1. Targeting the subject

On 1 May 2011, President Obama announced the death of Osama bin Laden. The al Qaeda leader and chief architect of the terror attacks of 11 September 2001 had been killed in his compound in Pakistan in an operation carried out by a team of US Navy SEALs. ‘Justice has been done,’ the president declared in a public speech at the White House:

The president's statement strikingly echoes some characteristic as well as controversial features of the juridical debate on targeted killing, which has become a crucial tactic in the US fight against terrorism. First, since targeted killing is accomplished in the name of national security, it does not appear to be a form of political assassination, but, instead, figures as a military concern. Second, targeted killing operations are supposed to respond to an enduring and potentially catastrophic threat that can be named but remains uncertain. Third, notable is the president's rhetoric of justice, which stands at odds with the administration's concern to present itself as acting lawfully.Footnote ² As one legal scholar observes:

The rhetoric of justice, evoking associations of punishment or revenge, places targeted killing on the edge of the rule of law. It may be taken as an indicator of the president's confidence about the operation's acceptance, but also that targeted killing still is in the process of being inscribed into the law. As a matter of fact, official statements immediately after the incident held that the operation was prepared ‘to take bin Laden alive, if he didn't present any threat’.Footnote ⁴ Justice in this sense would mean that the operation was not in fact an act of targeted killing, but an attempt to detain bin Laden with the use of lethal force as a last resort. According to an early account in the media, however, the possibility of an aerial bombardment, which would certainly have killed bin Laden as long as he had been correctly located, had also been considered in the run-up to the strike. It was finally rejected because it would have required excessive bombing that might have not only caused too much ‘collateral damage’, but, significantly, failed to provide physical evidence of the world's most wanted terrorist.Footnote ⁵ Instead, an official statement referred to the operation as a ‘surgical raid’,Footnote ⁶ which is a telling description given that the presumed precision of targeted killing operations points to their deliberateness, and the fundamental issue here is the legitimacy of states, or state-authorized agencies, to intentionally kill citizens.Footnote ⁷

Until recently, targeted killing was widely perceived as illegitimate and illegal, if not ‘unlegalizable’ in David Dyzenhaus's sense of an incontestable norm that deems a certain act in itself to be ‘morally wrong’.Footnote ⁸ State-sponsored killing is generally considered to be a last resort. Hence, the UN Charter grants states a very limited right to use force in relation to another state, by and large only for self-defence.Footnote ⁹ Furthermore, the elimination of world leaders is internationally outlawed in the face of an otherwise uncontrollable circumstance where, in theory, any government could designate an enemy as a legitimate target.Footnote ¹⁰ And, even if targeted killing operations are approved as an instrument of law enforcement, states are faced with the alternative options of detention and prosecution on the grounds of international human rights law and, in particular, the targeted subject's right to life.Footnote ¹¹

Targeted killing has been criticized as extra-judicial or arbitrary execution, assassination, or a pure act of revenge.Footnote ¹² While maintaining a critical stance towards the Israeli practice, the United States, from the 1990s onwards, nonetheless gradually moved towards openly adopting targeted killing as a tactic in the fight against terrorism.Footnote ¹³ Although remaining controversial, targeted killing, particularly in the aftermath of 9/11, received increasing political and, most notably, legal approval.Footnote ¹⁴ Rather than being accomplished extra-legally or as a normalized exception to the rule, as critics proclaim, targeted killing becomes inscribed into a law already prepared to accept it as a practice. This paper argues that targeted killing can be deciphered as a security dispositif – that is, as an ensemble of heterogeneous forms of knowledge and practices that become strategically interlinked in the governance of security problems and thereby establish certain regimes of visibility and frameworks of thinking.Footnote ¹⁵ The notion of dispositifs allows us to grasp how targeted killing both presents itself as being a suitable response to an allegedly urgent security problem and, as a particular technology, deploys its own rationalities and dynamics of action.Footnote ¹⁶ With their logic of pre-emptive and focused strikes, targeted killing operations relate to and at the same time redefine the register of established concepts that underlie international law, namely ‘state’, ‘war’, and ‘enmity’. International law, as a form of knowledge accommodating targeted killing as a mode of dealing with certain security matters, in turn encourages the establishment of a new security dispositif. Targeted killing itself thereby surfaces as a new phenomenon, namely as a military tactic of air strikes on a foreign territory employed in the fight against terrorism. It thus manages to present itself as a legitimate security technology and to gradually discard its historical association with the practice of political assassination. This will be illuminated in the following section 2. Targeted killing, then, is not external to the law. On the contrary, as a military tactic, it also carries legal claims about, for example, justified forms of killing individuals. Conceiving of law, in Foucauldian terms, as a practice that is indissoluble from the forms of knowledge both that enact it and that its enactment invokes allows us to trace the mechanisms of targeted killing's asserting itself as a legal practice. Precisely because law relies upon knowledge, it is susceptible to security concerns and shows itself to be prepared to accommodate targeted killing in legal terms. International law, as a consequence, is itself in transition. This account of law will be developed subsequently (section 3). If there is a conception of ‘the political’ in Foucault's writings to be discerned, it consists of inquiring how things and sites, as subjects of problematization and governmental knowledge, are being constituted and thus rendered accessible to intervention. The identification of threats provides a space for transforming the unknown into new forms of knowledge (section 4). Once it entered the public stage, targeted killing, as a state practice of resorting to the use of lethal force, incited the legal debate. Legal reasoning, however, couching the practice in legal terms and producing a normative reality of its own, contributes to targeted killing's legalization. Targeted killing and its law, in this sense, are mutually constitutive (section 5).Footnote ¹⁷

2. Targeted killing goes public within a new security dispositif

As a practice, state-sponsored targeted killing is not a new phenomenon. It can be seen as the continuation of a long history of assassinations of political leaders – a practice that, precisely due to its political nature, lies between legality and illegality. It has largely been entrusted to secret-service agents or military undercover units and is expected to be accomplished in a more or less clandestine manner.Footnote ¹⁸ The matter of the ‘reason of state’ becomes manifest here,Footnote ¹⁹ as the legitimacy of state-sponsored killing depends upon the sovereign state's defining it as such. This prerogative, to be sure, is a mutually conceded one. It forms part of the modern conception of the sovereign state itself.

As regards the US context, ‘assassination’ is prohibited according to Executive Order 12.333. As a consequence of criticism of the CIA's active involvement in assassination plots that arose in the mid-1970s, this executive order was finally signed in 1981 by President Reagan and still remains in effect today. Although leaving the term itself rather unspecified and thus with considerable scope for interpretation, it does not prohibit the targeted killing per se of terror suspects.Footnote ²⁰ Being framed as a military tactic, however, targeted killing rids itself of its historical association with the clandestine practice of assassinating political leaders and opponents. A brief look at some of the key discursive incidents on the subject may give an idea of how the term itself came to represent a kind of new phenomenon. One decisive incident was Israel's public admission in 2000 of employing targeted killing as a tactic in the Palestinian territories. According to the Human Rights Council,Footnote ²¹ this governmental statement made the term itself official for the first time, even though the practice of killing significant opponents was part of the conflict as far back as Israel's independence.Footnote ²² As regards the US policy, the Human Rights Council continues: ‘The first credibly reported CIA drone killing occurred on November 2002, when a Predator drone fired a missile at a car in Yemen killing . . . an al-Qaeda leader allegedly responsible for the USS Cole bombing.’Footnote ²³ It was, however, already during the presidency of Bill Clinton that missiles had been launched against al-Qaeda members in Afghanistan. In the aftermath of the embassy bombings in Kenya and Tanzania in 1998, a presidential finding authorized the use of lethal force in the name of self-defence.Footnote ²⁴ Yet, the subsequent strikes in response to terrorist acts, if they became public at all, were criticized as ‘armed reprisals’, which, for obvious reasons, ‘are almost universally regarded . . . as having been outlawed in contemporary international law’.Footnote ²⁵

The Bush administration did not yet turn targeted killing into an official programme, but it set the conditions for one when the president, in 2002, famously claimed the right of preventive or pre-emptive self-defence to be an essential and ‘enduring’ part of the national security strategy.Footnote ²⁶ The administration followed two legal references that were immediately provided after 9/11: first, the Congress's approval ‘to use all necessary and appropriate force’ in the fight against terrorism;Footnote ²⁷ second, the two UN Security Council Resolutions that recognized the right to self-defence pursuant to Article 51 of the UN Charter and expressed ‘its readiness to take all necessary steps to respond to the terrorist attacks of 11 September 2001, and to combat all forms of terrorism’.Footnote ²⁸ Although not automatically authorizing drone attacks, these decisions provided a suitable framework for the US government to consider the terrorist acts as constituting ‘an “armed attack” . . ., triggering application of the laws of armed conflict’.Footnote ²⁹

It was only with President Barack Obama's ‘drone program’Footnote ³⁰ that targeted killing operations were systematically and more openly employed in the fight against terrorism. Since Obama entered office, there has reportedly been a conspicuous increase of aerial strikes, mainly in Pakistan. Targeted killing became a subject of public knowledge and thus publicly visible.Footnote ³¹ As a security technology assigned to the context of military operations, the term itself then came to represent a rather new phenomenon of (mainly drone-launched) air strikes on a foreign territory – which, of course, does not preclude intelligence continuing to play a crucial role in the fight against terrorism and accomplishing respective missions.Footnote ³² Yet, within the ‘theatre of war’, as this telling phrase indicates, the practice of killing political opponents takes an entirely different shape. The exercise of sovereign power sees itself authorized to address corresponding bodies of law, notably around ‘self-defence’ and ‘armed conflict’. It is with these legal references that a justification of targeted killing operations apparently ceases to be required, according to US State Department Legal Advisor Harald S. Koh:

At the very same moment as targeted killing entered the public stage, it became legalizable. It did so as a security dispositif by locating itself within the legal discourse and at the same time relocating elementary conceptions of existing international law.

It was the identification of a new dimension of threats that, in the first instance, paved the way for targeted killing's surfacing on the political and legal stage. With the ability to utilize weapons of mass destruction or to display the capacity to invent such weapons, contemporary terrorism has been perceived as competing with the destructive power of states and, at the same time,Footnote ³⁴ being organized in transnationally operating networks, as an unforeseeable threat. The rationale of facing this threat no longer consists of deterring the attack by a known enemy state, but of pre-empting ‘the danger before it is known’ and before it has a chance to even emerge.Footnote ³⁵ If international law was prepared to accommodate targeted killing in legal terms, this was also the case because radical uncertainty, in the sense of the unforeseeable and possible, had already been introduced into legal reasoning.Footnote ³⁶ The precautionary logic constitutes a crucial feature of the new security dispositif – and a condition of possibility for targeted killing to be embraced by international law.

As a dispositif, targeted killing entails the claim of its being an appropriate response to the new dimension of threats. Its promise is that a limited, or ‘surgical’, intervention brings about the greatest effects. The rationale is to intercept, or pre-empt, any preparatory terrorist action and thereby figure out the source of the problem – in the present context, leaders and core figures of a terrorist organization – in order to disrupt the whole matter. Terror networks, through this lens, then, appear to be the mirror image of this dispositif. Rather than merely being a response to the presumed problem, targeted killing asserts that this kind of organization would in fact be its very structure. It thus relocates the legal notions of war and self-defence, once clearly attributable to ‘the political space of sovereignty’,Footnote ³⁷ within an entirely new constellation. What is at stake is no longer the idea of a confrontation between states, but rather the concerted acts of individuals.

If targeted killing could re-emerge as a new phenomenon and legitimate subject of legal debate, gradually losing resemblance to the classical forms of political assassination, this induced a distinctive kind of politico-legal question. The fight against terrorism, namely, is to be assigned to the legal sphere of either warfare or crime control. This decision makes a considerable difference as regards both the rights of state authorities to exercise lethal force and the due-process guarantees of the impacted individuals.Footnote ³⁸ It is, however, only the traditional notion of sovereign states that suggests a clear distinction be made between foreign and internal affairs, military and criminal cases, war- and peacetime, in accordance with bodies of law. Reluctance to accept targeted killing as a legitimate measure, even when basically assenting to ‘the morality of killing in the context of war’,Footnote ³⁹ within this framework seems quite rational – that is to say, once targeted killing is regarded as being an instrument of ordinary law enforcement.Footnote ⁴⁰ Those clear distinctions, however, have always been an idealization. Wartime, which is thought of as an exception to the norm, intrudes into everyday life through both memories and anticipation.Footnote ⁴¹ The presence of past wars in public debates is as much a testimony to this phenomenon as current political invocations to prepare for the next attack. There is also continuity between war- and peacetime that is reinforced by technologies and institutions.Footnote ⁴² The convertibility of military into civilian techniques, and vice versa, is to mention just one facet: the possible double use of drones in war- and in peacetimes another. It is only the awareness of boundaries being blurred that is a rather recent phenomenon.Footnote ⁴³ And, in fact, to the extent that targeted killing replaces the notion of assassination, the targets themselves are no longer civilian political leaders, but terroristsFootnote ⁴⁴ – a term that comes to be located within the juridical debate beyond the distinction of soldier or civilian. If targeted killing today in the fight against terrorism appears to be an appropriate security technology, embedded within international law, this acceptance in turn is evidence of a new security dispositif's becoming the norm.

Within a Foucauldian perspective, talk about a new security dispositif does not imply that one dispositif would replace the other altogether, but rather that established notions and practices become relocated and linked to new ones. Sovereign power thus in no way loses its significance, but sees itself confronted with new challenges and obligations, and endowed with new momentums of authorization. Targeted killing, in this sense, itself shapes state formation,Footnote ⁴⁵ namely our understanding of sovereignty, of the rule of law, and of what is a legal and an illegal practice. Rather than asking whether international law competes with the sovereignty of states, focus, within this perspective, is on how sovereignty transforms and constitutes itself anew by enforcing international law; how distinctions are being made, for example, between national and international legal matters or between laws of war and ordinary law enforcement; and what kinds of concept underlie legal norms and are being inscribed into the law.

3. A Foucauldian perspective on law

Foucault did not elaborate on a comprehensive theory of law – a fact that critics have attributed to his allegedly underestimating law's political and social relevance. Some statements by Foucault may have provoked this interpretation, among them his assertion that law historically ‘recedes’ with,Footnote ⁴⁶ or is being ‘colonized’ by,Footnote ⁴⁷ forms of knowledge that are addressed at governing people and populations. It is, though, precisely this analytical perspective that allows us to capture the mutually productive relationship between targeted killing and the law. In contrast to a widely shared critique, then, Foucault did not read law merely as a negative instrument of constraint. He referred, instead, to a particular mode of juridical power that operates in terms of repressive effects.Footnote ⁴⁸ Moreover, rather than losing significance coextensively with the ancient sovereign power, law enters new alliances, particularly with certain knowledge practices and attendant expertise.Footnote ⁴⁹ This linkage proves to be relevant in the present context, considering not only the interchange between the legal and political discourse on targeted killing, but notably the relationship between law and security.

According to Foucault, social phenomena cannot be isolated from and are only decipherable within the practices, procedures, and forms of knowledge that allow them to surface as such.Footnote ⁵⁰ In this sense, ‘all phenomena are singular, every historical or social fact is a singularity’.Footnote ⁵¹ Hence, they need to be studied within their historically and locally specific contexts, so as to account for both the subject's singularity and the conditions of its emergence. It is against this background that a crucial question to be posed is how targeted killing could emerge on the political stage as a subject of legal debate. Furthermore, this analytical perspective on power and knowledge intrinsically being interlinked highlights that our access to reality always entails a productive moment. Modes of thinking, or what Foucault calls rationalities, render reality conceivable and thus manageable.Footnote ⁵² They implicate certain ways of seeing things, and they induce truth effects whilst translating into practices and technologies of government. These do not merely address and describe their subject; they constitute or produce it.Footnote ⁵³

Law is to be approached accordingly.Footnote ⁵⁴ It cannot be extracted from the forms of knowledge that enact it, and it is in this sense that law is only conceivable as practice. Even if we only think of the law in ideal terms, as being designated to contain governmental interference, for example, or to provide citizens’ rights, it is already a practice and a form of enacting the law. To enforce the law is always a form of enactment, since it involves a productive moment of bringing certain forms of knowledge into play and of rendering legal norms meaningful in the first place. Law is susceptible to certain forms of knowledge and rationalities in a way that these constitute it and shape legal claims. Rather than on the application of norms, legal reasoning is on the production of norms.

Legality, within this account of law, then, is not only due to a normative authority that, based in our political culture, is external to law, nor is it something that is just inherent in law, epitomized by the principles that constitute law's ‘inner morality’.Footnote ⁵⁵ Rather, the enforcement of law and its attendant reasoning produce their own – legal – truth effects. Independently of the purported intentions of the interlocutors, the juridical discourse on targeted killing leads to, in the first instance, conceiving of and receiving the subject in legal terms.

When targeted killing surfaced on the political stage, appropriate laws appeared to be already at hand. ‘There are more than enough rules for governing drone warfare’, reads the conclusion of a legal reasoning on targeted killing.Footnote ⁵⁶ Yet, accommodating the practice in legal terms means that international law itself is undergoing a transformation. The notion of dispositifs is useful in analysing such processes of transformation. It enables us to grasp the minute displacements of established legal concepts that,Footnote ⁵⁷ while undergoing a transformation, at the same time prove to be faithful to their previous readings. The displacement of some core features of the traditional conception of the modern state reframes the reading of existing law. Hence, to give just one example for such a rereading of international law: legal scholars raised the argument that neither the characterization of an international armed conflict holds – ‘since al Qaeda is not a state and has no government and is therefore incapable of fighting as a party to an inter-state conflict’Footnote ⁵⁸ – nor that of an internal conflict. Instead, the notion of dealing with a non-international conflict,Footnote ⁵⁹ which, in view of its global nature, purportedly ‘closely resembles’ an international armed conflict, serves to provide ‘a fuller and more comprehensive set of rules’.Footnote ⁶⁰ Established norms and rules of international law are preserved formally, but filled with a radically different meaning so as to eventually integrate the figure of a terrorist network into its conventional understanding. Legal requirements are thus meant to hold for a drone programme that is accomplished both by military agencies in war zones and by military and intelligence agencies targeting terror suspects beyond these zones,Footnote ⁶¹ since the addressed is no longer a state, but a terrorist network.

However, to conceive of law as a practice does not imply that law would be susceptible to any form of knowledge. Not only is its reading itself based on a genealogy of practices established over a longer period.Footnote ⁶² Most notably, the respective forms of knowledge are also embedded in varying procedures and strategic configurations. If law is subject to an endless deference of meaning,Footnote ⁶³ this is not the case in the sense of arbitrary but historically contingent practices, but in the sense of historically contingent practices. Knowledge, then, is not merely an interpretive scheme of law. Rather than merely on meaning, focus is on practices that, while materializing and producing attendant truth effects, shape the distinctions we make between legal and illegal measures. What is more, as regards anticipatory techniques to prevent future harm, this perspective allows for our scrutinizing the division made between what is presumably known and what is yet to be known, and between what is presumably unknown and has yet to be rendered intelligible. This prospect, as will be seen in the following, is crucial for a rereading of existing law. It was the identification of a new order of threat since the terror attacks of 9/11 that brought about a turning point in the reading of international law. The identification of threats in general provides a space for transforming the unknowable into new forms of knowledge. The indeterminateness itself of legal norms proves to be a tool for introducing a new reading of law.

4. Uncertain threats, governed by law

While abandoning the rhetoric of war, the Obama administration, in fact, sees itself in an armed conflict with al Qaeda and thereby continues to draw on the same statutory authorizations as the previous White House.Footnote ⁶⁴ This becomes evident, for example, when US State Department Legal Adviser Harald S. Koh outlines the legal basis of targeted killing operations:

Self-defence may be understood as a reaction to an armed attack, but, at the same time, it aims to prevent the next strike. In this sense, it involves an anticipatory moment. Yet, the assumption of an ongoing catastrophic threat, which, within the legal discourse on targeted killing, tends to be presented as an undisputed fact,Footnote ⁶⁶ contradicts the conventional reading of the right to self-defence that, in a temporal sense, depends on the identification of an imminent threat: the threat is either manifest or at least ‘immediately’ emerging.Footnote ⁶⁷ It is, moreover, the premeditated nature of targeted killing operations itself that contests this reading.Footnote ⁶⁸ The question of legitimate killing, then, poses in a twofold manner, taking both the controversial reading of the anticipated threat into account and the fact that this supposedly emanates from an individual target. Traditionally, soldiers may be legitimately killed due to their belonging to the other side's army. Apart from whether they actually pose a danger, during armed conflict, ‘they are potentially lethal agents of that hostile party’.Footnote ⁶⁹ They represent an anonymous, ‘institutionalized danger’.Footnote ⁷⁰ The ‘high-profile terrorist’ whose name appears on a special target list, by contrast, poses an individualized threat.Footnote ⁷¹ Within this constellation, legal scholars contend that the legitimacy of killing an individual terrorist hinges on the subject's alleged role within the organization.Footnote ⁷² Both the moral and the juridical problems, therefore, consist of how to verify that a terrorist suspect's activities in facilitating and planning attacks, which have not yet proved harmful, in fact contribute to an intolerable threat in a way that justifies his killing.Footnote ⁷³ Rather than merely as a temporal matter, urgency here presents itself as a question of rendering the ‘unknowable threat’ intelligible and thus manageable.Footnote ⁷⁴

Any form of averting dangers entails uncertainty, and therefore a productive or speculative moment. It requires dealing with an unknown and, at the same time, presumably threatening future. Preventing harm thereby aims at having certain facts not eventuate at all.Footnote ⁷⁵ Danger is to be anticipated, even if difficult to conjecture. This is all the more the case the more abstract the threat and the greater the expected harm is. Indications of dangers or threats have to be taken seriously, while these threats at the same time are difficult to dispute, precisely because they escape determination. This is where the qualitative dimension comes into play: the identification of a catastrophic threat is based on the ‘ontological premise’ that it may manifest at any time without notice under an entirely new signature while causing intolerable harm.Footnote ⁷⁶ Yet, these processes of identification and definition, and of targeting, or focusing on, something are not merely questions of gathering knowledge about the addressed. They are also forms of creating knowledge that point to the one addressing the phenomenon in question.Footnote ⁷⁷ Hence, only once we accept the identification of a threat does it then require dividing the known from what is unknown and what is to be known.Footnote ⁷⁸ The unknown threat has to be approached and assessed in order to render it accessible to governmental intervention.Footnote ⁷⁹

This is the critical moment of law's susceptibility. It is, paradoxically, the state of uncertainty that enables security matters to be embraced by law. Law, on the one hand, relies on indeterminateness,Footnote ⁸⁰ so as to be open to interpretation and applicable to concrete cases and varying constellations. On the other hand, it is this indeterminateness of the norm that provides the space for security matters to embed themselves in the law. The law sees itself confronted with the ‘veridiction’Footnote ⁸¹ of a threat that can neither be ignored nor simply be disputed. The empirical norm thus tends to play off against the conventional understanding of the legal rule, which, in this case, concerns the question of self-defence in the face of an imminent threat.

If President Bush's launch of a fight against terrorism as an enduring threat has been reinforced by the Obama administration and therefore continues to be effective to the present day, this persistence is apparently driven by both actual evidence, namely the experience of a horrendous attack, and the presumption of a catastrophic threat that eludes determination. The problem with the ‘unknown unknowns’, as Donald Rumsfeld infamously identified the new dimension of threats in the wake of the 9/11 terrorist acts, is that an absence of any evidence of threats is not evidence of absence of threats.Footnote ⁸² It is particularly within the rationality of pre-emption that a critical link can be drawn between an individual target and a general threat, since the presumption of an unknown threat causing intolerable harm is irrefutable, while the suggestion is that there is an ‘overwhelming reason to believe that al Qaeda intends further attacks’.Footnote ⁸³ The crucial criterion the legal discourse has brought into play, thereby circumventing the notoriously difficult one of imminence,Footnote ⁸⁴ is the ‘gravity of the threat’ international terrorism poses, as Daniel Statman suggests: ‘A war of terror does not mean that all citizens are under actual attack all the time, but that such attacks are frequent enough and devastating enough to make life unbearable.’Footnote ⁸⁵ Targeted killing is inscribed into the law in so far as a temporal criterion (imminence) is converted into a qualitative one (gravity). The target, as the one who may be lawfully killed, thus at the same time sees itself included in and excluded from the law. If targeted killing was thought to be unlegalizable – not in the face of a superior norm, as Dyzenhaus has it,Footnote ⁸⁶ but rather given that the principle of the use of force as a last resort is indeterminable in advance precisely because it is always bound to a singular situation requiring an individual decision – here, the reverse has been proved. The unknowable threat that, by definition, only materializes in a radically singular, namely unforeseeable, constellation has been rendered accessible to governmental intervention through a novel legal determination.

The new security dispositif that becomes inscribed into international law to assist with targeted killing operations, appearing to be the appropriate solution to the problem of transnational terrorism, can now be characterized as follows: it rests not only upon a precautionary principle that aims at forestalling intolerable harm, but also on the notion of terror networks that relocates the traditional concept of a confrontation between states. As a focus of intervention, the territorial dimension fades away in favour of the alleged core figures of the network. If the new security dispositif thus paves the way for compliance with the jus ad bellum requirements, the effort to achieve compliance with the jus in bello principles, as we will see, indeed invokes the law and constitutes legal knowledge, and thus contributes to targeted killing's legalization.

5. Legal reasoning and law-making

The legal debate on targeted killing, particularly that referring to the US practice, has increased immensely during the last decade and even more so very recently, obviously due to a ‘compulsion of legality’.Footnote ⁸⁷ Once this state practice of resorting to the use of lethal force has been recognized as systematically taking place, it needs to be dealt with in legal terms. Whether this is done in supportive or critical terms, the assertion of targeted killing as a legal practice commences at this point. This is due to the fact that the law, once invoked, launches its own claims.

To insist on disclosing ‘the full legal basis for targeted killings’; on criteria, legal procedures, and ‘access to reliable information’ in order to render governmental action controllable; or on legal principles to be applied in order to estimate the necessity and proportionality of a concrete intervention at stake,Footnote ⁸⁸ not only involves accepting targeted killing as a legitimate subject of debate in the first place. It requires distinctions to be made between, for example, a legitimate and an illegitimate target. It invokes the production of knowledge and the establishment of pertinent rules. Indeterminate categories are to be determined and thus established as a new reading of positive law. The introduction of international human rights standards into the debate, for example, clearly allows limits to be set in employing the pre-emptive tactic. As Wouter Werner has shown with regard to the Israeli High Court of Justice's decision on the legality of targeted killing operations,Footnote ⁸⁹ this may well lead, for example, to recognizing the enemy as being not ‘outlaws’ but, instead, combatants who are to be granted basic human rights. Subsequently, procedural rules may be established that restrict the practice and provide criteria for assessing the legality of concrete operations.Footnote ⁹⁰ At the same time, however, targeted killing is recognized as a legitimate tactic in the fight against terrorism and is being determined and implemented legally.Footnote ⁹¹

When framed within the ‘theatre of war’, targeted killing categorically seems to be justifiable under the legal principles of necessity, proportionality, discrimination, and the avoidance of unnecessary suffering. This is true as long as one presupposes in general terms, as the juridical discourse usually does, both a well-considered proceeding along those principlesFootnote ⁹² and, accordingly, that targeted killing, by its very nature, is a ‘calculated, precise use of lethal force’.Footnote ⁹³ Procedural rules, like the ‘proportionality test’, that are essentially concerned with determination, namely with specifying criteria of intervention for the concrete case or constellation, certainly provide reliability by systematically inciting and provoking justifications. Their application therefore, we may say, contributes to clarifying a controversial normative interpretation, but it will never predict or determine how deliberation and justification translate into operational action. The application of procedural rules does not only notoriously remain ‘indeterminate’,Footnote ⁹⁴ but also produces its own truth effects. The question of proportionality, for example, is intrinsically a relational one. The damage that targeting causes is to be related to the anticipated military advantage and to the expected casualties of non-targeted operations. Even if there are ‘substantial grounds to believe’ that such an operation will ‘encounter significant armed resistance’,Footnote ⁹⁵ this is a presumption that, above all, entails a virtual dimension: the alternate option will never be realized. According to a Foucauldian perspective, decisions always articulate within an epistemic regime and thus ‘eventualize’ on the political stage.Footnote ⁹⁶ There is, in this sense, no mere decision and no mere meaning; and, conversely, there is no content of a norm, and no norm, independent of its enforcement.Footnote ⁹⁷ To relate this observation to our problem at hand means that, rather than the legal principles’ guiding a decision, it is the decision on how to proceed that constitutes the meaning of the legal principle in question. The legal reasoning, in turn, produces a normative reality of its own, as we are now able to imagine, comprehend, and assess a procedure and couch it in legal terms.

This is also noticeable in the case of the Osama bin Laden killing. As regards the initial strategy of justification, the question of resistance typically is difficult to establish ex post in legal terms. Such situations are fraught with so many possible instances of ambiguous behaviour and risk, and the identification of actual behaviour as probably dangerous and suspicious may change the whole outcome of the event.Footnote ⁹⁸ But, once the public found itself with little alternative but to assume that the prospect of capturing the subject formed part of the initial order, it also had to assume that the intention was to use lethal force as a last resort. And, once the public accepts the general presumption that the United States is at war with the terrorist organization, legal reasoning about the operation itself follows and constitutes a rationale shaping the perception of similar future actions and the exercise of governmental force in general.Footnote ⁹⁹ Part of this rationale is the assumption, as the president immediately pointed out in his speech, that the threat of al Qaeda has not been extinguished with bin Laden. The identification of a threat that emanates from a network may give rise to the question of whether the killing of one particular target, forming part of a Hydra, makes any sense at all.Footnote ¹⁰⁰ Yet, it equally nourishes the idea that the fight against terrorism, precisely because of its elusiveness, is an enduring one, which is exactly the position the United States takes while considering itself in an armed conflict with the terrorist organization. Targeting and destroying parts of a network, then, do not destroy the entire network, but rather verify that it exists and is at work. The target, in this sense, is constituted by being targeted.Footnote ¹⁰¹ Within the rationale of the security dispositif, there continue to be threats and new targets. Hence, at work is a transformation of laws through practice, rather than their amendment.

Giorgio Agamben maintains that a legal norm, because abstract, does not stipulate its application.Footnote ¹⁰² ‘Just as between language and world . . . there is no internal nexus’ between them. The norm, in this sense, exists independent of ‘reality’. This, according to Agamben, allows for the norm in the ‘state of exception’ both to be applied with the effect of ‘ceasing to apply’Footnote ¹⁰³ – ‘the rule, suspending itself, gives rise to the exception’Footnote ¹⁰⁴ – and to be suspended without being abolished. Although forming part of and, in fact, being the effect of applying the law, the state of exception, in Agamben's view, disconnects from the norm. Within a perspective on law as practice, by contrast, there is no such difference between norm and reality. Even to ignore a pertinent norm constitutes an act that has a meaning, namely that the norm is not being enforced. It affects the norm. Targeted killing operations, in this sense, can never be extra-legal.Footnote ¹⁰⁵ On the contrary, provided that illegal practices come up systematically, they eventually will effectuate the transformation of the law. Equally, the exception from the norm not only suspends the norm, transforming it, momentarily or permanently, into a mere symbol without meaning and force, but at the same time also impinges upon the validity of that norm. Moreover, focus on the exception within the present context falls short of capturing a rather gradual transitional process that both resists a binary deciphering of either legal or illegal and is not a matter of suspending a norm. As practices deploying particular forms of knowledge, targeted killing and its law mutually constitute each other, thus re-enforcing a new security dispositif. The appropriate research question therefore is how positive law changes its framework of reference. Targeted killing, once perceived as illegal, now appears to be a legal practice on the grounds of a new understanding of international law's own elementary concepts. The crux of the ‘compulsion of legality’ is that legality itself is a shifting reference.

Seen this way, the United States does not establish targeted killing as a legal practice on the grounds of its internationally ‘possessing’ exceptional power. Rather the reverse; it is able to employ targeted killing as a military tactic, precisely because this is accepted by the legal discourse. As a practice, targeted killing, in turn, reshapes our understanding of basic concepts of international law. Any dissenting voice will now be heard with more difficulty, since targeted killing is a no longer an isolated practice but, within the now establishing security dispositif, appears to be appropriate and rational. To counter the legal discourse, then, would require to interrupt it, rather than to respond to it, and to move on to its political implications that are rather tacitly involved in the talk about threats and security, and in the dispute about targeted killing operations’ legality.