A choking(?) engine of war: Human agency in military targeting reconsidered

Abstract

This article explores the question of human agency in military targeting. Targeting is one of the key drivers of war. When studied by academic disciplines, much interest has been devoted to the ethics and effects of military targeting. Less debated, but focused here, is the question of the conditions of human agency within military targeting. In the literature that does exist on this topic, there is a questioning of the traditional conception of human agency but at the same time a lack of closer conceptualisation of different kinds of articulations of human agency in the targeting process. In this article, we propose a recentring of human agency in critical scholarship on military targeting. With inspiration from Theodore Schatzki's work on ‘practice’, by analytically approaching targeting as a practice, and through various examples from Operation Iraqi Freedom, the article develops and illustrates a framework for the conceptualisation of human agencies in targeting. This framework distinguishes articulations of agency based on whether they furthered the (temporary) ordering of the targeting practice or challenged its internal organising elements. The study of military targeting is significant not least since the phenomenon is one of the key ‘engines’ and drivers of war's constant becoming.

Introduction

Military targeting is central to how wars unfold. It is one of the ‘engines’ of war. Without a continuous targeting process, any war could effectively come to a halt. From a military perspective, targeting is the process of making known, selecting, prioritising, and acting upon entities in the operational environment with the goal to reach certain ends.¹ For a critical scholarship on war and warfare, targeting is significant as it is key to the constitution of military power and since it, through its very function in military operations, is productive of military violence. Targeting does not express a given, pre-existing military power as much as it forms one of the conditions of possibility thereof.² Thereby it links closely to questions about the legitimacy and accountability of military operations and violence. In this way too, the critical study of targeting allows for a rapprochement to what Antoine Bousquet et al. called for ‘unbounded investigation into the emergent and generative character of war’, and for taking important epistemological and conceptual steps towards a deeper understanding of ‘war's incessant becoming’ in and through one of the key engines of contemporary war.³

Given its significance, it is noticeable that military targeting – with some key exceptions – has received relatively limited academic attention. When addressed, it is often from legal and ethical perspectives that focus on targeting's effects.⁴ Another debate, however, concerns human agency in targeting. This is the focus of the article at hand. In the following, we identify three research strands that in different ways inquire into the conditions and articulations of human agency during the operational and tactical phases of targeting – all in the midst of highly advanced military technological systems that, for many commentators, make human agency diffuse and hard to locate.⁵

Despite a relative richness of the aforementioned research strands and their questioning of traditional conceptions of human agency, we argue in this article the lack of closer theorisation and, specifically, conceptualisation of different articulations of human agency in military targeting. There is a conceptual neglect when it comes to mapping and understanding the multifaceted forms and articulations of human agency in targeting. Against this backdrop, we argue for a recentring of human agency in critical scholarship on military targeting. Given targeting's constitutive function for military power and for the production of military violence, it is precisely because human agency may be increasingly diffuse and hard to locate in the targeting process that it is vital that critical scholarship seeks to reveal it. Studying the ways in which human agency are articulated and can be conceptualised during the targeting process provides a key entry point to the constant becoming of war as well as to questions about accountability in military operations.

By drawing on yet also pushing extant research further, and with inspiration from Theodore Schatzki's work on ‘practice’, we analytically approach targeting as a practice and develop a framework for conceptualising articulations of human agency in the targeting process. Approaching targeting as a practice, in Schatzki's understanding, entails seeing the phenomenon as ‘an organised nexus of actions’ that through human doings may be maintained and reproduced yet at the same time more or less modified or transformed.⁶ That is, targeting consists both of a set of actions (or doings) and of a number of organising elements that keep the practice together. It is particularly when there are smaller or greater ruptures between the doings and the organisation of the targeting practice that articulations of human agency can be grasped and conceptualised, and – potentially – a new practice may be said to have emerged (this being always an empirical question). With the help of empirical snapshots and examples taken from operational and tactical joint targeting during Operation Iraqi Freedom (OIF), we propose a four-dimensional (not exhaustive) typology of distinct articulations of human agency, based on whether the agencies helped maintain and further the targeting practice as initially aligned with certain stated goals and modes of action, or modified and/or challenged some of its organising elements and thus produced internal discord. The ‘engine of war’ that targeting represents may in this sense be more or less choking.

Targeting during OIF is particularly interesting in this regard. Often considered an example of a highly technologised, automatic, rule-bound, and bureaucratised process, we show that it can be alternatively understood as a practice constantly in the making through the doings of humans and thereby subject also to frictions and challenges to its organising elements. The diffusion or disappearance of human agency in military targeting can thus never be taken at face value, especially not if critical scholarship wants to distance itself from the military's own idealised representation of targeting as a highly structured, uniform, and ‘clinical’ process.

The article is structured into four parts. The first briefly introduces military targeting and its main contemporary features, in a consciously sterile and standardised manner, which forms a first justification for why it is possible to consider this phenomenon a practice. The second part examines extant research on agency in targeting and accounts for the aforementioned argument. In the third part, we develop our main contribution in terms of a conceptual framework and recentring of human agency in the analysis of military targeting, drawing on Schatzki's work. Finally, we give a further justification for why targeting can be analytically approached as a (constantly emergent) practice and illustrate our argument and the usefulness of the framework through empirical snapshots of the targeting process during OIF.

We have chosen OIF for three reasons. The 2003 launch of the operation was preceded by a proclaimed ‘Revolution in Military Affairs’ (RMA). Advanced military information and communication technologies along with notions of speed and precision strikes were key to the development of a new operational ideal and targeting process.⁷ This development had begun (at least) with Operation Allied Force (OAF) in the former Yugoslavia and Operation Enduring Freedom (OEF) in Afghanistan, but became more pronounced during OIF.⁸ With OIF then, secondly, the targeting process had become increasingly technologised, formalised, automatic, standardised, rule-bound, and bureaucratised.⁹ This was significant also for military claims to legitimacy.¹⁰ Indeed, the US military considers OIF and its targeting process to have been highly successful.¹¹ In many ways, OIF represents a large-scale military operation in which it could be expected that (as extant research in various ways also suggests) targeting would form a uniform and standardised practice that leaves little room for human agency.

A third reason is the relatively large secondary material on targeting during OIF, particularly in the form of so-called ‘lessons learned’ reports produced by the military, think tanks, or individual scholars. It is often difficult for civilian researchers to gain access to classified targeting documents or travel to the locations in which wartime targeting occurs.¹² As a result, scholars have often relied on official doctrines in explorations of targeting during war and high-intensity operations. For our purpose, however, doctrines do not take us very far. What is needed, rather, are accounts of enacted targeting doings. While secondary materials such as lessons learned reports might offer somewhat ‘schematised’ accounts, they do provide an indirect account of enacted doings that enables a mapping and conceptualisation of agencies.

Introducing military targeting

Contemporary military targeting is described as ‘the deliberate application of capabilities against targets to generate effects in order to achieve specific objectives. It is about the application of means (weapons) of warfare to affect addressees (people or objects) using a variety of methods (tactics) that create effects contributing to designated goals.’¹³ Targeting thus links the ends and means of warfare. Today, targeting occurs in both specific theatres and beyond the confines of the battlefield. It involves classic kinetic lethal actions (such as the use of bombs, guns, torpedoes) as well as non-kinetic and non-lethal activities (such as financial effects and information operations).

Targeting's historical development is in many ways an embodiment of the intimate relationship between warfare and technology.¹⁴ The rise of aerial warfare, for example, dramatically altered the targeting process.¹⁵ In the latter half of the twentieth century, technological advancements continued to reshape targeting. Information- and weapons technology, for instance, increased the speed, accuracy, and destructiveness of firepower. While many commentators have expressed doubts over military ‘fantasies of high-tech omnipotence’,¹⁶ precision, and ‘ethical’ ways of doing war,¹⁷ information and communication technologies have continued to have a transformative impact on targeting.¹⁸

Moreover, the targeting process of Western militaries has turned increasingly formalised and bureaucratised, for example involving standardised procedures and vetting mechanisms.¹⁹ Contemporary targeting in the US military and for NATO consists of various steps and layers held together by a formalised targeting cycle.²⁰ The cycle provides an iterative procedure for targeting during an operation. The standard includes six mandatory phases: (1) analysis of Commander's intent, objectives, and guidance; (2) target development (including target analysis, vetting, validation, nomination, and prioritisation; (3) analysis of capabilities; (4) Commander's decision, force planning, and assignment; (5) planning and execution of the mission; and (6) assessment of the results.²¹ Further, throughout the targeting cycle (often referred to as ‘the kill chain’), a large number of actors, also non-military, are involved – such as policy analysts, weapons engineers, and intelligence officers.²² Military and civilian lawyers figure today heavily throughout the process.²³ Overall, the formalisation and bureaucratisation of targeting are meant not just to increase efficiency in an operation, but also to uphold legitimacy and legality.²⁴

Finally, the targeting process is typically divided into two categories: deliberate and dynamic. Deliberate targeting involves pre-planned and pre-authorised targets, and unfolds normally within 24–72 hours. Dynamic targeting occurs in a more compressed timeline during the battle (normally within 24 hours or less).²⁵ Recent targeting processes have increasingly been of the dynamic type, referred to by some commentators as the ‘compression of the kill chain’.²⁶ Dynamic targeting involves for example previously unknown, unexpected, or unanticipated targets that emerge during the operation. This is typically either a target that is considered time sensitive, and therefore requires a quick response, or Close Air Support (CAS), that is, when troops on the ground have encountered the enemy and request immediate air support. Craig Jones explains: ‘If a pilot has been briefed about a target before she or he steps into the aircraft, then it is a deliberate target. If a target appears after the aircraft has taken off and the pilot receives her or his orders while flying, then it is a dynamic target.’²⁷ Nonetheless, targets are to be vetted in both instances.²⁸

Summing up, and OIF is a good example of this, contemporary military targeting is on the one hand increasingly formalised and bureaucratised while, on the other hand, increasingly (but not solely) focused on dynamic targeting.²⁹ What this means, furthermore, is that while the targeting cycle and its organising elements – which we unpack further in the empirics – can be analysed as a particular yet constantly emergent practice, the aforementioned phases are the military's own idealised phases that represent how targeting should be executed. However, since targeting depends on actual doings, and not least given the shift towards dynamic targeting, the fog and messy realities of war and military operations often mean that phases overlap, that they play out in parallel, that some of the mechanisms are omitted, and/or that the organising elements are maintained and reproduced as well as challenged.³⁰

Uncovering military targeting

Academically, military targeting is often addressed from legal and ethical perspectives. For example, the respective targeting process during OAF and OEF brought about intense debates over what constituted legitimate targets.³¹ Recently, the use of drones for ‘targeted’ (or ‘extra-judicial’) killings has met strong criticism for providing (Western) states with a capability to survey, target, and kill individuals (rather than state forces) from afar.³² For Grégoire Chamayou, the use of drones produces a military vision that dehumanises the enemy and that enables an imperial policing that perpetuates global inequalities.³³ Relatedly, Maja Zehfuss and others have questioned the targeting rhetoric of ‘precision’ and ‘surgical operations’.³⁴ Of main interest here, however, is another debate: one that addresses the relation between military targeting and human agency.

Human agency in military targeting

Contemporary military targeting has turned into a global enterprise that besides specific theatres unfolds within a spatially and temporally unbounded battlespace. There is a sprawling literature that explores the conditions of possibility of targeting ‘assemblages’ (not least in a political-technological sense), and criticises their implications.³⁵ Whether or not explicitly, these works stress the significance of not falling back upon ‘technological fetishism’.³⁶ Likewise, works on the conditions of human agency in military targeting equally take this advice seriously. In the following, we explore three research strands that respectively explore the military targeting process at the operational and tactical levels of war from a perspective of human agency. While much academic attention has been directed at targeted killings executed by drones, it is important to remember that in large-scale military operations the targeting process is in itself very comprehensive. Drones are often working in parallel to ‘conventional strikes’ performed by a particular service branch, by service branches operating jointly, or by a multinational coalition. In this article, focus lies on human agency during the operational and tactical dimensions of joint military targeting and on conventional strikes too, rather than solely on warfare enacted by drones.

A first research strand focuses on the operational dimension of the military targeting process, exemplified by the work of Astrid H. M. Nordin and Dan Öberg. Nordin and Öberg discuss subjectification/objectification formations as these unfold within the highly bureaucratised process of targeting, which in turn prompt them to reconsider the ontology of war. For Nordin and Öberg, contemporary warfare has ‘become a reiterative process which strictly speaking lacks antagonistic engagement with “an enemy”’.³⁷ Ontologically, this means that targeting does not make sense from a perspective of critical war studies that conceive of war as generative of meaning, knowledge, and appearances.³⁸ Rather, contemporary targeting illustrates an ontological shift towards war as ‘disappearance’.³⁹ Targeting, more specifically, illustrates how war has turned into an endless processing, in which human subjects and the agencies of both targeteer and targeted disappear. The targeting process as such becomes the agent. The process absorbs and uses the people involved to the extent that agency disappears in ‘the excess of the process itself’.⁴⁰ As Nordin and Öberg say:

The target is a result of a medium that erases the very conditions of its appearance and forces it symbolically and literally to disappear as object. … When this is achievable in a seamless process – efficient, synchronised, integrated – by virtue of operational modelling, we have reached the perfect war.⁴¹

Richard Adams and Chris Barrie argue similarly in relation to the American covert drone programme. For them, the bureaucratisation and highly administered and organised character of drone targeting and its execution means that the act of covert killing has become impersonal and morally remote: ‘a pattern of killing reduced by “the system” to hum-drum routine.’⁴²

The bureaucratised, formalised, and seemingly seamless military targeting process is focused by a second research strand, which nevertheless also stress the process's embodied character and the multitude of people involved in it – from pilots, sensor operators, commanders, and planners, to intelligence officers, image analysts, and military lawyers. This strand has paid much attention to aerial targeting and/or remote warfare. Peter Adey et al., for instance, see targeting from above as an embodied, techno-cultural practice constituted by ‘emotions, affects, feelings and rationalities’, even forms of resistance.⁴³ As they further say, ‘this military theatre might appear comfortable and calm in relation to the messy and passionate world that it punishes’ yet ‘the assumption that passion must somehow lie outside of these apparatuses is no doubt problematic.’⁴⁴

Moreover, many works in this strand concern drone operators in remote warfare and the conditions of their activities.⁴⁵ Countering the notion of aerial targeting as predominantly bureaucratised, emotionless, and technocratic and thus as erasing subjectivities and agencies, Peter M. Asaro describes the newly formed subjectivities and emergent agencies that characterise the process:

A drone pilot, in consultation with the sensor operator and mission intelligence coordinator, must now consider whether a potential target is in fact a correct and valid target. The drone operators are much more aware of the complexities of making that judgment and the uncertainties inherent in it.⁴⁶

According to Asaro, drone operators suffer clear risks of post-traumatic stress disorder.⁴⁷ Part of the reason is that aerial targeting and drone warfare are characterised less by remoteness and distance, and more by a ‘special kind of intimacy’ built around a ‘hunter-killer’ relationship. Even if geographically distant, the people involved in targeting are closer than ever before to the ‘killing space’.⁴⁸

With the first two research strands addressing the characterisation of military targeting as bureaucratised and routinised, and reaching distinct conclusions regarding what the effects of this are on human agency, a third strand debates the conditions of agency within the technological and material configurations – or assemblages – that make up the targeting process. Today, ‘more-than-human’ assemblages made up of technological, biological, social, and material entanglements constitute the targeting process, both in its more conventional forms at the operative and tactical levels and in drone warfare.⁴⁹ Works in this tradition has paid attention to how material entities in the targeting process may ‘manifest certain agentic capacities’.⁵⁰ Jutta Weber, for example, examines the material agencies of the ‘disposition matrix’, that is, the US government's database of targeted individuals. Weber shows that ‘sociotechnical artefacts such as databases and algorithms are intertwined with human decision-making processes in the production of military targets.’⁵¹

Human-material intertwinement, and the question of what this do to our assumptions of human agency in targeting, is central to this strand. Looking specifically at drones, Caroline Holmqvist argues that contemporary targeting prompt reconsiderations of conventional understandings of human agency, even of what it means to be human in war. For Holmqvist, targeting and warfare by drones are deeply embodied, yet at the same time the agentic capacity of drones suggests a ‘blurring of corporeal and incorporeal’.⁵² Drone targeting and warfare force us to rethink human agency and consider ‘the way in which fleshy and steely bodies associate, interact, merge’.⁵³ Along similar lines, Bousquet suggests that the visual capabilities and perceptual techniques that condition the contemporary regime of global targeting – characterised, as he puts it, by a ‘convergence of perception and destruction’⁵⁴ – are key to targeting assemblages. These assemblages, moreover, make human agency understood as the ‘self-determining subject’ increasingly ‘diffuse and uncertain’.⁵⁵ As Bousquet further suggests, it is thus necessary to ‘decentre’ humans by inserting them in the broader sociotechnical assemblages that condition agency.⁵⁶

Derek Gregory has perhaps gone the furthest in researching and revealing the conditions of human agency in historical and contemporary targeting. For him, the fact that targeting today can be understood as an assemblage of ‘actors, objects, practices, discourses and affects, that entrains the people who are made part of it’ does not deny the human agency of the targeteers, even though it may be more diffuse and hard to locate.⁵⁷ By tracing the genealogy of contemporary targeting, with particular focus on remote warfare enacted by drones, Gregory details the conditions of human agency and subjectivity on both sides of the ‘kill chain’ and teases out what is indeed special about contemporary targeting.⁵⁸

In targeting, remoteness does not equal absolute distance but proximity, and (visual) intimacy rather than detachment. As Gregory says in relation to targeting during counterinsurgency operations: ‘commanders, advisers and analysts scan live video streams in order to push time-critical information to UAV crews and ground forces responding to emergent events. These developments reinforce the rush to the intimate that characterises counterinsurgency operations, but in this case the emphasis is as much on “the rush” as “the intimate”.’⁵⁹

In terms of the conditions of human agency, Gregory argues that the by military advanced technologies screen-mediated proximity, visual closeness, and near real-time embeddedness with forces on the ground may condition targeteers to take (lethal) actions that turn out highly problematic from a legal and normative sense. He gives an example from the Uruzgan province in Afghanistan, where in 2010 a Predator was tracking three vehicles travelling down a mountain:

The Predator crew in Nevada had radio contact with the Special Forces Joint Terminal Attack Controller and they were online with image analysts at the Air Force's Special Operations Command headquarters in Florida. At every turn the flight crew converted their observations into threat indicators: thus the two SUVs and a pick-up truck became a ‘convoy’, adolescents ‘military-aged males’ and praying a Taliban signifier (‘seriously, that's what they do’). After three hours’ surveillance two Kiowa helicopters were called in, and during the attack at least 23 people were killed and more than a dozen wounded. Only after the smoke had cleared did the horrified Predator crew recognize the victims as civilians, including women and children.⁶⁰

The example illustrates the implications in terms of responsibility and accountability when human agency is diffused and in a sense hard to pinpoint in what is a highly technologised and assembled targeting process with a multitude of people involved.⁶¹

As Gregory further details, the role of judge advocates in the targeting process is today more significant.⁶² An in-depth study by Craig Jones details the enabling role of the military-legal profession in military operations, and thus the intertwined relationship between war making and law making.⁶³ In his analysis of how the ‘kill chain’ has become a juridical field, and law has become productive of military violence, Jones gives an account of the doings and agencies of one of several key actors that are involved in military targeting. According to Jones, war lawyers play a central constitutive part in the production of military targets, and thereby in channelling and legitimising military violence – and this while simultaneously making the laws of war through certain doings and interpretations. As he says: ‘the later modern kill chain and its legalities are co-produced. In the CAOC and at other locations in the kill chain, technical registers mix with and animate law and legal advice, and vice-versa, in a co-constitutive and iterative process.’⁶⁴

Summing up, there is a relatively rich scholarship on the history and contemporary logics and dynamics of military targeting, albeit with an emphasis on drone warfare. Works by Gregory in particular (and to some extent by Jones) give an intricate look into both the genealogies of targeting and the conditions and articulations of human (and non-human) agency in contemporary targeting. With the exception of Nordin and Öberg, none of the works discussed denies the presence of human agency in military targeting. Moreover, while none of the works relies on traditional conceptions of human agency as intentional and subjectively enacted by freestanding individuals, most are less clear on their alternative conceptions of agency and its articulation during targeting. Again, Gregory is an exception, yet he does not fully conceptualise different expressions of human agency within the targeting process.

By taking our cue from Gregory, this article argues for a recentring of human agency in the analysis of how contemporary targeting unfolds in practice. Indeed, it is precisely because human agency in targeting is increasingly diffused and hard to pinpoint that it is vital that critical scholarship continues the attempt to locate and reveal it. Without denying the significance of non-human agencies in military targeting, we thus argue that extant research's decentring of human agency risks to foreclose investigations of the potentially multifaceted form(s) and articulation(s) of human agency in military targeting. This in turn is problematic since an understanding of the ways in which human agency are articulated during targeting provides a sense of doings, dynamics, and conditions of constant reproduction/change that are internal to the targeting practice rather than emerging ‘after-the-fact’ in investigations and/or media and public outcries in relation to specific military engagements. As mentioned above, this is closely linked to the becoming of war and constitution of military power more generally. While merely one – albeit central – engine of contemporary wars, and while we are in this article limited to the issue of how to conceptualise distinct articulations of agency in military targeting, researching targeting in this way facilitates an understanding of how wars become, and of alternatives forfeited.

Towards a conceptual ‘recentring’ of human agency in military targeting

Practice theory, which is not one but multiple (and at times incompatible) approaches,⁶⁵ offers intriguing avenues for a conceptualisation of human agency in targeting. The choice may seem unlikely. After all, practice approaches are often accused of dismissing agency and the possibility of malleable orderings in favour of the purely repetitive and reproductive. These charges, however, are misdirected. Practice theories are centrally focused on (constitutive) power, on that any particular order could always be otherwise, and on the becoming of agency through power and orderings.⁶⁶ For most approaches, power relations emerge and come into being in and though the enactment of practices and agencies, while at the same time power relations constitutively shape the conditions and enactments of practices and agencies.⁶⁷ Moreover, practice approaches do not dismiss human agency. Agency is typically treated as both constituted and made possible by a practice, while at the same time a practice (or a nexus of practices) is always constituted and made possible in and through agency.⁶⁸

While most practice approaches thus focus on ‘the interaction between both the emergent, innovative and the repetitive, reproducing sides of practice’ and concur on change as being actualised through human agency,⁶⁹ not all offer detailed reasoning on how to understand human agency in and of practice. One exception is Theodore Schatzki's work. Schatzki targets and reveals human agency within practices, without falling back upon conventional assumptions about intentionality, cognition, and freestanding individuals.⁷⁰ His work is methodologically coherent with a view of power (as emergent power relations) as an effect of and as constituting practices and human agency. For Schatzki, moreover, human agency is possible and finds articulation only from within a practice (or a nexus of practices).⁷¹ There is hence no binary between agencies as articulated by an individual or a group of individuals and submission to the structuring effects of a practice, not least since any particular empirical individual is likely enmeshed and engaged in various practices. Practices are social, non-individualist, and collectively achieved phenomena. Human agency as both conditioned by practices and conditioning practices does not deny submission to the components and ‘nexuses of doings and sayings’ that make up a particular practice. As Schatzki says, agency ‘does not invent the future wholesale from its own resources. Instead, it arcs through a variegated and folded landscape of variously qualified paths: Agency makes the future within an extant mesh of practices and orders that prefigures what it does – and thereby what it makes – by qualifying paths before it.’⁷² Immersion in and some degree of submission – even though this term is problematic – to the interrelations between the elements that compose a practice may thus be required for the articulation of agency in the first place, while at the same time agency ‘determines the instances and structures that allegedly determines it’.⁷³ Indeed, one of Schatzki's merits, for our purposes, is that he breaks down what a practice is and what holds it together, in terms of human agency as distributed across four elements that organise a practice and that condition its maintenance or transformation.⁷⁴

Agency and practice, practice and agency

Being located in and articulated from within a practice, Schatzki conceives of human agency as ‘doing’. Agency as doing ‘is the central motor’ of the constant emergence of a practice, with doings being as mentioned both constituted by and constitutive of a particular practice.⁷⁵ Agency is in this sense linked to responsibility – in the sense of human or non-human entities that perform certain doings and by doing so ‘commence or continue chains of events in the world’ – but not intentionality as conventionally understood in relation to assumed freestanding individuals.⁷⁶ For Schatzki, moreover, human agency resides uniquely in practices – in the bodily doings and sayings that form the actions that make up part of practices.⁷⁷ This does not mean that he denies that material arrangements and objects may co-constitute human agency or that non-human agency exists. Still, to grasp the distinctiveness of practices and of human agency in and of practices, Schatzki – and we follow him here – makes an analytic (but not ontological) distinction between practices and orders, with the latter being analytically sensitive to non-human agency.⁷⁸

A practice, in turn, is an organised nexus of actions. As such, a practice has two dimensions: activity and organisation. Activity can be understood as ‘bodily doings and sayings’, for instance jumping, weaving, or shaking one's head.⁷⁹ What distinguishes a practice from a singular action is, firstly, its social character. Any practice consists of a nexus of actions rather than individual ones, and are ontologically prior to both individual actions and the individuals performing them. As Raymond Caldwell points out, this makes problematic any ‘intentional and causal models of agency’ in which intentions, motives, or goals are assumed to be ontologically distinct from and causing actions.⁸⁰ Second, practices are distinct, social, and non-individualist phenomena given their internal organisation.⁸¹ In order to explicate the internal organisation of practices, the notion of ‘practical intelligibility’ is key. Practical intelligibility refers to what makes sense for any particular human being to do in a given situation. What makes sense, moreover, is continuously learned and carried out in and through collectively organised practices.⁸² Understanding the internal composition and organisation of practices is key to get a clearer sense of how human agency resides in practices. Specifically, human agency is located within and distributed across four elements that together hold a practice together and condition its maintenance, (re)production, or transformation: (1) practical understandings; (2) rules; (3) a teleoaffective structure; and (4) general understandings.

Practical understanding refers to the ability to know what to do in particular situations and to be able to carry this out. A practical understanding ‘executes the actions that practical intelligibility singles out’.⁸³ It refers to ‘knowing how to X, knowing how to identify X-ings, and knowing how to prompt as well as respond to X-ings’.⁸⁴ Practical understanding is thus the skill that makes certain activities (including reactions) possible, but it does not determine these activities. Actors may demonstrate more or less practical understanding in the sense of ability, but this is necessarily always in relation to the practice's wider intersubjective dynamic. In relation to military targeting, practical understandings may concern how to use and operate electronic equipment including computers and their related software, update databases, conduct data and image analysis, and scan images and live video streams to (among other things) ‘distinguish normal activity from abnormal activity’.⁸⁵

Rules refer to a seemingly straightforward aspect of practices, namely ‘explicit formulations, principles, precepts, and instructions that enjoin, direct, or remonstrate people to perform specific actions.’⁸⁶ Significant is that such rules are explicitly rather than implicitly formulated. For Schatzki though, this comes with a twist. Rules do not determine actions. Rather, what people do inform of how to understand the rules. As Schatzki says, drawing on Wittgenstein: ‘rules do not determine what people do; rather, what people do determines what following rules amounts to.’⁸⁷ Relevant rules for military targeting may be explicitly expressed in doctrines, rules of engagement, handbooks, manuals, instructions, planning documents, and guidelines.

The third element is what Schatzki calls a teleoaffective structure. A teleoaffective structure involves normativised projects, goals, and ends that specify accepted and expected doings (what is accepted to do, what ought to be done). The teleoaffective structure is not a feature of the individual performing the practice, but a normativised element of the practice as such – whether or not any specific individual who performs the practice is fully aware. Practices are thus both constituted and linked together by a certain teleoaffective structure, at the same time as practices help construct and reproduce (or transform) this structure. Indeed, teleoaffective structures may be contested, as these are no more or less than ‘recurring and evolving effects’.⁸⁸

The teleoaffective structure of targeting may form around ends and goals that are considered necessary and acceptable in order to achieve strategic, operational, and tactical success, and/or that the involved actors are expected to achieve through their performance of certain tasks (perhaps due to their position in a hierarchical authority). Such normativised ends are never given, but may be more or less openly contested and open to change. For instance, in her critique of precision weapons, Zehfuss proposes that the belief in precision has brought about a shift in the teleoaffective structure of targeting (although Zehfuss does not use this term). The use of precision technologies have meant that targets that previously would not have been seen as acceptable – such as enemy fighters hiding in residential areas in urban environments – are now considered adequate.⁸⁹

The fourth component is general understandings. This is perhaps the vaguest part of Schatzki's framework. Not least, a clear line distinguishing this element from teleoaffective structures is difficult to draw.⁹⁰ Still, the notion of general understandings refers to broader societal understandings that are not necessarily unique to any particular practice, but that may help compose it. Schatzki mentions a ‘sense of community’ and religious convictions as examples of general understandings,⁹¹ although not everyone participating in a practice necessarily hold such views. In other words, although not everyone who performs a particular practice may share a certain general understanding, it is often at least ‘encountered by’ most participating actors.

A general understanding that helps compose military targeting may – again – be the belief in precision weapons, and that these will reduce collateral damage and civilian casualties. Still, the central role that targeting plays for the very fulfilment of this belief⁹² means that any such general understanding is difficult to distinguish from the contemporary teleoaffective structure of targeting. This is why, for example, the mistaken NATO bombing of the Chinese embassy in Belgrade during OAF led to general public and media condemnations and internal critique that forced NATO to review in more detail the ‘political sensitivity of each target’.⁹³ The event went counter to general understandings about a new type of warfare and to the intrinsic teleoaffective structure of the OAF's targeting process.

Practice, agency, and a constant becoming, maintenance, and change

That human agency is located within practices and distributed across the activities and elements that hold a practice together imply that any practice's emergence, maintenance, and transformation is simultaneously conditioning and conditioned by articulations of agency. Indeed, while practices as made up of an organised nexus of actions are in a state of constant becoming does not mean that they are strictly repeated, nor that any practice continues unchanged. Practices may demonstrate regularities but are also ‘irregular, unique, and constantly changing’.⁹⁴ In other words, both the doings that constitute a practice and doings in relation to how a practice is organised in and through understandings, rules, and teleoaffective structures constantly transform, albeit more or less radically so. For instance, the same action and resulting practice may be performed by similar but not identical bodily doings and sayings.⁹⁵ Likewise, the rules that organise a practice may change. Overall, this means that any practice is always and constantly open to smaller or larger transformations.

One way (not the only one) to analytically locate articulations of agency within a practice is to explore whether any particular doing produces a predominant maintenance of or a challenge to the prevailing practice in terms of its internal, organising elements, its practical organisation.⁹⁶ Key is thus whether articulations of agency involve a maintenance, stimulation, and furthering of the practice in terms of continued harmony between its (necessarily temporary) organising elements, or doings that bring about clearer tensions and distinctions between the four elements. Indeed, what makes a practice a particular and reproduced practice is the ‘distinctiveness of the package of doings and sayings plus organization that each is’, and the continued harmony between actions and organising elements, understood as a ‘particular set of doings and sayings expressing a particular array of cross-referencing and interconnected abilities, rules, teleoaffectivities, and understandings’.⁹⁷ Human agency may, in this sense, both constantly maintain and reproduce the practice or generate a challenge, discord, and disharmony between its organising elements, which in the end – even if this question is beyond the scope of this article – may bring about the demise of a particular practice and the birth of a new one. As Schatzki says:

Constant doing must not be equated with change. Many human and non-human doings alike maintain the practice-order mesh as part of which they occur. Maintenance, accordingly, is not the absence of activity, but instead the occurrence of activity that perpetuates practices and reorders arrangements, minimally. Change, by contrast, comes about with activity that alters practices and orders more robustly.⁹⁸

Importantly, any change to and demise of a practice – understood as transformations of its internal activities or organising elements – are commonly gradual rather than revolutionary.⁹⁹ For Schatzki, major discords may arise when any particular organising element is questioned, challenged, or considered incompatible with the other elements. Such occurrences are most noticeable when it comes to rules and teleoaffective structures, whereas practical understandings often modify in a subtler manner.¹⁰⁰ Analytically, what the relation between articulations of agency – doings – and a practice's organising elements means in terms of the maintenance or demise of a practice is always an empirical question.

In the following, the proposed conceptual framework informs an empirical illustration and further conceptualisation of a multitude of agency articulations in military targeting, related both to a maintenance and furthering of the elements that make up the practice and to doings that, in a more challenging way, produce frictions and discord between the elements. Accordingly, we seek in the following not to explain any particular military engagement, nor to explain agency as such, that is, its prefiguration. Instead, we are focused on conceptualising and ‘typologising’ different articulations of human agency in the actual performance of targeting, and by that exemplify (that is, not give an exhaustive view) what distinct articulations of human agency in military targeting may look like.

Portraying human agencies in the targeting practice of Operation Iraqi Freedom

OIF was launched in March 2003. For the operation, an initial list of around 900 targets had been set up, which included both pre-planned targets and expected time-sensitive (TST) ones.¹⁰¹ The operation would come to carry out a historically high number of strikes, particularly against TSTs.¹⁰² Critics have stressed the high number of civilian casualties and the great material destruction that followed OIF's targeting.¹⁰³ For the US military and various military experts, however, targeting during OIF was a great success.

OIF targeting was claimed a success in terms of how the operation had managed to destroy the enemy over a short period.¹⁰⁴ In his testimony to the US Congress in July 2003, Donald Rumsfeld stressed the operation's success in terms of its speed and jointness in connection to the use of intelligence, advanced technologies, and precision weapons.¹⁰⁵ According to Osinga and Roorda ‘in some cases, it took as little as twelve minutes to destroy a confirmed target; at times, this was only five minutes after detection.’¹⁰⁶ The ‘immediate target prosecution’ went ‘beyond what had ever before been accomplished’.¹⁰⁷

The speed that characterised OIF targeting depended on advanced military technologies. Fyfe suggests that the integration of information technology into the Combined Air Operations Center (CAOC) and component headquarters ‘permitted joint visibility into the targeting process’, which eliminated ‘confusion, mistrust, and repetition’ while ‘enabl[ing] the efficiency and speed required to successfully F2T2EA [Find, Fix, Track, Target, Engage and Assess]’.¹⁰⁸ Anthony H. Cordesman, likewise, argues that OIF had an ‘unparalleled degree of near-real time situational awareness that shortened the “kill chain”’.¹⁰⁹ Nonetheless, this was a question not only of technologies but also of the ability to operate them. OIF targeting included a cadre of professionals who became the joint and service-level targeting experts. Many were handpicked and had previous wartime targeting experience from Afghanistan, not least from time-sensitive targeting as it had begun to institutionalise during OEF.¹¹⁰

The claimed success of OIF's targeting was allegedly due also to its decentralisation.¹¹¹ Planners at the CAOC at Prince Sultan Air Base in Saudi Arabia controlled the daily preparation of the target list and issued the daily Air Tasking Orders (ATOs). They also had the authority over collateral damage and causality estimations.¹¹² In addition, the authority over engagement decisions concerning immediate targets was (normally) decentralised to the components.¹¹³ All this contrasted sharply to previous operations’ centralised control at the US Central Command (CENTCOM) in Tampa, Florida.¹¹⁴

Moreover, the OIF heightened focus upon dynamic, time-sensitive targeting and emergent targets involved new doctrinal instructions, rules, and guidelines.¹¹⁵ A new joint targeting doctrine, the JP 3-60 Joint Doctrine for Targeting, had been approved in 2002. It included for instance detailed criteria for the positive identification (PID) of targets and for collateral damage estimates.¹¹⁶ Moreover, the Commander's Handbook for Joint Time-Sensitive Targeting made dynamic targeting increasingly rule-bound.¹¹⁷ A key normative goal of OIF was the avoidance of ‘collateral damage’. A multitude of military and civilian lawyers and analysts was involved in the vetting of targets.¹¹⁸ Key for the OIF was also its effects-based approach. In relation to every single target, the desired and expected effects of engagement on the operation's normativised objectives were meant to be specified.¹¹⁹

The OIF targeting process was thus very concerned with the operation's normative and moral standing. The process was more rule-bound and formalised as compared to previous multinational joint operations. In many ways, and despite the large number of people involved, targeting during OIF was thus a heavily bureaucratised, standardised, and streamlined endeavour subject to rules and principles as expressed in doctrines, handbooks, and operative manuals.¹²⁰ In this way, targeting during OIF may appear to be the uninterrupted and endless process(ing) driven by technologies and routinised procedures imbued with an explicit, rule-bound sense of what could and ought to be done that parts of extant scholarship on targeting emphasise.

The process, in short, does not appear to leave much room for human agency. Still, by reading the aforementioned features through a ‘Schatzkian’ practice lens, the very elements that made OIF targeting appear a stable, rule-bound, and uniform process can be shown to also and simultaneously have conditioned multiple articulations of human agency, where some were more clearly disruptive to the ordering and organisation of targeting during OIF. To analyse this, though, it is vital to first discuss in what sense the targeting process during OIF can at all be analysed as a coherent yet constantly evolving practice in Schatzkian terms. For this, the aforementioned characteristics of targeting during OIF can be reread as the organising elements of a practice, which human agencies both help constantly maintain and further as much as modify, challenge, and transform.

To begin with, OIF targeting was organised by practical understandings that enabled a decentralised organisation, a use of advanced technologies and software, and a quick striking of targets – in short, abilities of the people involved that made the practical (and more or less expertly) execution of the targeting cycle possible. Second, the targeting practice involved a large set of explicit rules that were formalised and expressed in doctrines, handbooks, and manuals, which as we are to see depended on the involved actors’ interpretations and doings. Third, the effect-based targeting approach and the stated goal to diminish collateral damage can be read as the teleoaffective structure of the OIF targeting practice. These were the normativised ends of OIF targeting that prescribed which types of doings were expected, accepted, and required in order to reach the operational and strategic goals, and that held OIF targeting together as much as the targeting practice itself – at least apparently so – helped reproduce these ends. Fourth, and even though a low collateral damage was intrinsic to the targeting practice (and targeting in turn considered the way to achieve this), this particular belief and normative end can be said to have informed the whole of the operation. Thereby it comes close to what Schatzki calls a general understanding. While the belief in the necessity to avoid or diminish collateral damage was not unique to the targeting practice, it conditioned which targeting ends and projects were deemed appropriate, accepted, and hence expected while at the same time amounting to a general understanding that most of the people participating in OIF if not shared so then at least encountered.

Based on this initial rereading of the organising elements of OIF targeting analysed as a (constantly emergent) practice, the following sections illustrate how human agencies and ways of ‘doing’ targeting can be differently conceptualised depending on whether the particular doing produces a maintenance of or a challenge to the elements that organise the practice. By no means an exhaustive list, we arrange different articulations and enactments of human agency at the operational and tactical levels into a fourfold typology. The typology distinguishes between different articulations of agency based on their respective relation to the practice's organising elements, that is, whether the agencies helped maintain and further the practice, or challenged one or more and thus produced a discord between the (temporary) organising elements. Rather than an internally uniform, rule-driven, and stable practice, targeting during OIF can be portrayed as a bundle of distinct types of agency that at times brought about frictions and direct challenges to the prevailing targeting organisation, and at others relieved tensions and thereby helped to predominantly maintain and further the practice (albeit always with minor transformations).

Maintaining and furthering the practice: Adjusting practical understandings in and through agency as ad hoc problem-solving doings

The ability to act with unprecedented speed that characterised OIF and its targeting process brought about some unexpected problems. For example, the MAAP cell (the targeting cell responsible for pairing up available air strike assets with fixed targets) quickly became unable to skilfully manage its workload. In order to cope and carry out the required tasks, the originally planned 12-hour work shift was extended into 24 hours. The head of the combat plans division explained:

The MAAP cell was never initially intended to be a twenty-four-hour operation. … What drove us to a twenty-four-hour MAAP process was the ensuing speed of operations. Normally, any changes required for an ATO, once pushed, would take place on the combat operations floor. But the ‘replanning’ cell was not sufficiently robust to handle that. So we had our MAAP staff on call twenty four hours a day to re-MAAP targets inside twelve hours of execution as necessary, even after the ATO had been pushed.¹²¹

The reshuffling of the MAAP cell was effectuated as a result of human agency as an ad hoc problem solving that involved a minor modification and adjustment of the targeting practice's practical understanding. That is, to be able to maintain the increasingly fast-paced targeting cycle, small adjustments of the temporal set-up were required. This was achieved in a manner that implied only a slight readjustment of one of the various practical understandings that constituted and organised the targeting practice, which illustrates an articulation of human agency that secured the maintenance of the practice and its constituent organising elements.

Maintaining yet indirectly challenging the practice: Adjusting practical understandings in and through agency as ad hoc problem-solving doings

The fast pace also meant that other problems arose, which would come to have greater indirect implications on the composition of the targeting practice. A senior legal advisor who served at CAOC is quoted in Jones explaining that when hundreds of sorties are flown each day, some of the emerging dynamic targets were struck without a preceding legal review. This was especially the case in situations of CAS targets, where the quickness of the response was seen as imperative.¹²² According to the senior legal advisor, he would not typically be awoken in the middle of the night for advising on a CAS target, it would be struck anyway, and given the intensity of the battle he claimed it was simply not possible to provide legal advice on all CAS targets.¹²³ From a Schatzkian perspective, this way of practically doing targeting by omitting a legal vetting of the target and thus neglecting some of the practice's formal rules and regulations can also be understood as an articulation of human agency that through ad hoc problem solving adjusts the practice's practical understanding and helps to maintain the practice. Yet, in this case, this act also indirectly challenged the rules organisational element of the practice.

Likewise, the speed of OIF created obstacles when it came to the assessment of the execution of targets, which is part of the final formal step in the targeting cycle. A senior CAOC staffer is quoted in Lambeth saying that ‘the air component's ability to … prosecute mass volumes of targets outpaced its ability to track and assess its progress.’¹²⁴ Following target execution, assessment results were supposed to be reported back to CAOC where an assessment team would analyse the outcomes. However, the ability to rapidly identify and destroy targets overwhelmed the people responsible for performing such Battle Damage Assessments (BDA). The production of BDAs could not keep pace with the speed of strikes, which in turn undermined the munitions effectiveness assessments and target re-attack recommendations.¹²⁵ Moreover, target executers often did not receive any visual feedback from the attack that could contribute to the assessment process. The imagery provided by the weapons and related IRS-systems did often not give adequate information as to whether the target was actually killed/destroyed. At times, this led to targets being (unnecessarily) retargeted.¹²⁶ Afterwards, this was deemed ‘inefficient and very dependent on duty officer workload’. Many of the people responsible for updating and putting in the right data were quickly ‘overwhelmed’.¹²⁷ To manage, an impromptu solution of how to practically perform targeting assessments was developed, which can be interpreted as an ad hoc adjustment of parts of the targeting practice's practical understanding. Instead of awaiting the BDAs, assumptions were made about which targets had been hit based on mission reports and on what weapons had been used.¹²⁸ Such ad hoc doings represented deviations from the formal rules of the practice, but were also what enabled the targeting practice to be furthered, even accelerated.

The first example of human agency as ad hoc problem solving illustrates an emergence of agency within the targeting practice that entailed that a problem was practically handled through minor modifications in how things were done, without upsetting the harmony between the organisational elements of the practice. The following two examples, however, illustrate instances of human agency as ad hoc problem solving that through minor adjustments of practical understandings did maintain and further the practice, yet in ways that implied deviations or neglects of the practice's rules and regulations. These doings thus indirectly challenged the internal uniformity and composition of the practice. Moreover, the latter examples also illustrate how adjustments of practical understandings may indirectly challenge also the teleoaffective structure. This is because, firstly, the legal review of targets was considered absolutely crucial for avoiding collateral damage and for upholding the operation's legitimacy. Secondly, accurate targeting assessments were considered crucial for achieving the principle of an effects-based operation that guided the OIF. In other words, these articulations of human agency threatened indirectly the teleoaffective structure of the targeting practice, and thus to destabilise and create discord in the practice's internal composition. At the same time, however, the maintenance and furthering of the targeting practice that these articulations made possible also reproduced the impression of an untouched teleoaffective structure.

Maintaining yet directly challenging the practice: Disharmony around or between organising elements in and through agency as fractious doings

The ability to act rapidly further brought about instances when different targeting doings created frictions that hindered a smooth maintenance of the ‘kill chain’. For example, at CAOC tensions arose between the team involved in time-sensitive targeting and the people tasked with replanning and developing new (also non-time-sensitive) targets.¹²⁹ For instance, the time-sensitive targeting team could have discovered a target that required immediate action. This would prompt them to override the already planned and ongoing target execution and take over a required attack asset already airborne in order to engage the time-sensitive target.¹³⁰ Such doings led to ‘occasional friction between TST and Offensive Operations personnel that detracted from the overall offensive operation efficiency, and in some cases delayed or inhibited the floor's ability to prosecute non-TST missions.’¹³¹

Another recurrent tension concerned the concept and practice of positive identification (PID). During OIF, PID was defined in the rules of engagement (ROE) as a ‘reasonable certainty’ that the target to be engaged was a legitimate military target. Doing PID as part of the targeting process thus entailed taking steps to (reasonably) ensure that the target was what/who intelligence said it was.¹³² Being central for achieving an effects-based approach, and more importantly, for avoiding collateral damage or fratricide, PID was therefore at the heart of the teleoaffective structure and general understanding of the targeting practice specifically, and of the whole operation generally. In less rushed deliberate targeting, PID is carried out throughout the cycle – from the development of the target to the planning and execution – and involves legal advisors (such as judge advocates) and intelligence analysts. In dynamic and time-sensitive targeting, however, PID becomes a highly rushed process. During the targeting practice of OIF, tensions arouse regarding who was responsible for carrying out PID in CAS situations. The ROEs did not specify whether forces on the ground that called in air strikes or the executing pilots had the final say on PID. The issued was controversial given situations of mismatch between how judge advocates of the ground and the air force respectively interpreted PID responsibility.¹³³

These examples illustrate that while some articulations of human agency helped maintain and further the internal elements that organised targeting, other articulations of human agency as fractious doings conditioned emergent disharmonies around or between the practice's organising elements. In the first example, different groups of people were acting in accordance with the given rules and regulations that to some extent held the practice together, but frictions nevertheless emerged due to differences between practical understandings with regard to differing or conflicting, priorities and/or uncertainties regarding which targets to pursue. In the second example, tensions arouse around how the formal rules were to be interpreted and understood in the first place. This illustrates how no rule is ever given, no matter its degree of formalisation. Differences in rule interpretations, particularly so in relation to dynamic targeting, also indicate a potential questioning of how well the teleoaffective structure of the targeting practice managed to hold the practice together.

When target executers performed doings that went counter the overall belief and goal of the effects-based approach, other instances of tensions and fractions arouse around targeting's teleoaffective structure. Fyfe describes how ‘there has always been a preoccupation in the fighter and somewhat in the bomber communities that coming home without any unexpended ordnance was a measure of success.’ Dropping bombs could frankly be the goal, without much consideration of the effects thereof.¹³⁴ During OIF, doings that were performed in adherence with such ‘measure of success’ directly contradicted and challenged the effects-based belief that was at the centre of the targeting practice's teleoaffective structure. There were, for example, instances when aircrews ‘went shopping for missions in order to ensure they expended their weapons prior to returning to base’, and when airborne strike assets that were planned to be stationed nearby Baghdad instead went ‘hopping to different land component controlling agencies in search of targets’.¹³⁵ A report by the RAND Corporation describes similar fractious doings. For instance, many aircrews allegedly preferred to work with the Marine Corps’ Direct Air Support Center (DASC) rather than the V Corps’ Air Support Operations Center (ASOC) because of the former's ample supply of targets.¹³⁶ The former offered better chances of returning to base without unexpended ordnance. There were also occasions on which aircrews ‘freelanced’ and actively went looking for targets in addition to those assigned to them. The ‘no weapons bring back’ norm meant that aircrews were so anxious to release munitions that they sometimes pressured air controllers for ‘dump targets’ such as abandoned vehicles.¹³⁷

The aforementioned examples of fractious doings that maintained the targeting practice while simultaneously creating discord around one or between multiple of its organising elements show the fragmented character of the kill chain as it engages a mesh of doings and elements that may encompass more or less contradictory priorities, interpretations, and normative, accepted, and expected belief and goals. The value in the fighter and bombing communities of returning to base without unexpended ordnance, for example, destabilised the belief in the necessity of an effects-based approach, and thus brought about a disharmony around the teleoaffective structure of the targeting practice.

Directly challenging the practice: Disharmony between organising elements in and through agency as defiant doings

During OIF, there were instances in which articulations of agency as defiant doings more clearly imposed a direct challenge to the prevailing targeting practice by producing a clear rupture between its organising elements. These doings actually broke the kill chain. One example concerns how performed actions set different standards for which rules should inform targeting and for how to act in accordance with them. In turn, this was closely related to distinct interpretations of the teleoaffective structure's emphasis on avoiding collateral damage. Compared with the previous OAF and OEF, Jachec-Neale explains, there were during OIF few open strategical controversies within the coalition over what kind of entity could constitute an accepted target. As targeting unfolded at the operational and tactical levels, however, disputes emerged between American and Australian forces over the risk of collateral damage. All of the proposed targets that the Americans suggested to the Australians were revetted by the latter. If the proposed targets were somehow inconsistent or went against Australians’ understandings of the operation's normativised ends, the Australians refused to participate.¹³⁸ On around forty missions, Australian pilots performed outright contrary actions and defied the orders of the American commanders. The pilots refused to drop their bombs, as for them the assigned targets were not valid and acceptable, often due to risks of collateral damage.¹³⁹

From a Schatzkian perspective, these defiant doings informed and decided the Australian forces’ interpretation of their ROEs. The actions performed by the Australian pilots set the standard for how their ROEs were to be understood, which greatly shaped the targeting practice. Again, this is an illustration of how no rule is ever given no matter its degree of formalisation. Squadron leader Pudney describe how pilots refused orders to drop their bombs: ‘Each guy would have made that decision once to half a dozen times in the conflict. It was presented as being just one pilot in one incident, but it was all of us several times.’¹⁴⁰ At times, the impetus to act differently came from Australian commanding officers.¹⁴¹ In other words, the Australian forces’ defiant doings brought about disharmony around the rules of the targeting practice, as well as discord and rupture between the formalised rules and the teleoaffective structure of the practice, with some ROEs deemed more in line with the normativised ends than others.

Conclusion

By drawing on and pushing extant research further into a conceptual discussion, the article has addressed the question of human agency in military targeting. By analysing targeting from a Schatzkian practice perspective, we have illustrated how articulations of human agency from within the targeting process can be distinctly conceptualised depending on whether the doings condition a maintenance of or an indirect/direct challenge to the phenomenon's internal organisation – in other words, whether or not the doings in question involve a choking of the engine of war that is targeting. Based on various empirical snapshots and examples, we proposed a fourfold typology, in which articulations of agency are distinguished based on their particular interrelation to the four elements that organise the targeting practice and that simultaneously condition and are conditioned by agency.

Overall, the analysis illustrates how OIF targeting – which at first glance may appear to have been stable, standardised, rule-bound, and internally coherent process – was a practice constantly in the making. Throughout its constant becoming, human agencies either conditioned a maintenance and furthering of key elements of the practice or challenged its coherence in the sense of bringing about a disharmony around a certain organising element or between multiple elements. In the more radical example, the challenge implied a clearer rupture between some of the organising elements, as explicitly defiant doings implied contrasting interpretations of the rules of targeting and brought about a disconnect between the formalised rules and the normativised ends of targeting during OIF.

This article has hence studied military targeting as a practice in order to develop, illustrate, and contribute with a framework to map and conceptualise articulations of human agency from within the targeting process, and the conditions of a potential transformation that these may or may not bring about. That said, the article cannot draw any firm conclusions with regard to the stability of the OIF targeting practice or the potential rise of a new practice in the sense of the four organising elements, as this would require a diachronic analysis. However, the development of targeting following OIF could be one avenue for future research, as could further explorations of the links between the practice of targeting and other military practices that jointly condition and constitute military power. Indeed, targeting is one of the more significant engines in any war's constant becoming, but necessarily alongside, for instance, military logistics. Future research could do well to examine in both conceptual and empirical terms the relations between targeting and other key phenomena that make military power and the military enactment of violence possible.