2.1. The drone actant

The idea that that material objects such as drones may have actor-like qualities has a long pedigree. Pace Latour, it is not confined to ‘premodern’ thinking.Footnote ¹² Rather, thinking about material agency has been alive and well not only in anthropology and feminist studies but also in the disciplines Latour is most harshly chastizing for forgetting it, namely sociology and philosophy.Footnote ¹³ This said, there can be little doubt that the interest in material agency has seen a strong renewal in all areas of the social sciences as captured by the growing number of references to ‘post-humanism’, and ‘new materialism’ and by the impact left by Latour's own work.Footnote ¹⁴ Two recurring ideas drawn from the resulting debates are particularly important for approaching the agency of drones.

The first is that technologies are not merely passive tools in the hands of those who use them but have agency. As Latour puts it, technologies and things more generally ‘can authorize, make possible, encourage, make available, allow, suggest, influence, hinder, prohibit and so on’.Footnote ¹⁵ They have performative effects that escape human intention and control. This is the core contention in Latour and Woolgar's study of ‘Laboratory Life’, which argues that we need to understand ‘hybrid’ (social/material) relations if we want to understand science and its development.Footnote ¹⁶ But technological agency is not only of relevance to the hard sciences. Kittler has shown that also language, music, art, and literature are created through evolving technological agency. He has, for example, demonstrated the role of the typewriter in co-creating literature.Footnote ¹⁷ Analogously, in security studies, Aradau, following Barad, suggests that we need to study the material/social ‘intra-actions’ to grasp how the security of critical infrastructure has developed.Footnote ¹⁸ In all these renderings of hybrid agency the technological is not simply an extension or mirror of human social practices. Even if a technology has an origin and was intended for a particular purpose, this does not confine it to that origin or that purpose. Just as discourses and ideas transform and diffuse, so do technologies. Technologies ‘travel’, ‘translate’, ‘make associations’, ‘enrol’, ‘rhizomatically spread’, and become ‘boundary objects’.Footnote ¹⁹ In the process the technologies may retain traces of their origins and excavating these may be important to understand their nature and consequences. However, their role and agency are not reducible to their origin any more than is the agency of a person. In that sense, scholarship focused on fetishes provides useful insights about material agency (such as that of drones). It recalls that the role of objects and their agency may have little to do with their origins, which are often forgotten, and everything to do with the way they have been woven into practices to the point where they become essential sources of authority and objects of veneration in their own right.Footnote ²⁰

The second recurring idea is that technologies generate meaning. It is not only through discourses and social practices that meaning is reproduced but also through material agency.Footnote ²¹ Indeed, technologies generate classifications, ways of seeing and ways of presenting. They are at the origin of (re)interpretations and representations of things. The microscope produces a representation of the world and the algorithm is at the origin of data classifications. Latour and Woolgar (as many others) draw on this to underscore the absolutely central place of instruments of observation and measurement in the ‘construction of the scientific fact’.Footnote ²² They insist that the material has a role which is ‘symmetrical’ to that of language and practices in the generation of meaning. Meaning then ceases to be an exclusively human-based notion and instead becomes something reproduced also by the material.Footnote ²³ The material may hence be ‘performative’ in the sense of generating new subjectivities and imaginaries.

Nothing in the above suggests that material agency implies that things/machines are ‘human’ or that they are becoming more so. On the contrary, material agency should be understood on its own terms and in context. Latour therefore uses the term ‘actant’, borrowed from semiotics, to distinguish the technological agent from the human actor. Aradau (echoing Barad) insists on the term intra-action (to capture the hybrid material–human relation). Kittler is explicitly critical of the fact that ‘people are far too anthropomorphic when it comes to discussing technology’.Footnote ²⁴ This point has bearing for the omnipresent argument that as long as drones do not act as humans there is no need to worry about them; they are ‘just’ weapons. Understanding material agency from the vantage point just introduced, as the performative effects of technology in a context, underscores that drones do not need to become humans to ‘do’ things any more than do Latour's microscopes, Aradau's critical infrastructure, or Kittler's typewriters.Footnote ²⁵

2.2. Drones in the field of legal expertise

If drones can be actants in the dual sense just suggested (of having agency and of generating meaning) it is reasonable to wonder how they may be intervening with the many practices that constitute them; that is, how drones are involved in co-constitutive processes. How, for example, do drones intervene with the legal expertise constituting the legal status of the US drone program? To look for an answer, it is useful to think in terms of field analysis. Field analysis has often been used to analyse co-constitutive processes, such as the ‘mysterious’ ones by which the minister creates the community s/he represents.Footnote ²⁶ It usefully highlights first the contested character of legal expertise and second the mechanisms by which this expertise is likely to evolve.

A field is a social space held together by a stake at stake, such as the disagreement over what legal expertise is and who can (therefore) claim to be an expert.Footnote ²⁷ The practices surrounding this stake at stake reproduce understandings of what is valuable − which resources (or capital) and what dispositions (or habitus) − and hence a hierarchy based on ‘symbolic power and violence’.Footnote ²⁸ The more natural and uncontroversial the hierarchy the more stable it is likely to be. If gendered or racial (dis)positions (or the legal expertise constituting the drone program as an object of legal regulation) are natural and uncontroversial, the hierarchies on which they rest will be more stable than if they are controversial and contested. The very existence of a field requires a degree of misrecognition (an illusio) of the extent to which the stake at stake (for example of legal expertise) is itself a historically situated construct imbued with power. Much field analysis has therefore been geared towards looking at the historical emergence and stabilization of fields, including the ‘juridical field’.Footnote ²⁹ However, field analysis is not (contrary to what is often contended) a static theory of ‘domination’.Footnote ³⁰ It is geared to diminish misrecognition and hence pave the way for contesting relations of symbolic power and violence.Footnote ³¹ It does so by focusing both on the historical constitution of the field's illusio and on its reproduction. The former focus directs attention towards the existence of processes of co-constitution in general (legal expertise and drones co-constituting each other). The latter moves on to questions about which (of the possible and hierarchically related) form a specific co-constitution takes. As such, field analysis remains open to the possibility that legal expertise and the drone program may not co-constitute each other so as to generate a ‘targeted-killing dispositive’ (Krasmann's expressionFootnote ³²) in singular and hence may not be trapped in the kind of ‘legitimacy paradox’ discussed by de Goede and Sullivan in this issue. Rather, field analysis would see the precise nature of the co-constitutive relationship as defined by the hierarchical relations between the multiple and evolving legal expertise(s) and therefore as not only open but also potentially plural.

As a logical correlate of this emphasis on plurality, openness, and non-fixity, field analysis has an elaborate view of the mechanisms through which ‘fields’ and the struggles within them evolve and change. For the present analysis, change through agency is particularly central. The question is how the rhizomatic, boundary-object, association-making, travelling agency of drones refashions the field of legal expertise. On this account, two mechanisms of change appear particularly central. The first is changes to the boundaries of fields; altering the boundaries of who/what is included in a field also alters its dynamic and logic. Bourdieu's account of the expansion of the state in France, which eventually comes to shape all spheres, including the most intimate ones of personal life, love, and death as epitomized by the livret de famille (the French family registry) and the military service, is a case in point. Through the agency of bureaucrats as well as through the agency of statistical registrations and the spread of administrative, bureaucratic, and legal technologies the state comes to retain a ‘monopoly on symbolic violence’ which is far more central than its monopolization of physical violence and taxation (usually emphasized). The monopoly on symbolic violence is core to the other monopolization processes.Footnote ³³ The second mechanism is change within the field; that is, changes to the hierarchies of ‘resources’ (capital) and ‘dispositions’ (habitus). This type of mechanism is core to Bourdieu's analysis of the emergence and subsequent changes in the literary field in France. New ideas about place and standing of literature promoted by (the agency) of authors such as Baudelaire and Flaubert generated a space for artistic literary production proper in the mid-nineteenth century.Footnote ³⁴ Thinking analogously about how the agency of drones refashions legal expertise, in other words, leads to concentrating on how this agency redraws field boundaries and the hierarchies of valuation in the field. Of course, these processes are bound to be intertwined; reshuffling what is valued also changes who is an actor and vice versa, as also confirmed in Bourdieu's analysis of the state and of the literary field. Separating the two core mechanisms is therefore entirely artificial. However, it is a useful heuristic move focusing attention on one mechanism at a time. It will be used in this way in the analysis below.

To sum up, the analysis below traces the actant role of drones in legal expertise with the help of the theoretical building blocks just introduced. It focuses on how the drones ‘act’ (have agency and generate meaning) in ways changing the field of legal expertise through the two mechanisms of change just outlined (change of field boundaries and change within fields). The argument, in other words, departs from the questions Michael Lynch considers fundamental to the study of expertise: ‘who's in the field and what are they? . . . and which fields [areas of expertise] are relevant?’Footnote ³⁵ However, it moves on to also explore how the answers reshape the hierarchy of expertise(s) and hence how legal expertise and the drone program co-constitute each other.

3.1. Drones making associations with legal experts

Drones play a role in pulling in and consolidating the role of legal expertise that has no obvious (or even currently accepted) place in the discussion about the targeted-killing/extrajudicial-execution status of the drone program. The possibly most obvious case in point is the way in which the agency of drones has been core to bringing a range of civil-society associations such as Dronewars in the UK (at the forefront of the discussion about the legal status of the drone program in Europe) or the American Civil Liberties Union (which played a core role in the Awlaki caseFootnote ³⁶) and made them engage as legal experts. However, the discussion below illustrates the argument with two other associations, namely those drones have made with the legal expertise of intelligence services and of commercial contractors. I focus on these partly because they are less obvious but also because they are ‘harder cases’ for the argument that drones and legal expertise co-constitute each other, in a way that reproduces an understanding of the drone program as a program of extrajudicial assassination and not only as one of targeted killing.

Drones have generated associations with the legal expertise related to (civilian) intelligence. Drones have not only given intelligence a far more central role but blurred the line separating it from the armed forces.Footnote ³⁷ In fact, the blurring seems to have become so normalized that lawyers working in the respective institutions see little difficulty in crossing the boundaries. Hence, CIA director Panetta (presumably on the basis of the legal expertise of the agency) did not seem to find it problematic to publicly state that he had been in charge of Operation Neptune Spear killing Bin Laden (carried out by Navy SEALs).Footnote ³⁸ Nor does the director of the National Counterterrorism Center seem to find it problematic to publicly state that it does not really matter whether an operation is carried out under the authority of the armed forces, the CIA, or the JSOC but that who is in charge should depend on what is legally most advantageous.Footnote ³⁹ As this indicates, there may be a number of reasons to wish and push for the blurring of lines, including precisely the limiting of legal oversight and accountability. This said, ‘technological trends drive convergence’.Footnote ⁴⁰ Drones change the relationship between intelligence gathering and military operations. Intelligence, surveillance, and reconnaissance have come to play a central role in directly supporting and organizing the operations of armed forces on the ground to the point where many suggest that they have restructured the way US armed forces are organized and work.Footnote ⁴¹ By the same token they have increased the scope and relevance of the intelligence and paved the way for the current trend to form hybrid (military–civilian) teams and organizations. While this operational closeness has not wiped out differences or competition between these institutions, it has made them draw upon and cultivate the same kinds of legal expertise. It hence seems not only warranted but progressive to deplore the superficial training and relationship of (civilian) intelligence institutions to international humanitarian law and the law of armed conflict.Footnote ⁴² Drones have, in other words, made associations that blur the lines between the military and the intelligence services in the US. Drones have brought the civilians into an area of military expertise and, in the process, constituted civilians as actors in the field of military legal expertise. They have hence shifted who can take part as an expert of what kind of law in the disagreement surrounding the legal status of the drone program, hence shifting the boundaries of this field.

Second, and similarly, drones have shifted the boundaries of the field of legal expertise by making associations bringing commercial contractors into it. Companies have a role in developing drones, in training those using them, and in providing maintenance and support.Footnote ⁴³ This presence of companies is usually presented − by the companies as well as by observers − as a matter of technical competence and procurement with little or no relation to the more controversial deployment of drones.Footnote ⁴⁴ However, this ‘technical’ presence brings contractors closer to the drone program than is usually acknowledged. Technology development in relation to drones is not something merely done in the lab, before technology is used. Rather the demand for drones has been such that they have been deployed ‘without lengthy testing procedures’ and that they are ‘developed and tested in use’.Footnote ⁴⁵ The implication is that the commercial contractors remain involved in the drone program and that ‘training’ assumes a long-term, continuous character. ‘Training’ no longer takes the form of a course in which contractors communicate the necessary knowledge once and for all and then disappear. To this one should add that manning drones and training enough pilots has been a recurring problem that has been solved by engaging contractors to fill the gaps.Footnote ⁴⁶ All in all, it is therefore not surprising to find that those responsible for flying drones think their operations would collapse without contractors and that they often are uncertain about the professional status (civilian or military) of a person responsible for a specific task.Footnote ⁴⁷ In this situation, one would expect things to occasionally go wrong, thus forcing the legal expertise of these contractors into the heart of the drone program. This expectation is borne out, for example, in a 2010 story published by the Washington Post about a drone that mistakenly killed ‘at least 15 civilian Afghans’,Footnote ⁴⁸ including women and children. The Washington Post informs its readers that the army officer investigating the incident ‘was surprised’ to find out that the woman responsible for misinterpreting the video feed worked for a company (SAIC).Footnote ⁴⁹ This information (obtained under the Freedom of Information Act and echoed across the US press) sums up what is available about the incident. The SAIC incident may not have been linked to targeted killings/extrajudicial executions. There is no way of finding out. However, it confirms the closeness of contractors to the drone program. This closeness, in turn, entails unwritten or written rules specifying (including by limiting) the role of contractors in relation to military operations. These rules participate in forming the understanding of the legal status of the drone program and of contractors within it. How exactly is impossible to say, as this legal expertise is made invisible to the public.

For the argument here, what matters is its consolidation and presence. Drones have pulled in and consolidated the presence of legal expertise linked to commercial contracting just as they have with the legal expertise of civilian intelligence and issue-based NGOs. In that sense, the agency of drones contributed to the redrawing of the boundaries of legal expertise involved in constituting the legal status of the drone program.

3.2. Drones providing legal expertise

A second, rather more provocative, thought is that drones may shift the boundaries of the field of legal expertise by creating space for technological forms of legal expertise that originate from the drones themselves. However, this thought deserves serious consideration. Drones have come to occupy positions as experts partly by inadvertence and partly because they are placed there.

Drones have come to offer their legal expertise partly because the questions regarding who, in practical terms, ends up making legal evaluations when drones are involved have not received the attention they deserve. The official position in all countries using drones (and not only the US) is that there will always be a ‘human in the loop’, a person who must consult relevant legal expertise if s/he does not have it.Footnote ⁵⁰ However, in practical terms the human in the loop may be so far removed that the legal evaluations end up resting with the drones. Who, for example, is responsible for the legal judgement and evaluation when drones misinterpret situations and action is taken on this basis as happened when two marines returning to their base were killed because their way of advancing was deemed threatening on the basis of images produced and classified by drones?Footnote ⁵¹ Similarly, what happens when there is an interaction of codes or systems that works in unexpected ways, creating ‘emergent problems’ (problems rooted in the system as a whole)?Footnote ⁵² It could be argued that responsibility rests with the person(s) who wrote the codes or planned the systems. Alternatively, it could be located with person(s) using them in ways or contexts they were not intended for. These people could then be held accountable for their insufficient mastery of relevant legal expertise or for their unwillingness to refer to such legal expertise. However, the loop is long and filled by very reasonable arguments for shuffling responsibility elsewhere and especially onto the technology itself. An illustration of what is involved is provided by the litigation between two companies where one, Integration Systems Inc. (IISi), alleged that the other, Netezza, had reverse engineered its software coding and sold it to the CIA for use with Predator drones where, according to IISi, it could not work because of systems incompatibility issues. A core worry (and motive for the litigation) as expressed by IISi's chief technology officer was liability ‘in case that code kills people’.Footnote ⁵³ Even if the phrasing ‘in case that code kills someone’ was used figuratively, it indicates the sense of the technology officer that the code's incompatibilities would be acting on their own, escaping the control of its originators. Whether or not this would lead to the killing of people, it indicates the officer's sense that codes increasingly become independent, and hence practically speaking responsible, potentially also for judgements that normally would involve legal expertise. This is neither because the chief technology officer, as the originator of the code, intended this, nor because those who reverse engineered the software or those who employed the drone intend to grant it this responsibility. Therefore none of these people would be likely to accept that they were responsible; instead judgements involving legal expertise would be located with the drone. This example shows that the shift displacing legal expertise to drones, or ‘a code’, as the chief technology officer puts it, is an outcome of the way technologies move and are used for multiple purposes. It has little relation to a reflected displacement of legal expertise from the human ‘in the loop’ to the technology and much to do with inadvertence.

Second, drones are also invited in a more reflective and conscious manner to provide legal expertise. The reason is the increasingly common argument that allowing the technology to make the judgement is better. One of the reasons used to justify the (already ongoingFootnote ⁵⁴) use of automated weapon systems is that these systems can and do make judgements in situations where ‘the humans in the loop’ cannot make them, for example because of technical problems with signalling, because the required reaction time is too short, or because the information to be analysed is too massive and too complex (there is a ‘data crush’ and/or a ‘skills lag’).Footnote ⁵⁵ It is hence suggested that not relying on the automated judgements would worsen the protection of civilians and increase collateral damage.Footnote ⁵⁶ To this is added that, unlike people, technology systems are not emotionally engaged and hence do not react out of fear, anger, exhaustion, or despair. On the contrary, they can be programmed to be self-sacrificing.Footnote ⁵⁷ The argument is often pushed further to suggest not only that it is appropriate to leave some important judgements to technology at present, but that even more should be left to them.Footnote ⁵⁸ The hesitation to move down this route is correspondingly presented as resting on an emotionally and irrationally grounded fear of technologies taking over; a matter of ‘domestic and diplomatic sensitivities’ or, more banally, ‘misunderstandings’ that will diminish over time.Footnote ⁵⁹ Certainly, the resistance against a greater autonomy seems to be receding. In the UK the idea of building the legal review process into the machines is already debated.Footnote ⁶⁰ Similarly, the connotation that technology really ultimately knows best is frequently mobilized in current speak. For example, already in 2009, a

The emphasis put on the expertise contained in the algorithm may be a figure of speech, not a statement of how targeting decisions are made. However, even if it is, it is indicative of the willingness to rely on technologies for expertise on complex legal issues related to the targeted-killing program. This acceptance of drones as relatively independent providers of legal expertise is no less contested than is their role in hybridizing legal expertise.Footnote ⁶² As has been intimated throughout the preceding discussion, there is no shortage of suggestions that their role is unwarranted, in need of regulation or even banning.

Disagreements over the boundaries of a field are integral to the struggles in any field. As just shown, drones intensify these disagreements. They make associations that constitute, consolidate, and strengthen the weight of civilian intelligence and commercial company-related legal expertise in the disagreements surrounding the drone program. They also alter the boundaries by offering up their own legal expertise. This dual move may appear as a recipe for bolstering the targeted-killing position which treats the drone program as falling within the realm of legality. However, this is not necessarily so. Not only is the presence of intelligence, commercial, and technological actants mobilized to bolster the extrajudicial-execution position treating the drone program as entirely outside legality by those who consider this presence itself as proof of the extrajudicial nature of the program. The extrajudicial-execution position may also be adopted from within the legal expertise tied to intelligence and commercial actors. In intelligence, there are recurring indications of dissent about the drone program as became clear (again) in the worry that Petraeus’s successor might try to cut it.Footnote ⁶³ Similarly, commercial actors may also adopt an extrajudicial-execution position on the drone program. It removes the suspicion that companies are involved in activities of dubious legal status. By the same token, an extra-legal position can consolidate the reputation of the drone industry, which has been tarnished by the controversies surrounding the drone program.Footnote ⁶⁴ Finally, technological expertise may well be more readily received and integrated if paired with a clear extrajudicial-execution position. The way the agency of drones is redrawing the boundaries of the field of legal expertise, in other words, leaves considerable scope for controversy and politics. The associations they make and the expertise they provide do not unambiguously make the field slide toward a targeted-killing position, on the contrary. However, this may be of limited significance if drones weigh on the understandings of what is valuable in the field of legal expertise in ways that produce a systematic bias in favour of the extrajudicial-execution position. The next section therefore looks more closely at how the agency of drones has co-constituted the ‘rules of the game’ in the field.

4.1. Drones imposing transparency

Drones have made transparency figure centrally in relation to the struggle over legal expertise in relation to the drone program for two reasons. The knowledge drones generate and the organizational processes they trigger have both turned transparency into an inescapable issue.

Legal expertise surrounding the targeted-killings–extrajudicial-executions program rests heavily on arguments referring back to the transparency ensured by the technological knowledge generated by drones. The legal expertise mobilized to defend the legality of the drone program constantly refers to the high quality of the knowledge provided by the drones. On their account, the information generated by the ‘unblinking eye’ of the drone is superior to that generated by the human eye, notably because the drone needs no sleep and therefore can generate continuous massive flows of information. The drone can also perceive in detail, over distances, in conditions, and over time in ways that the no human eye possibly could. The drones make the otherwise complex and opaque picture of a faraway reality, possibly clouded by the ‘fog of war’, translucent and graspable to an extent normally impossible. For the same reason drones create the possibility of acting in this reality with ‘surgical’ precision.Footnote ⁶⁹ The emphasis on the transparency provided by the technologically generated knowledge is taken to its extreme in the relation to the so-called ‘personality’ or ‘signature’ strikes where decisions to strike a target are founded on the analysis of the behaviour (personality or personal signature) of people.Footnote ⁷⁰ Inversely, precisely because it plays such a central role in the argumentation, legal experts striving to demonstrate the extrajudicial status of the drone program are intent on demolishing this image of transparency. They hence recall that technologically generated knowledge may not be the main, let alone only, form of knowledge drawn upon. Instead they underscore the role played by conventional informants and point to the ‘multiple reports’ of cases where people have misinformed about their enemies with fatal consequences.Footnote ⁷¹ More centrally, they also challenge the infallibility of technological knowledge. They recall and insist on the many recurring problems of technologically generated knowledge. Common issues in this regard include the difficulties involved in handing the complex and overwhelming amounts of information that lead to missing what is significant and misunderstanding that which is not missed. They will also point to the problems such as those created by ‘system incompatibilities’ (which blur images), by the ‘soda straw effect’ of drone knowledge (one can only see the area covered by the censors which is as narrow as a ‘soda straw’), and by time-lag effects that may result in an inability to adjust to changes (such as the appearance of civilians close to a target or the disappearance of the target).Footnote ⁷² The overall image the critics present of technologically generated knowledge is far removed from the one allowing for all-encompassing transparency. It is an image of partial, confused, and imperfect understanding; a reality that remains opaque. Rather than a transparent world, it is an inscrutable uncertain one, analogous to the one Wouter Werner finds in the Tallinn Manual in this issue. The energy that goes into constructing this image as well as its opposite of perfect transparency shows the extent to which transparency has become core to bolster legal expertise.

The centrality of transparency to legal expertise has been further accentuated by the organizational processes drones have generated. By contrast to conventional aircraft flown by a single person in often strained physical and emotional conditions, drones are operated by teams of people who can take turns and consult each other in relative comfort. Over 150 people are involved in flying Predators and Hawks. The video feeds from drones are digital and can be stored.Footnote ⁷³ The likelihood that knowledge can/will remain the preserve of one person or even a small group diminishes drastically. Legality itself has come to resemble a drone hovering over all drone-related activities, as a journalist puts it.Footnote ⁷⁴ On this account, drones have increased the role of law in warfare. As one observer put it, drones have made ‘laws, rules, and regulations pervade almost every aspect of the modern U.S. military institution . . . war itself has evolved into a legal institution, lawyers have predictably become a common fixture in all of its dimensions’.Footnote ⁷⁵ Reportedly lawyers are reviewing and approving the processes in the US drone program.Footnote ⁷⁶ Those adopting a targeted/killing position on the drone program highlight this pervasive presence of law to argue the increasingly stringent standards, as well as an increased pressure to live up to these.Footnote ⁷⁷ In their view, the organizational transparency generated by drones ensures not only that the drone programs are legal but that they are more legal than conventional forms of warfare. Inversely, those adopting an extrajudicial-execution position reverse the claims. According to them, far from generating transparency, the multiplicity of actors with their overlapping competencies, relying on ill-specified co-ordination, and hampered by institutional competition, generate opacity and confusion. Involving legal expertise merely accentuates this, as it is co-opted and instrumentally manipulated to mask mistakes and pre-empt legal accountability. A ‘no blood, no foul’ poster hanging in a site jointly operated by the CIA and JSOCs is taken as an iconic illustration of the relationship with law of the drone program.Footnote ⁷⁸ On this account, far from generating transparency, the organizational transformations engendered by drones have undermined it.

Drones have, in sum, imposed transparency as a core resource and disposition for legal expertise pertaining to the drone program. They have done this through the forms of technological knowledge and the organizational process they generate and in a way that affects legal expertise whether articulated to defend or contest the targeted-killing position on the drone program. This might appear disconcerting to those who assume that transparency mainly weighs in favour of those contesting the secret services and the security establishment. The next section is bound to reinforce this sense of consternation. It argues that the technological agency of drones has made secrecy increasingly central, suggesting that implications for whether the drone program is constituted as one of targeted killings or of extrajudicial executions is analogously open.

4.2. Drones steadying secrecy

Along with transparency, drones have also made secrecy increasingly central to the rules of the game in the field of legal expertise. Transparency and secrecy consolidate and define each other. They have a symbiotic relationship as the one presupposes the other. In relation to the drone program, mobilizing secrecy is part and parcel of giving weight to virtues of transparency that the justification of legal expertise in relation to drones relies on.

Secrecy is one of the most potent ways of contesting the credibility of the legal expertise defending the legality of the drone program; it empties the arguments of their substance. Drones might have created a graspable and transparent world that can be acted upon with precision in ways enhancing legality. However, when no information is provided about the legal argumentation, the evidence gathered, or the legal procedures involved, such claims lose their credibility. There is just no way of verifying. It is therefore not surprising to find a critique of the way ‘due process’ is (not) provided in the drone program couched as a plea to ‘lift secrecy’ and ‘clarify’ the procedures and standards of evidence used.Footnote ⁷⁹ In harsher terms, critics of the program, see secrecy as nothing more than a thin disguise for what is, at best, a wish to evade legal discussions. Hence, the proposal for a regulatory mechanism for the ‘targeting of citizens’ is presented as an attempt to counter the executive's wish to preserve the secrecy surrounding the drone program or its demand for a legal ‘blank check’ to cover its operations.Footnote ⁸⁰ Finally, and most harshly, the extensive reliance on secrecy is sometimes couched as a thinly veiled cover-up which constitutes an infringement of existing law and/or the development of appropriate regulatory procedures and practices. Along these lines, Cole argues that:

Cole concludes with no less than two full paragraphs of strictly legal questions that would require answers but can and do remain unanswered because of the reference to secrecy.Footnote ⁸²

It would, however, be profoundly misleading to conclude that secrecy can only be mobilized to consolidate the authority of legal expertise contesting the targeted-killing position on the drone program. It is also pervasively mounted in its defence. A recurring argument is that the nature of the knowledge and the kind of processes generated by drones make secrecy warranted. Wise judgement about what to make publicly available is a necessity. Because drones make reality so transparent and governable, the knowledge provided by them must also be carefully guarded. Openness might undermine the efficiency of specific military/intelligence operations, including those tied to the drone program, for example by making it possible for those targeted to avert the strikes or mount counteractions. Moreover, it creates the risk that images are used by ‘terrorist groups and rogue states’ to provide ‘evidence of alleged law-of-war violations as part of a growing practice that some describe as “lawfare”.Footnote ⁸³ Not respecting secrecy might in fact become a source of danger in its own right. If more were known about the drone programs and the precise and detailed information gathered by the drones, it might jeopardize the operations and expose those involved, as well as civilians. Even more strongly, secrecy is mobilized as a source of expertise in its own right. Referring to secrecy is a way of affirming superior judgement about what can reasonably be made accessible; it is too complex, too sensitive for the uninitiated. Only insiders can judge. Hence, technological complexity is frequently used to dismantle arguments to an extent where a legal scholar feels compelled to protest that ‘there is nothing to be fascinated by; they [the drones] are only weapons’.Footnote ⁸⁴ This way of drawing on secrecy to affirm expertise parallels the way religious authority is established. Only the initiated can and should have access to knowledge and perhaps even the initiated may not fully understand it. It also dovetails nicely with a long tradition of political thought stressing that arcana imperii (secrets of empire) are essential for good government. Even if that tradition is currently unfashionable, its connotations of good governance can be and are mobilized when information is authoritatively presented as too sensitive to be made public.Footnote ⁸⁵ Secrecy can, in sum, also be mobilized as a weighty argument to reaffirm a targeted-killing position on the drone program.

To sum up, drones have co-constituted secrecy and transparency as essential resources and dispositions in the field of legal expertise. The technological knowledge and organizational processes generated by drones make it important for legal experts to mobilize and rearticulate secrecy and transparency. This holds for experts on both sides of the targeted-killing–extrajudicial-execution divide. The secrecy–transparency couple does not clearly favour one side over the other and is effectively mobilized by both positions. Who will be able to do this most effectively remains an open question.