provoke a vitriolic response, and not only from right‐wing conservatives. For example, the Euston Manifesto published in London 2006 by left‐liberal intellectuals supportive of the US–British‐led invasion of Iraq, sets itself firmly against what it considers the irresponsible yearning to ‘understand’ terrorism and terrorists:

We decline to make excuses for, to indulgently ‘understand’, reactionary regimes and movements for which democracy is a hated enemy – regimes that oppress their own peoples and movements that aspire to do so. We draw a firm line between ourselves and those left‐liberal voices today quick to offer an apologetic explanation for such political forces. ¹

Likewise, in the United States, so‐called ‘Just‐War liberals’ such as Michael Walzer (2002) and Jean Bethke Elshtain (2003b) have interpreted the attempt to account for the milieu out of which political violence emerges as nothing more than an ‘exculpatory strategy’ that reflects the ‘automatic oppositionism’ of the academic left (Elshtain 2003b, p. 8).

Both Walzer and Elshtain analyze terrorism under the paradigm of Just War theory. They privilege a definition of terrorism as the intentional targeting of innocent civilians. Understood in this way, terrorism always contravenes principles of justification and legitimacy. By contrast, state‐led military intervention is ascribed a presumption of legitimacy, and not necessarily only on the grounds of self‐defence (Elshtain 2003a).

The targeting‐based definition of terrorism has been challenged on the grounds that it forecloses a consideration of the intentional structure that is distinctive of terrorist action (see Goodin 2006, Waldron 2004). For example, Virginia Held (2004) challenges definitions of terrorism that prioritize the deliberate killing of innocents. This type of definition holds ‘that terrorism is so morally unjustifiable as a means that we do not even have to consider the political objectives of those who engage in terrorism’ (Held 2004, p. 62). Building targeting of innocent civilians into definitions has the effect of automatically establishing terrorist violence as unjustified, while implying that state‐sanctioned counter‐terrorism is justified self‐defence (Held 2004, p. 65).

An alternative definition of terrorism emphasizes the sense of terrorist violence as a type of public action. Held proposes that terrorism be understood as action that seeks ‘to spread fear among a wider group than those directly harmed or killed’ (Held 2004, p. 63), and as ‘political violence that usually spreads fear beyond those attacked, as others recognize themselves as potential targets’ (p. 68). The importance of restoring the sense of terrorism as a form of public action lies in relocating this type of political violence within an evaluative field in which questions of justification remain open:

Instead of accepting a definition of terrorism that sees it as necessarily targeting civilians and hence always and inevitably unjustifiable, we should consider the aims of terrorists and of those who use violence to thwart those aims. (Held 2005, p. 180)

Insisting that terrorism can and should be evaluated with reference to the purposes of its perpetrators draws into focus the importance of public deliberation as a medium in which to respond to this form of political action. One feature of post‐9/11 political debate has been the argument that the balance between liberty and security needs to be recalibrated – that individual rights need to sacrificed in order to secure the protection of all. The argument that liberty and security can be traded off against each other draws on apocalyptic scenarios of ‘ticking bombs’ and imminent attacks by rogue WMDs to generate a generalised sense of urgency which justifies the curtailment of deliberative reflection (see Waldron 2007). In a public sphere dominated by such horrific scenarios, Ronald Dworkin (2006, pp. 49–51) observes that the only virtue expected of citizens and governments alike is one of ‘terrified prudence’.

The closing down of deliberation reflects what Iris Marion Young (2007) argued is the development of a security state in the United States after 9/11. This development has depended on a particular framing of the relationship between states and citizens, security and liberty. She calls this framing ‘the logic of masculinist protection’: the state that wages war abroad in the name of securing citizens’ safety at home expects obedience from its citizens, in a manner analogous to a patriarchal model of a male protector of the domestic space inhabited by vulnerable women and children. This logic mobilises an affective repertoire to present the state as a virtuous and responsible guardian. In this logic, citizens are primarily positioned as dependent subordinates:

Central to the logic of masculinist protection is the subordinate relation of those in the protected position. In return for male protection, the woman concedes critical distance from decision‐making autonomy. (Young 2007, p. 120)

Young sees the logic of masculinist protection as a threat to the practices of participatory self‐determination that distinguish democratic self‐government.

It can be argued that a heightened sense of risk and fear requires more deliberation, not less. This argument follows from understanding security not as an individualized good, but as ‘a collective good that makes possible an environment in which we can live and deal with one another’ (Waldron 2007, p. 44). On this understanding, curtailing individual liberties threatens to undermine the capacity to discern and act upon risks in appropriate ways. If one prioritizes the self‐governing capacities of citizens to identify and reflect upon appropriate courses of action in relation to risks and fear, then social scientific procedures of explanation should be ascribed a central place in public culture.

In contrast to the denunciation of explanation as tantamount to exculpation, there are a range of philosophical responses to terrorist violence and counter‐terrorist strategies that affirm two related principles of analysis: firstly, that considering the purposes and intentions of the deployment of violence is crucial to understanding terrorism; and secondly, that it is perfectly possible to distinguish causal explanations from normative evaluations. Nevertheless, maintaining that deliberation, explanation, and justification are necessary responses to violence is difficult in a climate in which such a task is routinely condemned. Judith Butler suggests that the effect of loud invocations of ‘terrorist threats’ is to drown out the kind of thinking that is required by any responsible analysis of political violence:

To condemn the violence and to ask how it came about are surely two separate questions, but they need to be posed in tandem, held in juxtaposition, reconciled within a broader analysis. Under contemporary strictures on public discourse, however, this kind of dual thinking cannot be heard. (Butler 2002, p. 186)

Butler’s identification of the need for a style of dual thinking provokes the thought that ‘what can be heard’ in the public sphere is not only determined by the master‐frames of official discourse. It might also have something to do with the tone in which explanations are articulated. Martin Shaw (2002) has suggested that too often the response of the left to 9/11 and subsequent events has displayed a troubling tendency to provide ‘misplaced comparisons’ and ‘fallacious contextualization’ as substitutes for rather than accompaniments to the task of formulating an appropriate moral response to events of human suffering. John Brenkman (2007, p. 80) has also identified as a difficulty for the left the ‘inability to express terror and pity in the same breath as dissent and criticism’.

The point is not that comparison and contextualization are misplaced per se. It is that they need to be articulated alongside acknowledgements of the weight of particular harms and traumas suffered through the acts of violence. Shaw raises the problem of what the relationship between providing an explanation and formulating a response to atrocious acts of political violence should be (see Shaw, no date). This distinction between explaining and responding requires finding a way of thinking about two overlapping imperatives of responsibility: one that tries to understand the conditions and causes of violence; and one which endeavours to bear witness to the suffering, trauma, and harm that is the enduring consequence of violence. In the same spirit, Jacques Derrida insisted in the wake of 9/11 that

one can describe, comprehend, explain a certain chain of events or series of associations that lead to ‘war’ or to ‘terrorism’ without justifying them in the least, while in fact condemning them and attempting to invent other associations. One can condemn unconditionally certain acts of terrorism (whether of the state or not) without having to ignore what might have bought them about or even legitimated them. (Borradori 2003, pp. 106–107)

The double task of judging and explaining that Derrida insists upon, as separate but equally compelling imperatives, requires us to think of the task of responsibility as similarly double: as involving both an imperative to explain and an imperative to formulate a response.

The next section considers the arguments of Giorgio Agamben, which provide one of the most compelling and influential philosophical analyses of the significance of the War on Terror. Agamben’s work is an example of political theory which ontologizes violence as a foundational feature of human affairs. It is argued that this style of analysis forecloses the possibility of understanding the double responsibility outlined above. The ontologization of violence elides the publicity conditions through which the political deployment of terrorist and counter‐terrorist violence works, and drastically narrows the grounds for criticizing these practices.

Ontologies of violence

One way of making violence intelligible as a feature of human affairs involves ascribing it a primal status as the dynamic of life in general. This argument is associated with an ontological claim that all political action is rooted in violence (see Hanssen 2000). Giorgio Agamben’s (2005) account of the ‘the state of exception’ provides one of the most resonant resources for critics seeking a philosophical vocabulary in which to locate the War on Terror. Agamben builds on Carl Schmitt’s derivation of a theory of the state and of ‘the political’ from a theory of law, and on Schmitt’s view that all law is constituted in the concrete situation of application. For Schmitt, relations of enmity and hostility are the very essence of politics. Schmitt held that all legal norms rest ultimately on the arbitrary ‘decision’ of a sovereign, who either applies laws universally, or exercises the arbitrary power to pronounce exceptions to the law. Schmitt’s ideas on the political (2001) and on sovereignty (2006) have been revived as a way of critiquing the relationships between liberal constitutionalism and the internal and external deployment of violence by democratic states (see Scheuerman 2006a, 2006b). Agamben uses Schmitt to help in developing an argument that the ultimate foundation of political authority is the asserted will of a dominating sovereign actor, able to exert control over the vitality of life itself by exercising the power of death over excluded others (cf. Norris 2003).

Agamben contributes to the revival of interest in decisionistic interpretations of ‘the political’. Decisionism refers to the supposition that the absence of foundationalist grounds of legitimate validity in knowledge, law and politics means that all power is essentially arbitrary – the imposition of force through the assertion of individual or collective will. ² The decisionistic interpretation of ‘the political’ raises the question of whether violence can be rationalized through public mediums of legitimacy and justification. On a strongly ontological interpretation that naturalises violence, violence and legitimacy cannot be squared at all (see Frazer and Hutchins 2007).

Agamben’s account of political power seems to provide a particularly apposite vocabulary for analysing a geopolitics centred on assertions like George W. Bush’s ‘I’m the decider’ and the neoconservative doctrine of unitary executive authority. The Bush administration has even been described as crypto‐Schmittian (Critchley 2007, pp. 133–150). But this apparent decisionistic quality of contemporary state policies should cause us to stop and wonder whether a Schmittian‐inflected analysis of the War on Terror does not just mirror back the normative self‐understanding of neoconservative ideologues. The ascendancy of Agamben’s account of the state of exception is an example of the peculiar convergence of left and right around Schmitt’s account of modern power.

Agamben’s account of the state of exception builds on an already well‐established political interpretation of Derrida’s account of the undecidable foundations of the law and justice. Derrida’s (2002) account of the ‘aporia’ between law and justice considers Walter Benjamin’s messianic reflections on the constitutive relationship between violence and law. Benjamin’s (1978) ‘Critique of violence’ holds that violence always begets the law, and castigates political compromise as unmanly and unworthy of the generative force of violence in human affairs. It invokes an eschatological ‘pure’ violence as the only force able to resolve human conflicts. Benjamin’s essay stands in close, and controversial, proximity to Schmitt’s decisionistic account of legality, sovereignty, and the political (Weber 1992).

Derrida’s ‘Force of law’ (2002) re‐reads Benjamin through the prism of his own distinctive understanding of J.L. Austin’s analysis of the force of performative utterances. Derrida held that Austin’s account of the force of performatives should be understood as ‘a thought that is nothing less than Nietzschean’ (Derrida 1982, p. 322). There is a recurrent tendency in Derrida’s allusions to Austin to run together force, power and violence. This rendering leads to the claim that the foundation of law is a performative act in the sense that it involves a coup de force, involving ‘performative and therefore interpretative violence’ (Derrida 2002, p. 241). Derrida’s reading undoes any clear‐cut distinction between law as enforcement and the constitutive force of founding law that Benjamin had established.

This deconstructive reading of the concept of ‘force’ in ordinary language philosophy is picked‐up in Agamben’s account of the state of exception. Agamben (2005, pp. 37–39) reiterates Derrida’s ‘Force of law’, but crosses through the term ‘Law’. The effect is to imply that the lesson of Derrida’s reading of the performative force of legal foundations is that law ultimately has no validity at all. Agamben argues that the ‘mystical’ element that Derrida addresses in ‘Force of law’ – the relationship between legal norm and application, between potentiality and act – still requires conceptual clarification. The clarification which Agamben provides simply confirms a Schmittian interpretation of sovereignty as the arbitrary authority to assert and impose exceptions to the law.

Agamben asserts that the history of the law of emergency, concerned with the conditions under which the rule of law can be suspended in order to guarantee its own existence, reveals the general condition that governs the relation of law to reality. Following Schmitt faithfully, Agamben supposes that juridical orders are characterised by ‘an essential fracture between the position of the norm and its application’ (Agamben 2005, p. 31). The very possibility of application is revealed by the powers invoked in emergency situations (pp. 85–86): application is reduced to a Schmittian interpretation of sovereign decision, understood as a wilful imposition of force that secures its own conditions of success.

Agamben transposes this decisionist interpretation of legality into a universal theory of political power and the state. The Schmittian hypothesis about sovereignty is bolstered by the appropriation of Foucault’s notions of biopolitics and biopower. Sovereignty, once revealed as the essential form of all political power, is understood to cut through normative niceties by exerting violent force directly over the vital conditions of life and death itself. The analysis of the state of exception builds on and confirms Agamben’s analysis of bare life in Homo Sacer (1998), in which all community is understood to be constituted through the violent extrusion of others into spatialized zones of exclusion. Agamben claims that all power rests on the ability to take life, so that power is always grounded on the possibility of enforcing death over subjects stripped of all status apart from that of ‘bare life’ itself. In a drastic reduction and reversal of Foucault’s ideas, Agamben equates biopolitics with the suspension of law in the exercise of sovereignty. Biopolitics is understood as referring to a mode of domination, expressed in the exercise of the power to take life. And this is presented as the paradigm for all political power.

Agamben’s account of the state of exception does not seek to mobilise a philosophical vocabulary to analyse contemporary geopolitics. It selectively interprets the contemporary conjuncture to confirm a sweeping philosophical claim about the nature of political power in general, wherein the ultimate grounding of all state power lies in wilful and violent assertions of sovereign authority. Post‐9/11 security measures, such as the internment of prisoners at Guantanamo Bay or the US Patriot Act, are presented as stylized facts which confirm ‘[t]he immediately biopolitical significance of the state of exception as the original structure in which law encompasses living beings by means of its own suspension’ (Agamben 2005, p. 3). The only thing that this situation can be compared to, he claims, is the legal situation of the Jews in Nazi camps (p. 4). This assertion conveniently confirms Agamben’s claim that ‘the camp’, on the model of Auschwitz, reveals the logic of modern power in general. The contemporary example of the War on Terror reveals the paradigmatic status of the sovereign assertion of exceptions as the ground of all power. On this interpretation, the state of exception ‘not only appears increasingly as a technique of government rather than an exceptional measure, but it also lets its own nature as the constitutive paradigm of the juridical order come to light’ (pp. 6–7).

Agamben’s view that the exercise of power over life and death is the essence of the modern state has become a basic reference point for a revival of interest in the topic of sovereignty in the social sciences and humanities. On this view, any rationalization of violence through law is always already won at the cost of externalizing violence onto various demonized ‘others’. Agamben’s conceptual vocabulary of ‘the camp’, ‘bare life’, and the ‘exception’ has proved instantly applicable to cultural analyses of the logics of the War on Terror. However, there are two sets of reasons to remain dubious of this type of analysis: empirical reasons related to the failure to acknowledge the degree to which sovereign claims of exception are made effective in part by being justified as necessary; and theoretical reasons concerning the selective interpretation of grand theorists such as Derrida and Foucault in Agamben’s account of the state of exception, biopower, and sovereign power.

In contrast to Agamben’s construal, Brennan and Ganguly (2006, p. 25) observe that the Bush administration ‘sought not to suspend the law but vigorously observe its letter’. Extravagant efforts have been made to find legal justifications, in international law and constitutional law alike, for the use of torture (see Luban 2007), for the legal definition of the status of ‘enemy combatant’, and the use of Guantanamo Bay as a detention camp (Brennan and Ganguly 2006, p. 26). Brenkman (2007, pp. 55–56) also challenges Agamben’s analysis, on the grounds that it conflates the assertion of the necessity of making exceptions with the successful achievement of these claims. There has been and continues to be a political and judicial contestation of the Bush administration’s legal stratagems. The recourse to ‘dodgy dossiers’ or the presentation of questionable intelligence before the UN to justify the invasion of Iraq likewise indicates that the public realm remains a crucial arena in which justifications for emergency measures are made and challenged. The relevance of these practices of public justification for the exercise of sovereignty is entirely elided in Agamben’s account of the state of exception.

Agamben recuperates Foucault’s notion of biopolitics into an account in which all forms of power are derivative of the exercise of sovereignty. Foucault’s concepts of discipline or governmentality are presented as merely instrumental mediums for the more effective exercise of sovereign control over mortality itself. Rabinow and Rose (2006, p. 198) observe that Agamben has become one important authority for a reductionist understanding of ‘biopower’ as underwriting ‘a politics that is fundamentally dependent on the domination, exploitation and, in some cases, elimination of the vital existence of some or all subjects over whom it is exercised’. On this understanding, biopolitics is identical with ‘thanatopolitics’ (Rose 2006, pp. 56–58). Foucault’s writings on biopower hardly authorize such a drastically one‐dimensional analysis of power. He defined ‘biopower’ quite precisely, as distinct from the sovereign exercise of ‘the right to take life and let live’, operating instead as a modality of ‘the right to make live and let die’ (Foucault 2003, p. 241). Biopower is understood to be quite distinct from thanatopolitics (see Foucault 2000, p. 416). Foucault also clearly distinguished power from violence, and explicitly resisted the reduction of all power relations to this single mode. Violence and power are held to stand at opposite ends of a spectrum of relations of interaction (see Foucault 2000, p. 341).

If Agamben drastically simplifies Foucault’s work on power then nor should Agamben’s allusion to Derrida in the analysis of the state of exception be accepted as the authoritative interpretation of deconstruction’s relevance to contemporary geopolitics. It is common to find deconstructive topics such as undecidability interpreted, with the help of Schmitt, to support an account of ‘the political’ in terms of arbitrary force and unfounded decisions (see Barnett 2004). The decisionistic interpretation of the performative force of foundations tends to take Schmitt’s account of sovereignty as the power to decide on the exception as axiomatic. In Schmitt’s original formulation and in Agamben’s faithful reiteration of it, the sovereign is defined as the one who decides on the exception. The circularity in this definition (Bull 2004) effaces the interval between claiming the necessity of making an exception – of suspending the law – and the successful enactment of this claim. The absence of objective foundational validity to norms is, through reference to the Derridean theme of undecidability, made to imply that application and judgement are a purely arbitrary imposition of force by the strongest will. This interpretation misconstrues the practice of judgement that the deconstruction of foundations is meant to throw into new relief (see Brenkman 2007, pp. 59–77; Zerilli 2005).

In Derrida’s writings, the absence of foundations is not the occasion for a depressed resignation to a decisionistic interpretation of sovereign power at all. Rather, a distinctive sense of responsibility is made visible by the undecidability of normative foundations. Derrida does not locate the ‘mystical’ foundation of authority in the self‐assertive will of the sovereign who decides on the exception, but in the exposure to the demands of others (see Barnett 2005). Derrida clearly articulated his doubts about the sustainability of Schmitt’s account of sovereignty and the political, on the grounds that it depends on a logic of conceptual purity in which enmity is always found to be the original source lying behind all apparently reciprocal relationships (Derrida 1997, pp. 246–247). In his own reflections on the philosophical resonances of contemporary geopolitics after 9/11, Derrida undermines the model of sovereignty as wilful self‐assertion upon which the state of exception analysis rests. The circularity of the Schmittian account of sovereignty maintains the assumption of an ‘ipseity’ of the self, defined by the security of an ‘I Can’ (Derrida 2005, pp. 11–13). Against this self‐founding circularity, Derrida affirms the spacing between asserting sovereignty and the enactment of decision which is implied by thinking of sovereignty as performative:

To confer sense or meaning on sovereignty, to justify it, to find a reason for it, is already to compromise its deciding exceptionality, to subject it to rules, to a code of law, to some general law, to concepts. It is thus to divide it, to subject it to portioning, to participation, to being shared. (Derrida 2005, p. 101)

For Derrida, pure sovereignty on the model of decisionist exceptionality does not and cannot exist, in so far as it would exclude the possibility of sharing upon which sovereignty depends. In Derrida’s account, sovereignty is divided at is origin by a spacing that implies sovereignty is always distributed, shared with others (Derrida 2005, p. 45).

The deconstruction of the punctual ipseity of decisionistic sovereignty has consequences for understandings of concepts of force, power and mastery (Derrida 2005, p. 17). If violence is a possibility inhabiting all relations, then so equally is its opposite, non‐violence. While Derrida acknowledged that ‘a certain force and violence’ is irreducible in argumentation and discussion, he argued that ‘none the less this violence can only be practised and can only appear as such on the basis of a non‐violence, a vulnerability, an exposition. … I do not believe in non‐violence as a descriptive and determinable experience, but rather as an irreducible promise and of the relation to the other as essentially non‐instrumental’ (Derrida 1996, p. 83). Reversing and displacing the order of priority found in decisionistic interpretations of deconstruction, Derrida affirms that violence is always entangled in ‘a non‐appropriative relation to the other that occurs without violence and on the basis of which all violence detaches itself and is determined’ (1996, p. 83). The point of calling attention to this entangled relationship of violence and non‐violence in Derrida’s work is to underscore that if the relationship between norms and authority, justice and law, is understood to be ‘undecidable’, then this should not be construed as authorizing the sort of conceptualisation of power as violently sovereign developed by Agamben (see Johnson 2007, Wortham 2007).

What do these empirical and theoretical reasons for remaining sceptical of Agamben’s diagnosis of modernity reveal about the relationship of this strand of philosophical reflection to the task of conceptualising the double responsibility discussed in the first section of this paper? The state of exception argument is informed by and also confirms a model of action that is freed from any horizon of responsibility whatsoever. There are two dimensions to this disavowal of responsibility.

First, with respect to political responsibility, this analysis affirms that all constituted power is tainted with violence and death. On Agamben’s understanding, the actions of modern states always rest on arbitrary decision and violent force. This understanding provides resources with which any and all actions of states can be traced back to this fundamental root (Brenkman 2007, p. 66). In turn, the expression of dissent is thereby freed from any sense of the legitimacy of struggling over the control of institutions (Brennan and Ganguly 2006, p. 27).

Second, this abjuring of political responsibility is matched by the account of ethical action that likewise leaves no ontological or normative space for responsibility. Agamben takes his distance from Schmitt only in so far as the latter sought to recuperate foundational, sovereign violence within the scope of the law (Agamben 2005, p. 59). As Leys (2007, p. 161) observes, for Agamben the notion of responsibility is felt to be inherently unethical because it is ‘irremediably contaminated by law’. In line with the persistent drive for conceptual purity which Derrida identifies in Schmittian political theory, responsibility is therefore abandoned in favour of a model of untainted ethical action modelled on Benjamin’s messianic vision of ‘pure violence’, which seeks to assure violence ‘an existence outside of the law’ (Agamben 2005, p. 59). Benjamin’s concept of the pure or divine violence of revolution is recuperated as a preferred figure of authentic ethical action freed from subordination to the law (Agamben 2005, p. 88). Pure violence is a figure for ‘action as pure means’, with no relation to an end (Agamben 2000). Agamben takes Benjamin as the inspiration for a model of action entirely freed from any implication in public practices where responsibility even arises, and not least freed from the scenes of exposure to alterity upon which Derrida seeks to erect a deconstructive sense of responsibility.

The ontologization of violence leaves little scope for considering the possibility that power can and does circulate in normatively justified and justifiable forms. This posture does little more justice to the potential of concerted action in the public sphere than does the constant trumpeting of unending ‘terrorist’ threats. But there is no good reason to suppose that violence is not a contingent feature of human affairs. A decisionistic view of power is little help in understanding the processes of learning and repair that maintain the fabric of peaceable pluralistic societies (see Spelman 2002). Nor is a decisionistic view of power very much help when it comes to thinking about the possibility of addressing issues of security, international terrorism, or war democratically, that is, through apparatuses of deliberatively generated influence (see Dryzek 2006, Habermas 2006, Young 2007, Ch. 5). The value of practices of reasoning and reasonableness which are recognizably democratic is derived from their fragility, not their firm foundations in grounds of certainty (Keane 2004, p. 3). It is this fragility that is threatened by recourse to violence as a political instrument. What is required, therefore, is to think about the publicity conditions that function as more or less effective, but always fragile contexts of legitimation and justification that can serve as the basis for further progress towards the democratisation of violence (Keane 2004, pp. 154–164).

Justifying violence

The ontologization of violence stretches the concept of violence well beyond the use of physical force to include ‘such phenomenologically elusive categories as psychological, symbolic, structural, epistemic, hermeneutical, and aesthetic violence (Hanssen 2000, p. 9). The expansion of the concept of violence to include a range of structural effects of inequality, exploitation and domination can certainly be effective in providing relevant context in which to explain the emergence of political violence. For example, Johann Galtung’s (1980) expansive reconceptualisation of violence informs a non‐violent politics of peaceful conflict resolution. However, this perspective does still not address the difficult question of whether it is ever possible not only to recognize violence in human affairs, but to justify the use of violence for political ends.

The reason to avoid the conceptual expansion of violence is to keep open the possibility of asking which modalities of action are capable of being subjected to ‘normative steering’ by forms of concerted public action (Arendt 1969, pp. 44–46). ³ There are good reasons to resist the expansive use of violence, both normative and empirical reasons. Keane (2004, p. 35) provides a more restrictive definition of violence as ‘the more or less intended, direct but unwanted physical interference by groups and/or individuals with the bodies of others, who are consequently made to suffer a series of effects ranging from shock, speechlessness, mental torment, nightmares, bruises, scratches, swellings, or headaches through to broken bones, heart attacks, loss of body parts, or death’. Normatively, this definition allows us to acknowledge the variety of forms that violence takes, but without losing sight of some crucial features that enable us to retain a perspective in which questions of justification, legitimacy and responsibility can be addressed. And empirically, we should resist the temptation to over‐extend the term because ‘to spread the term “violence” across all interpersonal relations and solitary actions of which we disapprove actually undermines the effort to explain violence’ (Tilly 2003, p. 4).

In this light, we might usefully follow Hannah Arendt’s (1969, p. 38) argument (echoed by Foucault) that power should not be thought of simply as a ‘form of mitigated violence’. Violence, for Arendt, is instrumental and strategic. Therefore, its use is certainly not irrational. Power, understood as action‐in‐concert, is an end in itself. If power is conceptualized instrumentally, then one is lead either to suppose that the most effective form of power is violence, or at least that various forms of power relation are really just sublimated relations of violence.

Arendt (1969, pp. 51–52) distinguished between the legitimacy and justification of violence: legitimacy refers questions of validity to the appropriateness of the origins of action; justification refers validity to the conduciveness of actions to attaining an end. Justification, in short, is more contingent than legitimacy. For Arendt, violence could be justified, but never legitimate; and its justification ‘loses in plausibility the farther its intended end recedes into the future’ (1969, p. 52). Reflecting on the relevance for the analysis of contemporary geopolitical events of Arendt’s argument that violence can be justified but never legitimate, Young (2007, Ch. 4) argues that any specific deployment of violence for political ends always requires rigorous justification, irrespective of the apparent legitimacy of the state or non‐state actor involved. The latter cannot substitute for or guarantee the validity of the former. If the normative question most at stake in considering terrorism and counter‐terrorism should be justification, then responding to this question necessarily requires consideration of empirical issues of causality, explanation, prediction and consequences.

Michael Ignatieff (2004) offers one account of political violence that gives priority to contexts of justification. He argues that there are important discriminations to be made between the different modalities of violence pursued by ‘democratic states’ and ‘terrorists’. These discriminations turn on the publicity conditions to which different actors are subject:

The liberal state and its terrorist enemy stand under very different obligations to justify their actions. The agents of a constitutional state are aware that they may be called to defend and explain their actions in adversarial proceedings, possibly even in court. Terrorists do not stand in any institutional setting that holds them accountable. They may have an informal moral contract with their base of support, a tacit set of understandings of what types of violence are acceptable, and, in particular, which kinds will expose their base of support to reprisal. But this is not the same as an institutional obligation to render an account of your actions. (Ignatieff 2004, p. 109)

On these grounds, Ignatieff (2004, pp. 82–111) suggests that it is possible to develop an ethical test for the use of political violence, one that would allow for justified political violence ‘as a last resort, in the face of oppression, occupation, or injustice’ (Ignatieff 2004, p. 110). He suggests that the ANC’s military campaign against apartheid within South Africa passes such a test. ⁴ Ignatieff holds that such recourse is justified if it is constrained by a wider web of publicity and accountability conditions. He claims that many contemporary examples of terrorist violence are likely to fail any reasonable test of justification, in so far as they are directed against a culture of dissent, pluralism, deliberation, and compromise, not at extending these values further.

Ignatieff does not develop this same argument very vigorously in the opposite direction. Considering the grounds for justification of the deployment of violence as a political instrument of state action might lead to much more stringent restrictions on such action even in contexts of ‘humanitarian intervention’ than Ignatieff acknowledges. The putative beneficial consequences of such action might be rather more indeterminate and difficult to predict than is often supposed (see Young 2007, pp. 100–104). As already pointed out, if justification is understood to be an empirical matter, drawing into view judgments of causes and explanations and unintended consequences, then the stringency of justification for putatively legitimate states to deploy violence should be considered strengthened rather than undermined. Arendt’s claim that violence can be justified but never legitimate throws the weight of justification squarely on to a consideration of the consequences of any specific deployment of violent means. The empirical dimension of justification implies a much more stringent set of criteria than is often supposed by contemporary apologists for ‘liberal interventionism’ (Young 2007, pp. 100–104). Their disdain for the task of explaining terrorist violence is only matched by their naïve faith in the ease of predicting the outcomes and risks of military interventions.

Ignatieff is marked‐out on the anti‐war left as a ‘liberal imperialist’, having supported the invasion of Iraq, and in turn having then issued a mea culpa for this position. For some that will be enough to invalidate his argument as an apologia for a moral sell‐out to the new imperialism. Ignatieff allows that the judgment of purposes has a place in reasoning over the justification of terrorism that is disallowed by Just War perspectives that view terrorism as inherently unjustified by virtue of the means it adopts (cf. Held 2004). Methodologically, Ignatieff’s position is not significantly different from Ted Honderich’s argument regarding the situations in which political violence can be justifiably deployed as a means of extending values of equity, democracy, and justice. Honderich’s arguments became embroiled in the politics of what it is permissible to say and hear in the post‐9/11 public sphere. His Violence for equality, first published in 1980, has recently been reprinted and updated, as Terrorism for humanity (Honderich 2003b). His After the terror (Honderich 2003a) became the subject of controversy when the author donated £5000 of royalties to Oxfam, only for the charity to refuse to accept them following a media‐induced campaign of outrage that a development charity should be seen to be accepting the profits from a book interpreted as condoning terrorism. ⁵

Honderich holds to the plausibility of talking about ‘democratic violence’, on two grounds: that the use of violence in some circumstances can advance progress towards freedom and equality; and because some forms of political violence share attributes allowed by the criteria of democracy, not least attributes of persuasion (Honderich 1980, pp. 165–170). Most pertinent to the discussion here is how Honderich locates the issue of responsibility in relation to political violence. Responsibility arises in his analysis through the philosophical topic of acts and omissions. It is a commonplace to argue that omissions can cause suffering as intensively as can directly intentional acts. The move to widen the range of concerns for which actors can be held responsible often rejects the distinction between acts, omissions, and unintended outcomes. These distinctions are accused of serving as excuses for restricting the scope of responsibility (e.g. Singer 1993, pp. 206–213). However, Honderich argues that the importance of the distinction between acts and omissions lies in showing how the usual condemnation of political violence is incoherent. It supposes that only intentional acts are causes of harm that can justifiably give good cause to other people to take up violent measures. From the perspectives of perpetrators of political violence and their supporters, violence might well be justified on tu quoque grounds – that is, on the grounds that by virtue of ‘our’ ordinary omissions, ‘we’ contribute to the reproduction of insufferable hardship, harms, and oppressions that can only be changed by recourse to violent means (Honderich 1980, pp. 96–100). From the point of view of ‘us’, this seems to flatten any distinction at all between what ‘we’ intend to do and the side effects of intentional actions. This is a distinction that Walzer (2002), for example, insists must be upheld for any moral course of action to remain conceivable. But Honderich’s point is precisely that acknowledgement of perspective is central to reasoning about questions of responsibility. From the perspective of those who suffer from the unintended consequences of actions, the distinction between intended and unintended harm might appear a little less robust as a ground for moral justification than is often supposed. Honderich’s argument reverses the perspective from which questions of responsibility and justification arise, moving away from the intentions of actors and towards the perspective of victims. ⁶ Any consideration of the relationship between justifications of violence needs to address both actor‐oriented perspectives and victim‐oriented perspectives.

Despite their different stances on contemporary events, neither Ignatieff nor Honderich suppose that the main criterion for demarcating between justified violence and unjustified violence is simply that the former is the action undertaken by democratically legitimate states. They maintain stringent criterion of justification, primarily referring to a set of publicity conditions embedded across the state/civil society boundary that are meant to ensure accountability and proportionality. The space they share does suppose that there are circumstances in which the use of violence can serve democratic, even ‘humanitarian’, objectives. One reason, then, for affirming the importance of causal‐explanatory knowledge is that it provides a very stringent test of the justification of violence. But it is important to underscore that the arguments considered in this section do not suppose that the rightness of any violent action can solely be judged by reference to predicted ends and outcomes. The criteria for a ‘democratic ethics of violence’ are not consequential considerations alone. As I noted earlier, Keane (2004, pp. 159–160) insists that the idea and possibility of ‘democratic violence’ depends on the existence of a robust network of institutions of public accountability and an open and free civil society. These publicity conditions are essential to ensuring that the violence deployed in the name of democratic ends remains contingent and removable. An infrastructure of effective public culture and accountability is thus ascribed a kind of deontological function of acting as the restraint on excessive use of violence for justifiable political ends (see Nagel 1979).

Ignatieff and Honderich both give weight to the importance of public practices of reason‐giving as the fragile medium through which justification and democratic legitimacy are formed. But this emphasis needs to be radicalised in order to supplant the self‐centred model of responsibility that both intention‐centred and actor‐centred models reproduce. Both the intention‐oriented, actor‐oriented perspective and the effects‐oriented, victim‐oriented perspective tend to be abstracted from consideration of the conditions in which claims of responsibility are levelled. What is needed is a conceptualization that redistributes responsibility as a form of action‐in‐concert that is summoned in and through scenes of public address. The final section considers how recent feminist theories of responsibility, written in the shadow of contemporary global politics, help to develop this type of understanding.

Responsibilities

The first section outlined the polemical terrain in which explanation is dismissed as mere exoneration, and outlined the imperative of a double responsibility required to connect explanation to acknowledgement. The second section explored how a decisionistic interpretation of the ontological foundations of violence as a uniquely generative feature of human affairs squeezes any space for normative public action out of its political ontology. In the third section, we saw how the acknowledgement of violence in political philosophy is related to consideration of practices of justification and legitimation that do give weight to public action, but do so only by limiting the constitution of public concerns to self‐generated intentions or effects. In order to elaborate further on the dual thinking called for in the first section, this final section draws on the affirmation of the normative force of publicness intimated by the writers considered in the third section, but detaches this from the model of action that restricts the attribution of responsibility for justice to the actions of self‐mastering actors.

The idea that social science explanations of political violence lay the basis for undermining practices of responsibility is a long‐standing philosophical suspicion (see Turner 2006, pp. 11–12). The kernel of truth to this idea lies in the fact that such causal‐explanatory knowledge challenges a model of free will that is central to understandings of responsibility which centre on the assignment of blame. But the previous section indicates why the equation of explanation with exoneration is wrong‐headed: once justification is distinguished from legitimacy, then explanatory knowledge becomes quite central to the normative task of justification, which always requires consideration of causes, effects, and consequences.

Distinguishing between causal‐explanatory knowledge and normative evaluation is, then, a central task for developing the double responsibility discussed in the first section. One starting place for thinking about this double responsibility is a distinction that Judith Butler makes in admitting that focusing on explanation can slip into a style of exoneration if it elides questions of responsibility. She suggests that questions of responsibility are kept alive rather than collapsed by thinking in terms of the conditions of terrorist violence rather its causes. As she puts it, conditions do not act, agents do. But no agent acts without conditions. Individuals are responsible for the acts they bring about, and this principle follows from insisting on the principle that people ‘are not dupes or mechanisms of an impersonal social force, but agents with responsibility’ (Butler 2002, p. 186). But Butler goes on to suggest that acknowledging the conditions in which agents act is important not because the task at hand is that of holding actors responsible. It is important in accounting for and helping to justify the possibility that actors can take responsibility for events that do not derive from the range of their effective, intended or unintended causal action at all.

Butler’s distinction between causes and conditions is in line with a broader range of feminist moral philosophy which develops a ‘forward‐looking’ sense of ‘taking responsibility’ (Card 1996, pp. 27–29, 143–151). This approach does not suppose that questions of responsibility start and end with the self‐caused actions of which individuals are the authors. The concept of taking responsibility requires rethinking the relationships between the determinant conditions of action and the modes of responsible agency which they enable (e.g. Lloyd 2000). This type of conceptualisation of taking responsibility is not associated with a disavowal of assigning responsibility in moral terms. Card’s account of taking responsibility is linked to her reconstruction of a secular theory of evil. On what she calls ‘the atrocity paradigm’, evil is defined in terms of foreseeable intolerable harms produced by culpable wrongdoing (2002). This theory holds together the sense that certain forms of harm are both perpetrated and suffered. It is on these grounds that Card (2003) unambiguously describes the 9/11 attacks as evil. Card delimits evil around the notion of culpability, and this frees up responsibility from a restriction to notions of blame and guilt. Harms which are not tied to culpable wrongdoing are still harms (2002, p. 7); furthermore, atrocities can give rise to ‘blameless guilt’ over actions of others in which one had no say but were still implicated in (p. 202). In both cases, responsibility starts where intended action or unintended consequences end.

The distinction between holding responsible and taking responsibility is also worked up in Iris Marion Young’s work on political responsibility. Young’s starting point is the importance of de‐centring a blame model of responsibility, and she uses examples such as global labour solidarity movements (2004) and responses to Hurricane Katrina (2006) to elaborate on this conceptualization (see also Massey 2006). The blame model of responsibility tends to focus on individuals, searches for malevolent intent, generates defensiveness and blame‐switching, and tends to limit attention to the small time‐frames of determinate events. Therefore it detracts from focusing on background conditions that generate harms on specific occasions and from the task of collective mobilization in response to these conditions. In Young’s reconstruction of political responsibility, being directly responsible for harmful consequences is differentiated from benefiting from certain actions and from being empowered to act to change structural patterns of disadvantages or vulnerability (Young 2007, Ch. 9). Making these distinctions helps to focus attention on the question of effective agency in future‐oriented action as the critical issue in assigning responsibility.

These feminist conceptualizations of taking responsibility draw attention to the degree to which the idea that responsibility is threatened by causal‐explanatory knowledge remains restricted to a blame‐centred model of agency. In the final analysis, this model is indifferent to suffering, harm and loss. The practice of taking responsibility cannot be cultivated by either holding fast to a strict notion of individual responsibility, acts and omissions, or the doctrine of double effect; nor by simply supposing that questions of an individual or collective actors’ intentions can be wholly disregarded, and that responsibility can simply be deduced from a consideration of effects and the distribution of benefits and suffering. In one view, the agent of responsibility is viewed as isolated and self‐contained; in the other view, the agent of responsibility is expected to be able to act in response to a potentially infinite range of avoidable harms. And on both views, the agent of responsibility is still presented as a self‐mastering, omnipotent benefactor of less fortunate passive beneficiaries. It is on the basis of this sort of understanding that Young (2003) takes Habermas (and Derrida) to task for continuing to suppose that the kind of cosmopolitan responsibility they call for as a response to contemporary geopolitics can be centred on Europe’s own self‐understanding of its historical responsibilities (see Habermas and Derrida 2003). What is needed, she argues, is a style of response that starts by listening to voices from beyond this centre.

The concept of taking responsibility implies a more modest concept of responsible agency. The attribution of responsibility arises as a political problem in contexts in which demands for action are publicly articulated (Barnett et al. 2008). Feminist theorists of forward‐looking responsibility hold that it is necessary to discriminate between types of action in order to keep open the public spaces in which it is possible to reason about demands to take responsibility for outcomes, harms and suffering that are not an actor’s own doing but which they might nonetheless be empowered to act on, to alleviate, prevent, or repair. Butler spells out how attributions of responsibility should be rethought once the question of effective agency is foregrounded:

Our acts are not self‐generated, but conditioned. But we are acted upon and acting and our ‘responsibility’ lies in the juncture between the two. What can I do with the conditions that form me? What do they constrain me to do? What can I do to transform them? Being acted upon is not fully continuous with acting, and in this way the forces that act upon us are not finally responsible for what we do. (Butler 2002, p. 187)

In Butler’s (2005, pp. 84–88) formulation of an ethic of responsibility, this excess of potential agency over self‐caused actions is generated in scenes of address by others. It is in these spaces that a subject’s susceptibility to being acted upon prior to active choice is made the condition for responsibility even arising as a possibility in the first place.

It is a basic assumption in arguments in favour of notions of collective responsibility that there are real differences between the outcomes an actor intentionally brings about and the effects in which an actor is anonymously implicated (Lloyd and Gatens 1999). Feminist theorists writing in the context of contemporary global political events identify two additional aspects of responsibility: the idea that attribution follows from consideration of the distribution of effective agency; and the idea that responsibility arises in response to the claims of others for acknowledgement of various types. Taking responsibility is not premised on abandoning the importance of intentional action in favour of a consideration of effects. Taking responsibility is not an expression of atonement for an actor’s own actions. Nor does it depend on an objective assessment of the relative weight of harm an actor is responsible for. Rather, it is a practice that prioritizes formulating a response which is attentive to the agency of others and mindful of uneven capacities for agency. It is, therefore, a practice that draws on and re‐enacts the fragile spaces of public action. Formulating such a mode of attentive responsiveness needs to go hand in hand with causal‐explanatory inquiry. Not only does cultivating both of these practices not involve excusing terrorism, but it also helps in the task of ‘not excusing those who wilfully fail to understand it’ (Held 2004, p. 72).

Conclusion

I have argued that one strand of contemporary critical theory that seeks to address the philosophical significance of the War on Terror, exemplified by the work of Giorgio Agamben, tends towards the ontologization of violence. In seeking to understand the ineluctability of violence in human affairs, the metaphysical commitments of this approach lead to the elision of the publicity conditions through which the political deployment of terrorist and counter‐terrorist violence works in practice. In contrast to this approach, I have argued that acknowledging that the force exercised by political violence depends on its being performed in a public medium provides an entry point for theorising the practical possibility of what John Keane calls the democratization of violence. Drawing on recent interventions by writers including Keane, Virginia Held, and Iris Marion Young, it has been argued that rather than focus solely on the question of the legitimacy of the recourse to violence, more attention should be given to criteria of justification. Focussing upon criteria of justification draws into view an understanding of the relationship between violence and the publicity which does not simply look at whether a democratically constituted public has provided the a priori validation of legitimate political action. Rather, criteria of justification require thinking of publicity as a dynamic medium in which the means, ends and consequences of action are constantly open to question. In order to further appreciate the possibility of action being normatively steered by publicly acknowledged responsibilities to others, I have argued that recent feminist literature provides valuable insights into the theorisation of the future‐oriented dimensions of responsibility.

Note on contributor

Clive Barnett is Reader in Human Geography in the Faculty of Social Sciences at The Open University, in Milton Keynes, United Kingdom. He is the author of Culture and democracy: media, space and representation (Edinburgh University Press, 2003), co‐editor with Murray Low of Spaces of democracy (Sage, 2004), and co‐editor with Jenny Robinson and Gillian Rose of Geographies of globalisation (Sage, 2008).

Notes

1. Available at: http://eustonmanifesto.org/the-euston-manifesto/.

2. The term decisionism has long‐standing importance in debates around deliberative and radical democracy more widely (see Honig 2007).

3. On the importance of differentiating forms of power for making significant judgements about the location of responsibility, see Connolly (1993, Ch. 3).

4. Held (2005) discusses the degree to which groups who can claim to be legitimate representatives of oppressed peoples can justifiably have resort to terrorist violence.

5. Honderich’s account of this affair is at http://www.ucl.ac.uk/~uctytho/.

6. Honderich’s argument shares features with Galtung’s (1980) approach. Galtung does not accord primacy to the acts carried out by individuals or collectives in assigning responsibility for what he calls structural violence; it adopts the perspective of victims rather than perpetrators.