Creating common sense: getting NATO to Afghanistan

Abstract

This article raises the question of how NATO became bogged down in Afghanistan. I scrutinise how the alliance became involved in Afghanistan and how it formulated its strategy. In doing this, I follow the general premises of practice theory. However, instead of the common focus on diplomats and their everyday doings, this article suggests an approach that pays more attention to the structure of the field of positions. I demonstrate that the actions of permanently seconded representatives of member states and of NATO’s administrative cadre were crucial in drawing the alliance into Afghanistan. I argue that their actions significantly contributed to the creation of a fatal common sense: namely that the alliance had to become and remain engaged even in the absence of clear political goals. This provided the basis for a means-focused and endless mission.

Introduction

After more than one and a half decades, NATO appears bogged down in Afghanistan. It is still the largest current NATO operation by far. Even at first glance, the alliance’s course appears erratic: engaged in Afghanistan since 2003 when NATO took command of the International Security Assistance Force (ISAF), the alliance initially intensified its engagement. NATO then reduced troop numbers in the early 2010s and transitioned to the downsized Resolute Support (RS) mission, only to increase it again in 2017 and begin reducing it by 2020. The Afghanistan engagement has turned into a “fruitless but never-ending effort”, as Walt aptly characterised it in a recent article.¹ Most independent scholars agree on the core problems of NATO’s Afghanistan engagement: the majority of studies point out that member states did not (or not early enough) agree on a clearly defined common goal that would have allowed the organisation to set priorities. They only defined a vague, overambitious mandate that unrealistically sought to help transform Afghanistan in a relatively short timeframe into both a peaceful country, free of “terrorists”, and a modern liberal state.² Apart from this debate on strategic mistakes, scholars also point out that “on the operational level” member states did not apply the appropriate approaches and never fully cooperated or were willing to dedicate enough resources to the enormous tasks.³

The focus of the literature on strategic mistakes and operational shortcomings provides a convincing explanation for why NATO has struggled to execute and successfully end its engagement in Afghanistan. Yet, while the problems of strategy-making and gathering domestic political support and other questions of the execution of the mission have been well addressed, we still lack an understanding of why the mission started at all. The main question should be: Why did NATO become at all involved to such an extent in Afghanistan without a clear and widely shared purpose? I divide this broad question into two sub-questions: Why did the alliance ever engage in a mission with an unclear purpose (the main reason behind all strategic problems)? Why did it expand its engagement despite the lack of clear goals and most member states’ unwillingness to risk the lives of an adequate number of soldiers?

To answer these questions, it will be necessary to move beyond approaches that focus on states and international organisations as unified actors. Such approaches struggle to provide fully convincing explanations for the alliance’s course of action in Afghanistan. Instead, it will be helpful to pay attention to groups of real-world actors who constitute what we usually perceive and interpret as states and international organisations. With this perspective and in response to the research question, my main argument is that the alliance’s paradoxical moves were caused by the common sense held by NATO’s “administrative cadre” (i.e. the “secretariat”) and member states’ “seconded representatives” (i.e. diplomats, military officers etc.), according to which the alliance had to be intensively involved in Afghanistan. Their actions alone did not weigh heavily enough to cause fundamental decisions on the direction of the alliance. But combined with occasional support from the national leadership (“major national representatives”) of important member states, they provided the necessary weight in favour of the alliance’s paradoxical decisions. I argue more generally that, in order to understand NATO, the practice of these three groups of actors (namely the administrative cadre as well as seconded and major national representatives) has to be drawn into the focus. This perspective will advance our theoretical understanding of NATO, which has, so far, mostly either rendered the alliance a reified entity or only differentiated between member states’ preferences and those of the alliance secretariat. In addition, I will contribute to recent practice-oriented approaches that have already focused on seconded representatives.⁴

In the following sections, I position my approach and arguments within current theoretical and conceptual debates, especially with regard to the booming field of Pierre Bourdieu-inspired practice theory and research in International Relations (IR) scholarship. In contrast to the current mainstream within IR practice theory and research, and in line with Bourdieu’s original concept, I argue for a more structural perspective that relates the ideas and actions of actors more strongly to their position. To this end, I depart from the common micro-focus on practices (plural)—meaning specific things actors do on an everyday basis. Similar to bureaucratic politics and organisational theory approaches that argue that “actors stand where they sit”, I choose instead a broader perspective and relate the positions of actors to the structure of the field of positions. Through this, I seek to understand their practice (singular)—meaning a structured mode of action that results from a specific and largely habitualised common sense.

In my empirical analysis, I answer the main research question (why did NATO become involved to such an extent in Afghanistan without a clear and widely shared purpose?) by showing that the alliance’s administrative cadre and seconded representatives were unconsciously guided by a common sense to preserve and increase NATO’s importance by participating in the Afghanistan intervention and expanding the engagement. They also cooperated with elements of the member states’ governments who were in favour of expanding the mission. As will be elaborated in more detail below, this study is based on multiple sources, including mostly declassified documents and numerous interviews. Finally, I draw conclusions about the role of member state representatives and the administrative cadre in NATO.

A Bourdieusian structuralist perspective

What distinguishes practice theory from most mainstream approaches is that it takes real-world actors instead of theoretical entities like “states” as level of analysis. Practice theorists are interested in what these actors actually do instead of analysing norms or policy outcomes. Broadly speaking, they assume that practice evolves from social context rather than from a rational cost–benefit calculation. Yet, beyond these commonalities, there is still a lively debate about the exact place and content of practice theory.⁵ One of the major fault lines between practice theorists seems to run between those who see practice as a result of structure and those who analyse practice in its own right. There is no universal understanding of “structure” in practice theory and it is impossible to elaborate on all available interpretations. I will therefore elaborate on the concept provided by Pierre Bourdieu who is probably the most frequent theoretical reference for practice theory. Next, I only briefly address alternatives.

For Bourdieu, the meaning of structure is most closely associated with his concept of field, a web of “objective relations between [social] positions”.⁶ According to Bourdieu, positions are more important than the actors who occupy them since, in his concept, it is the specific structural position that grants a relative amount of power to its occupants. The constellations of actors on differently endowed positions on a field also create specific stakes all players common sensically internalise through socialisation. From this evolves a field-specific practice in which actors compete for power in struggles they do not necessarily understand as such and in which they do not fully strategically plan their actions (position-takings).⁷

Bourdieu’s concept to explain the internalisation of structural conditions and its consequences is the habitus, the sum of a person’s dispositions. The habitus tends to be adjusted to specific fields, as all dispositions—including field-specific stakes—are acquired through socialisation on the fields. In consequence, the habitus provides actors with schemes of perception that filter reality in such ways that actors favour decisions that appear to benefit them in struggles on the field. The habitus therefore works as a “structuring structure”: it leads actors to internalise and—for the most part unconsciously—reproduce the structure of the field.⁸

For instance, Bourdieu demonstrated that scholars’ political preferences (conservative or revolutionary) tend to reflect their position in the academic apparatus (powerful or weak) and that taste for specific styles mirrors actors’ positional location on the field.⁹ While actors also strategically calculate their actions, the habitus makes them do these calculations only within the narrow boundaries of a common sense defined by the structure of the field of positions. Decisive in Bourdieu’s concept is that actors’ subjective preferences can only be understood with reference to the objective conditions under which they have been socialised and that they rest on a largely unconscious common sense.¹⁰

There is some debate among Bourdieu experts about the theoretical nature of his approach. It appears that most scholars consider him a structuralist—at least at some point in his career.¹¹ This is because the position of actors on the field, particularly with regard to its material socio-economic foundation, is so important in Bourdieu’s theory. This holds especially true for his assumption that actors are (mostly) unable to grasp reality since the filters for their perceptions are determined by structure. Also his conviction that only an outside observer (or a reflexive researcher) is able to properly analyse practice—if he or she properly applies methodology to overcome structural bias—is seen as structuralist.¹²

Especially practice scholars who do not refer to Bourdieu have criticised his structuralist approach as being too static.¹³ Even scholars whose interpretation of practice theory is mostly informed by Bourdieu often do not (fully) embrace his structuralist assumptions and conceptualise practice as being (more) independent from structure, “not merely outcomes to be explained but also explanans”.¹⁴ In the same vein, one of the forerunners of IR practice theory, Emanuel Adler, while sharing many aspects of Bourdieusian theory, recently distanced himself from Bourdieu’s focus on socio-structural and material factors and the idea that actors cannot properly reflect their actions.¹⁵

Yet, these Bourdieu-inspired scholars still use most of Bourdieu’s theoretical repertoire, which renders their interpretation of practice theory somewhat inconsistent.¹⁶ But even non-Bourdieusian practice theorists still assume that practice is, to some degree, structured or organised and point to recurrent features.¹⁷ However, in contrast to Bourdieu, alternative non-structuralist approaches lack a clear idea of what precisely it is that structures practice. Since they refuse to give much weight to the structure of the field, non-structuralist practice theorists have been assuming a micro-perspective focusing on actors’ practices as everyday activities in international relations. A prominent question is usually how specific actors “navigate a social milieu successfully”.¹⁸ By contrast, I understand practice as what actors actually do with a sense of agency, yet wherein they are decisively limited and intuitively led by structure as incorporated in their habitus. Structural conditions are decisive for determining the logic of practice of a field that results from the field’s particular distribution of positions and the character of dispositions habitualised by the actors on it.

Practice studies of the non-structuralist strand of literature have greatly enhanced our understanding of precisely how international relations are conducted, including NATO’s ISAF mission.¹⁹ Yet, their focus on how actors pursue strategies in the “engine room of world politics”²⁰ is at the expense of why they do so. Also, their understanding of practice as both explanans and explanandum at the same time has made the approach vulnerable to critique that it is impossible to provide cause and effect explanations that still underlie all scholarly work.²¹ The strong focus on micro-level interactions between permanent representatives or diplomats of different nations has also prevented a broader outlook on the field on which these actors interact and a relational perspective vis-à-vis other actors.

Practice scholars who stick more closely to Bourdieu’s original theoretical idea have demonstrated the explanatory power of a structuralist approach with regard to national foreign policy and European Union policy.²² I will use my empirical study to demonstrate that a structural approach also furthers our understanding of NATO policy. Its usefulness in this regard as well underlines the broader benefits of taking practice theory into a more structuralist direction.

The structuralist approach shares core strengths of bureaucratic politics (BP) and organisational theory IR approaches and further advances them. BP is to be credited for opening up the black box of the state by demonstrating that foreign policy is not the result of rational decision-making by one unified actor (“the” state). Rather, the varying interests of state representatives lead to often paradoxical compromises. BP also demonstrated that “actors stand where they sit”, that is, the interests of state representatives result from their position in the state apparatus.²³ Yet, unlike Bourdieusian practice theory, BP does not provide the epistemological depth to explain why this is the case. Also unlike Bourdieu, the majority of BP approaches—mostly implicitly—assumed that actors act consciously and rationally.²⁴ Furthermore, BP has only been applied to national foreign policy actors and not to those of international organisations or those working at the interface between them. The most convincing critique, however, refers to BP’s failure to explain the emergence of political ideas that actors want to see implemented.²⁵

Organisational theory scholars convincingly demonstrated that international organisations like NATO are not mere instruments of states, but increasingly develop policies independent of their member states.²⁶ They have focused on the relationship between states as patrons and organisations as principals who usually strive for more autonomy.²⁷ By doing so, they have also assumed that organisations act rationally. As most non-BP scholars did with regard to states, organisational theorists have reified organisations as black boxes. Organisational theory therefore does not allow for a differentiation between groups of actors who constitute organisations—like the administrative cadre and those who are only temporarily seconded to them.

A Bourdieusian structuralist approach fills significant blind spots of BP and Organisational Theory. First and foremost, its concept of “common sense” helps to understand how political ideas evolve among practitioners and how they are shaped. This concept also helps to incorporate insights from earlier studies that show how actors often do not decide rationally—without getting lost in a post-structuralist “anything goes” theme. Finally, due to its focus on relations, the “field” enables and demands a close analysis of the relationships between pertinent actors who actually constitute those processes that appear as interactions between member states and NATO.

Mapping the structure of the NATO field

Practice scholars have already demonstrated the analytical value of conceptualizing NATO as a field. Yet existing studies have mostly conceptualised NATO as one field in relation to other fields (e.g. EU, NGOs etc.) without scrutinizing its structure.²⁸ Some recent works on NATO have been somewhat more differentiated but still remain limited to scrutinizing fields that appear to be only populated by diplomats (who I include in my definition of seconded representatives);²⁹ or they do not differentiate between the positions and dispositions of diplomats and those of the alliance’s administrative cadre.³⁰ The relations between heads of state and government and sometimes ministers for foreign and security affairs were mostly left to traditional IR scholarship. However, a structuralist approach demands that one has to scrutinise the whole NATO field with all its inhabitants in order to understand its practice.

This means that the relations between the relevant positions have to be determined. These relations are crucial for understanding why the “occupants” of specific positions develop a certain view on issues, desire certain sorts of capital and take stances (position-takings).³¹ In my conception, the NATO field is the field on which the alliance’s direction is determined, that is, where the most momentous decisions like major organisational reform or war and peace are made. One should keep in mind that Bourdieu conceptualised the social space as three-dimensional and fields as overlapping, depending on the relations of its inhabitants. The NATO field is therefore also a node on which fields formed between NATO representatives and actors as different as journalists, IR scholars or members of the pro-NATO Atlantic Treaty Association overlap. Yet, I limit my study to the positions most pertinent to influencing NATO affairs.

Evidently, the most powerful positions on the NATO field fall to actors that I term major national representatives: heads of state and government, ministers of foreign affairs and defense and their deputies. On the few occasions during which they are personally dedicated to NATO affairs, they form a field (identical with the North Atlantic Council, NAC) on which they struggle for unanimous decisions on the most important issues with the major national representatives of the other member states.³² Given their important domestic role, major national representatives only occasionally convene on the NATO field at summits or ministerial meetings. Since they are rarely exposed to the NATO field, their habitus is more strongly formed by their domestic foreign policy fields.

For those national actors who reside more permanently than major national representatives on the NATO field, I chose the term seconded representatives.³³ This analytical term should not be confused with the legal practitioner term for government representatives who are seconded for a certain time period to an international organisation. Here, the term seconded representatives encompasses all actors who only temporarily serve at NATO, for instance those who are legally “seconded” as officers in the command structure. Because I have observed a similar practice among them (as will be discussed in more detail below), I extend the term to the national civilian and military representatives who belong to the permanent representation of their home country at NATO, led by an ambassador.³⁴

Seconded representatives deal with NATO issues on two different sorts of fields. They form one field with their superiors in their domestic foreign and defence ministries headed by major national representatives. Seconded representatives also form fields with the seconded representatives of other member states they interact with. The superiors expect not only permanent representatives and their deputies but also officers in the NATO command structure to enhance the “national interest” on the NATO fields. Superiors sit on more powerful positions since they rate and may relocate seconded representatives. Yet, as practice and some principal-agent studies have demonstrated, seconded representatives sometimes also follow policies that are not perfectly congruent with orders from their domestic superiors. Due to their direct and daily involvement in NATO affairs, they can draw on superior information and shape reports for their domestic superiors.³⁵ The habitus of seconded representatives is formed on these actors’ domestic and the alliance’s fields, as is most obvious with military officers who often have trained or served in a NATO context.³⁶ Bourdieu’s concept of hysteresis helps us understand why former seconded representatives even speak out for NATO long after their service.³⁷

The third group of actors sits on positions that formally belong to what I term the NATO administrative cadre. Officially, they only organise the national decision-making process of member states representatives and execute their decisions. However, numerous studies have demonstrated how “the secretariat” of NATO follows its own policies and tries to achieve maximum autonomy from member states.³⁸ Since they tend to work more permanently (up to a working-life time) on the NATO field, they can often draw on even more information and direct involvement than seconded representatives.³⁹ Members of the administrative cadre also have considerable leg room for interpreting policies in ways that further their goals; such room is left to them by the often vague decisions that emerge from compromises between member states. They form fields with representatives of all member states. Yet, previous research demonstrated that the administrative cadre’s practice is most strongly shaped by the secretary general’s relations with major national and seconded representatives of powerful member states. The secretary general is therefore forced to maintain good relations with them, in particular with NATO’s most important military commander, the US-appointed Supreme Allied Commander Europe (SACEUR).⁴⁰

After having assessed the structure of what I conceptualise as the NATO field, it becomes necessary to establish the shared stakes for the three most pertinent groups of alliance actors. It appears that these stakes boil down to one widely shared commonsensical conviction: namely the idea that the alliance is necessary to ensuring security and that it has to be maintained. Therefore, major national representatives of all member states prefer to be seen by others as contributing to this end. They want their respective country to look like a “good ally” and make efforts to avoid appearing like a free rider or spoiler working against NATO cohesion.⁴¹ They do this by trying to avoid minority positions in voting and (even if only symbolically) participating in common efforts. Evidence from NATO’s ISAF politics has provided further proof of this logic of practice among major national representatives.⁴²

Yet, as the vast literature on differing perceptions of the purpose of the alliance among member states demonstrates,⁴³ while major national representatives broadly agree on the general relevance of the alliance, they often disagree on the exact purpose and relevance of NATO for a specific policy issue. It appears that such disagreements are due to the different logics of domestic fields where “the national interest” is formulated. Apparently, they strive to reconcile both logics and are vulnerable to influence by actors who try to achieve decisions more in favour of the NATO or domestic field. I will not elaborate on the national logic of practice that is not determined by NATO positions, but only refer to it in the empirical part as far as it effected key decisions made on the alliance’s field. In contrast to seconded representatives and the administrative cadre, major national representatives do not hold any kind of capital whose value is determined by NATO’s role, they therefore seem to have a more instrumental view on the alliance.

As they have been socialised on and are rated by actors on their domestic fields (with consequences for their respective careers), seconded representatives share the basic elements of their major national representatives’ stakes in that they try to promote the “national interest”. Yet, judged by the evidence of detailed historical studies and other non-political science accounts it appears that their judgements often show pro-NATO tendencies.⁴⁴ For instance, historians found that NATO diplomats formed informal networks with other NATO diplomats and its secretariat during the Cold War and thereby worked as a bridge between national governments and the alliance. These diplomats often promoted a different, more NATO-friendly policy than their government.⁴⁵ It therefore appears that seconded representatives advise their superiors and filter information in ways that underline NATO’s importance. Rarely do they give advice that would, in effect, make the alliance appear less relevant.

Bourdieu’s concept of cultural capital helps to understand this practice. It describes actors’ potential benefits from acquired knowledge like training and expertise. As pointed out by Mustafa Emirbayer and Victoria Johnson in their application of Bourdieu’s concepts to organisational theory, cultural capital “is a key source of power in organisations”, since it is hard to acquire.⁴⁶ In his analysis of the French public administration, Bourdieu demonstrated that (subaltern) actors fight to preserve the value of their cultural capital because much of their power rests on it.⁴⁷

It appears that this is why the habitus of national NATO experts like permanent representatives or desk officers responsible for NATO affairs in national ministries makes them intuitively prefer options that strengthen the alliance’s international relevance. It is precisely here that Bourdieu’s assumption of an overlap between objective position and subjective meaning comes into play. Accordingly, one should not assume that actors necessarily strategically plan to increase or maintain the value of their cultural capital. Rather, the habitus’ mechanism of sorting and judging information makes them unconsciously adopt a common sense that NATO should be an important player. It is this common sense that defines their stakes on the NATO field. In contrast to the administrative cadre, the perspective of seconded representatives marries the importance of NATO with their member state’s aspirations. For instance, US seconded representatives tend to see NATO as enhancing the power of the USA instead of constraining it as American critics of the alliance argue.⁴⁸

Members of NATO’s administrative cadre internalise very similar stakes. Even more than in the case of seconded representatives, the value of their cultural capital rests on the relevance of NATO—their direct employer. The very same Bourdieusian social processes make them embrace ideas and policy-options that give the alliance a more important role. This has been most thoroughly demonstrated by studies on the heads of the administrative cadre, the secretary generals. Especially after the Cold War, they have lobbied for involving the alliance in the solution of international crises and for better resourcing of its activities. Bourdieu argued that, if actors are in homologous situations, they often share similar interests and follow common strategies even if they are mostly based on different fields. For instance, workers and leftist intellectuals mostly act on very different fields, but while leftist intellectuals mostly belong to the privileged classes, they are on a homologously low position on the field of the privileged as workers are in society as a whole.⁴⁹ The same appears to apply to members of NATO’s administrative cadre and seconded representatives who—from very different positions—share the interest to protect and elevate the relevance of NATO.

Research approach and methods for the empirical study

To answer the question of why NATO became engaged in Afghanistan without a clear political purpose, I will look at what I consider key decisions that led to this result. These were: the invocation of Article 5 of the North Atlantic Charter (collective defense) after the attacks of 11 September 2001; the assumption of the command of ISAF by the alliance in August 2003; and the formulation of a strategy for and geographical expansion of the mission from 2003 until 2006. In accordance with Bourdieusian methodology, as a first step I will seek an understanding of the common sense that guided the administrative cadre and seconded and major national representatives by reconstructing their position-takings in terms of decisions, recommendations and internal deliberations.⁵⁰

Next, I spell out the unquestioned common sense implicitly inherent in these position-takings. To be precise, this is my empirically grounded interpretation of NATO actors’ common sense, which is made possible by an observant and relatively disinterested outside position. My position is “relatively disinterested” in the sense that I do not share the stakes shared by the actors whose practice I analyse; however, I am interested in writing about NATO and in making my interpretation known. Then, I contrast NATO actors’ common sense with a “rational” problem-solving solution to NATO’s impasse in Afghanistan in order to point out how actors’ position-takings were actually products of their habitus and its relation to the conditions they acted on. In other words, I use this “rational” problem-solving solution as a necessary analytical construct to render visible the structural conditions that unconsciously underlay actors’ practice. I realise, of course, that even this analytical construct is positioned and can be contested. But its use should be acceptable as long as its positionality is acknowledged and made transparent.⁵¹

Such a complex methodology has to be based on multiple sources. Primarily, I rely on declassified documents from several online archives like the National Security Archive, The Rumsfeld Papers, and the Freedom of Information Act online reading rooms of the US Departments of State and Defense and some that I received from the German Federal Chancellery and Federal Foreign Office pursuant to the German Freedom of Information Law.⁵² I also analysed the memoires of decision-makers who held positions in NATO and relevant member states during this time period. However, while actors often consciously or unconsciously rearrange or confuse historical events in retrospect, formerly classified documents in particular have the advantage of enabling researchers to uncover the actors’ position-takings at the time under scrutiny. They and other written sources can also be read by other scholars who can thereby test the validity of my arguments. Furthermore, in the years 2011 until 2020, I conducted semi-structured face-to-face or telephone interviews with individuals who worked as civilian staff and military officers in a NATO role at the time of the interview or shortly before. I also scanned through the almost 600 interviews the US Office of the Special Inspector General for Afghanistan Reconstruction (SIGAR) conducted with decision-makers in the Afghanistan engagement up to the ambassadorial and four-star general level.⁵³

Whether in the form of official documents or interviews, I do not simply take actors’ statements as evidence for their actual practice. I only accept their accounts of practice as actual practice if I have found other sources that confirm these accounts. Due to space limitations, I am not able to elaborate on each single case and on my decision to accept an actor’s account—or not. Instead, I focus on relating actors’ statements to their position within the alliance and to the positions of other NATO representatives. This allows me to draw attention to relationships between actors and to the “common sense” aims that inform and guide their practice.

Afghanistan, NATO and ISAF

Making Afghanistan a NATO issue

To understand how NATO became involved in Afghanistan, it is necessary to analyse the practice of the most powerful actors after the 11 September 2001 attacks. In the immediate aftermath of the events, actors on the NATO field and on the national fields of the alliance’s member states made very different position-takings. There is no evidence that actors on the governmental fields of NATO member states considered to consult the alliance at that point in time. Senior members of the US administration were occupied with emergency measures and—since nobody declared responsibility—assessing intelligence on the possible perpetrators of the attacks.⁵⁴ Decision makers in allied governments condoled and contemplated how they could support the US or protect their own countries from similar attacks.⁵⁵

The only actors who instantly referred to NATO as news of the “9/11” attacks spread were seconded representatives and members of the administrative cadre. Most witnesses agree that it was the Canadian dean of NATO ambassadors, David Wright, who in a conversation with the US permanent representative, Nicholas Burns, brought the idea on the table to invoke Article 5 of the North Atlantic Charter that would call for collective defence.⁵⁶ The NATO ambassadors soon agreed on this initiative and reached back to their capitals to receive authorisation.⁵⁷ At the same time, senior members of the NATO administrative cadre discussed the Article 5 option and also reached consensus on its invocation.⁵⁸ Their head, Secretary General George Robertson, enthusiastically embraced the idea and started to lobby for it among the major national representatives.⁵⁹ SACEUR General Joseph W. Ralston also supported the invocation of Article 5.⁶⁰ Unsurprisingly, there is evidence that no more than one day after the attacks, seconded representatives at NATO started contingency planning for “an assault [that] would involve tens of thousands of ground troops”.⁶¹

It is worth noting the ease with which seconded representatives and members of the administrative cadre “quickly satisfied [themselves] that there was a good case for declaring that the attacks had triggered the Washington Treaty’s collective-defense provisions”, as one of them recalled.⁶² Designed for an “armed attack against one or more of” the members states that constitutes a breach of “international peace and security”, the article, in fact, clearly referred to inter-state war and not to acts of “terrorism” by non-state actors like those behind the attacks. To include the vague category of “terrorism” could have involved the alliance in contested internal struggles like Turkey’s Kurdistan issue. Especially the major national representatives of France and Germany had therefore prevented an unambiguous reference to counter-terrorism as a NATO task in its 1999 Strategic Concept.⁶³ Another argument against the invocation was that the supposedly “attacked” state did not ask for collective defence. The US rather had to be urged to do so. But the habitus of seconded representatives and members of the administrative cadre disqualified all these objections in favour of a common sense that gave way to a situation which increased the value of their cultural capital. Accordingly, about a week after the attacks, US Ambassador Burns pointed out in a NAC session that the invocation of Article 5 “did not only send out a clear political signal, but it also made every single US citizen aware of the importance of the alliance”.⁶⁴

However, supporters of the Article 5 invocation had to overcome the resistance of major national representatives. Most important was the support of the “attacked” state, the US, for the proposal. Yet, Secretary of Defense Donald Rumsfeld opposed the idea.⁶⁵ Eager to keep US military forces under his authority, he had unambiguously written in an internal memo in February 2001:

The command structure must be something we can control – not UN control or a joint or collective command structure, where command decisions are made by others or by a committee. Neither NATO, the UN or any other coalition should be in a position to control US decision-making.⁶⁶

As the value of his cultural capital rested on the relevance of multilateral relations, Secretary of State Colin Powell perceived collective defence more positively.⁶⁷ Burns was therefore able to convince him to support the invocation of Article 5 and, after President George W. Bush’s approval, received a corresponding guidance.⁶⁸

Also in other member states, advocates of the Article 5 invocation had to overcome the reluctance of major national representatives. The latter were looking for ways to show solidarity with the US and participate in her reaction to the attacks. But, contrary to the most widely known narrative about the events of these days,⁶⁹ their position-takings were not too different from Rumsfeld’s.⁷⁰ Like him, most heads of government and ministers of defense preferred to see Article 5 as a declaration of mere political support. They were reluctant to give a military carte blanche not only to the US, but also to NATO. This indicates that they did not want to contribute numerous troops to the massive ground operation some on the NATO field envisioned at the time.⁷¹ From the perspective of major non-US representatives, the formalised force generation process posed another disadvantage since it was harder to escape than a bilateral agreement.⁷² In some cases, they also feared that the domestic political opposition might criticise the soon to come mission as military adventurism in favour of the unpopular and “unilateralist” Bush administration. Therefore, a significant number of them only approved the Article 5 request on the condition that the US would provide evidence on the attackers. Also, they highlighted the provision of Article 5 that they would only have to provide “such action as it deems necessary”. Only under these conditions did NATO decide on the self-defence clause on 12 September 2001.⁷³

On the US administration’s field, actors identified Usama-bin Laden as the mastermind of “9/11” and unsuccessfully demanded his host—the Afghan Taliban regime—to extradite him. In consequence, President Bush decided to attack Afghanistan and—as Rumsfeld had demanded to be able to act quickly and avoid interference of NATO allies in operational planning—that this operation should not be conducted under NATO command but through the US-led Operation Enduring Freedom (OEF).⁷⁴ Yet, probably as a concession to the multilateralists, the Bush administration asked NATO for help outside Afghanistan and used its institutions as a forum to bilaterally gather Afghanistan-related support.⁷⁵ On 2 October 2001, US representatives also provided evidence on the attackers, which representatives of other member states accepted as a condition to invoke Article 5. Some did so reluctantly, but also did not want to be seen as “bad allies”.⁷⁶ Even though NATO did not fully participate in the Afghanistan intervention, the position-takings of the majority of seconded representatives and of the administrative cadre were decisive steps to get NATO involved in the first place. They contributed to the emergence of a common sense that was tailored to their positions on the alliance’s field. This common sense demanded that NATO had to be engaged in Afghanistan. The invocation of Article 5 made it harder for national representatives to legitimise decisions that did not involve NATO in Afghanistan.

Assuming responsibility against all odds

On 7 October 2001, the OEF intervention into Afghanistan started and, with the help of Afghan opposition forces, within a few months led to the collapse of the fragile Taliban regime.⁷⁷ Early on, major US representatives struggled over the direction of post-conflict policy. The multilaterally-oriented representatives of the Department of State (DoS) favoured a preferably UN-led, long-term, countrywide peacekeeping and nation-building project significantly supported by the US.⁷⁸ Representatives of the Department of Defense (DoD) also considered a peacekeeping force, but clearly spoke out against nation-building.⁷⁹ Furthermore, they suggested that the “US should not commit to any post-Taliban military involvement, since the US will be heavily engaged in the anti-terrorism effort worldwide”.⁸⁰ They also feared that heavy US involvement would take incentives from other states to contribute. Therefore, Rumsfeld ruled against a US military participation in a UN- or NATO-led post-Taliban peacekeeping force, which might create an “in together, out together”-logic that could eventually pull the US into it. To avoid interference with the heavy-handed OEF, the peacekeeping force (of yet undetermined shape and origin) should be limited to Afghanistan’s capital Kabul and its immediate surroundings.⁸¹

Again, not too different from Rumsfeld, most allied major national representatives did not have big appetite for becoming engaged in a long-term peacekeeping force in Afghanistan. Most forward-leaning were those of the United Kingdom, who tried “to be seen as the senior partner to a US-led war”.⁸² Therefore, the British declared that they would lead the peacekeeping force, but demanded that it should be integrated in OEF. However, major representatives of Germany and France and other member states fought this proposal in order to avoid strong US control over the mission. Finally, the British decided to lead the mission anyways, but only until June 2002. ISAF—established in the 2001 international Bonn agreement—deployed in January 2002 to Afghanistan. It was limited to operating in the Kabul area, where it patrolled and trained Afghan security forces.⁸³

Contrary to Rumsfeld’s assumption, American absence from ISAF rather decreased incentives for allies—who mostly sought proximity to the US—to contribute costly capabilities and take on the burden as a lead-nation. Not bound to any international organisation, ISAF increasingly struggled to gain those assets. The US even had to materially support Turkey in exchange for its willingness to succeed the United Kingdom as lead-nation in June 2002.⁸⁴ Representatives of the DoD therefore began to revise their stances on ISAF. Especially as violent conflicts between armed factions increased after the fall of the Taliban, they now also considered to expand ISAF beyond Kabul and/or to bring it under the authority of an international organisation. Yet reservations against an expansion of ISAF and its subordination to an international organisation remained in principle. Rumsfeld’s stances on these issues after the fall of the Taliban regime were “a little inconsistent” as his close former advisor Marin Strmecki later admitted.⁸⁵ In April 2002, Rumsfeld spoke out against ISAF expansion.⁸⁶ Yet, in June 2002, he wrote that he had “no problem with expanding the ISAF […]. If there are folks who want to do it, I agree with you, let them do it. We need to keep our focus where it is”.⁸⁷ During the summer of 2002, the US government decided to expand US instead of multilateral efforts in Afghanistan. The means to this end were US inter-agency civil-military Provincial Reconstruction Teams (PRT) that, one after another, started to become operational in January 2003.⁸⁸

In this undecided and dynamic struggle, seconded representatives at NATO and members of its administrative cadre made position-takings that decisively supported those domestic representatives of member states who supported a NATO assumption of ISAF command and its expansion beyond Kabul. NATO ambassadors from the first two lead-nations, the United Kingdom and Turkey, viewed ISAF positively and advised major representatives back home to let the alliance assume command.⁸⁹ Yet this did not help overcome resistance in member state capitals. NATO policy on ISAF only started changing with the next two lead-nations. In the summer of 2002, the governments of Germany and The Netherlands agreed to assume the lead-nation role, but only if they received NATO command support.⁹⁰ Secretary General Robertson endorsed this proposal and successfully gathered support among the member states, who approved the request in October 2002.⁹¹

NATO’s support role provided the decisive basis for its eventual assumption of ISAF command, which was still a contested policy among major representatives of the member states. For instance, the French even initially opposed the German-Dutch proposal and “force[d] the ISAF issue back to the Military Committee” as a DoD official noted.⁹² Especially the seconded representatives at the alliance’s major command, headed by the SACEUR, Supreme Headquarters Allied Powers Europe (SHAPE), who now assisted ISAF, proved crucial in pushing NATO’s engagement. This was not quite in accordance with established procedures: to ensure that military planners in the NATO command structure do not constrict member states’ political decisions, they require an “initiating directive” from the NAC to start planning. Yet, SHAPE planners did not wait for such a directive. The only reasonable explanation for this being that they had a strong interest in NATO assuming command. They started to draft options for NATO assumption in early 2003 and presented them to the secretary general, who supported and lobbied for assumption,⁹³ and to member state representatives. This provided the basis for an official request in March to assume command by Germany, The Netherlands and Canada—who had voiced willingness to assume command as next lead-nations if ISAF were posed under NATO command.⁹⁴

To make NATO assume command of ISAF and overcome resistance among the member states required the support of the most powerful alliance member—the USA. For this, US seconded representatives Ambassador Burns, the SACEUR and other NATO experts in the US government were crucial. Burns recalled that “particularly in the second half of 2002 […] we push[ed] hard, my mission and I, assisted by my military colleagues in Europe [the SACEUR is dual-hatted as Commander of US European Command⁹⁵], to convince Washington that we ought to think of a NATO deployment”.⁹⁶ A former US National Security Council (NSC) staffer observed at the time that “NATO was anxious to expand their role, and so were people inside DoS and DoD working on European/NATO directorates who were suddenly important again”.⁹⁷ One of them was a senior DoS official responsible for European affairs who also advocated NATO assumption and identified Frank L. Miller as another crucial NATO expert who during this time “kept insisting that we needed NATO”.⁹⁸ Miller was Special Assistant to President Bush and Senior Director for Defense Policy and Arms Control on the NSC staff. Immediately before these postings, he had served at NATO HQ and previously had been responsible inter alia for NATO issues at the DoD.⁹⁹

By early 2003, in the struggle for NATO assumption, Rumsfeld had become more favourable towards an alliance command over ISAF.¹⁰⁰ Yet, while the per se more multilaterally-oriented DoS representatives endorsed NATO assumption, Rumsfeld still did not dare to make a final decision.¹⁰¹According to a US military officer and eyewitness, the efforts of the most senior US military seconded representative, SHAPE’s SACEUR General James Jones as successor to Ralston, eventually won him over. Aided by DoD officials with responsibilities for NATO, who acted as “NATO lobbyists” in the Pentagon, Jones finally “sold Rumsfeld on turning Afghanistan over to NATO”.¹⁰² Eventually, the numerous position-takings of the seconded representatives and administrative cadre created the necessary momentum to make major national representatives of all member states agree on 16 April 2003 to assume command of ISAF by 11 August.

Expanding ISAF without strategy

After NATO had assumed ISAF command, the question of an expansion of the mission beyond Kabul remained contested. Alliance representatives strongly lobbied for expansion. To achieve this, SHAPE officers proposed a broad interpretation of the area of operations provided in ISAF’s UN mandate and drafted plans for expansion. Yet, most member states did not support expansion.¹⁰³ Especially DoS representatives saw the need to expand,¹⁰⁴ but Rumsfeld and his closest aides still feared to be pinned down in Afghanistan through NATO while they needed troops for the looming Iraq intervention.¹⁰⁵ They therefore called for their allies to assume some of its OEF PRTs, but most Europeans shied away from getting involved in costly operations in the Afghan countryside. Only its closest ally, the United Kingdom, and New Zealand agreed in early 2003.¹⁰⁶

During this time, major Belgian, French and German representatives criticised the unlawful and unilateral tendencies of the US “war on terror” and rejected American attempts to rally UN and NATO support for their Iraq invasion plans. In the eyes of most contemporaries, the conflict seriously “damaged” transatlantic relations.¹⁰⁷ After the US had begun the Iraq intervention in March 2003, the Germans therefore “sent numerous signals they wish to move the relationship [with the US] forward” as a DoS official noted.¹⁰⁸ To this end, Chancellor Schröder, who had declined similar requests before, in May 2003 vaguely promised Secretary of State Powell to advocate an expanded role of ISAF. Afterwards, Minister of Defense Struck indicated that Germany could assume command of a PRT. Yet, for the following months, serious discord between the German ministers of foreign affairs, defence and development, who all had to contribute to the PRT, jeopardised the whole plan. The ministers of foreign affairs and development insisted on a location far distant from Kabul to increase the “visibility” of their effort. Contrary to that, the minister of defence—relying on military security assessments—judged these areas as too dangerous and instead sought proximity to the capital.¹⁰⁹

In this conflict, the only military member of the alliance’s administrative cadre, Chairman of the NATO Military Committee, General Harald Kujat, appears to have made the decisive position-taking to achieve ISAF expansion. Having served as German chief of defence before, he contacted his former superior Struck and recommended Kunduz Province as a proper location for a PRT.¹¹⁰ Though Kujat’s assessment later proved wrong in light of the intensifying violence in this province, Struck’s decision for Kunduz created the necessary inter-ministerial consensus since this province was far enough away from Kabul and therefore “visible”.¹¹¹ Glad that Germany had made a first step by assuming command of an ISAF PRT, other major national representatives followed suite and decided on ISAF expansion through a net of PRTs in October 2003.¹¹²

Looking at the practice of seconded representatives at NATO and members of its administrative cadre in the period outlined in this section, it becomes evident that they were driven by a common sense that NATO should assume command of ISAF and expand it to the whole country. Viewed from an observant outsider’s perspective, this made sense in respect to the fact that NATO command brought more continuity and that major Afghan issues could not properly be addressed from Kabul alone. Yet, the core question the observant outsider would pose is: What issues were these? What exactly did NATO want to achieve in Afghanistan? The answer is that, as demonstrated, major representatives of the member states mostly did not have Afghanistan-related goals or—if they did—they did not agree on these goals with their allies. They were also not willing to provide the necessary resources. In a rational decision-making process, seconded representatives at NATO and members of its administrative cadre would then have concluded—and advised major national representatives accordingly—that it makes no sense to engage the alliance in Afghanistan.

What happened instead was that this “rational” option appeared untenable through the filter of the habitus of the seconded representatives and the representatives of the administrative cadre. What appeared sensible to them instead was an option that—at least in the short run—maintained or increased the value of their cultural capital as NATO experts. After the decision to expand ISAF, they drafted a “Longer-term Strategy” that depicted the mission as doable but only included a very broad goal:

A self-sustaining, moderate and democratic Afghan government, in line with the relevant United Nations Security Council resolutions, able to exercise its authority and to operate throughout Afghanistan, without the need for ISAF to help provide security.¹¹³

As the US Central Intelligence Agency had already criticised with regard to the same formulation in the US Afghanistan policy guidelines of that year, the terms “moderate and democratic” were too vague to serve as a measurable end state for what the Afghan state should eventually look like.¹¹⁴ It was also not clear whose “security” to what extend was meant. The description did not prioritise between the competing goals of creating a “moderate and democratic” state on the one side and one that was authoritative as well. In addition (and even more fatally), the NATO strategy did not establish why these changes to Afghanistan mattered for the alliance’s members. Reflecting on this issue 13 years later, Ambassador Burns admitted that they had not considered this fundamental issue:

After 2003 and 2004, once we were fully engaged in both wars, I can’t remember us ever saying: “Should we be there? Are we being useful? Are we succeeding?” […] I don’t remember us asking those questions, really until President Obama’s administration, and I was out of government by then, and I think just having looked at it, I think we would have done better if we had made some more specific, strategic assumptions.¹¹⁵

Besides this lack of concrete political goals on the side of major national representatives, the strategy was deficient because “[t]here was no sense of common purpose” according to a NATO official.¹¹⁶ In particular, the US representatives gave more weight to counter-terrorism as a goal while the (continental) Europeans tended to follow “a vague idea about the mission that it was something like ‘protected development.’”¹¹⁷ In effect, they only agreed on a broad goal that was too vague to serve as basis for a coherent strategy. Based on this broad goal, NATO representatives only formulated sets of actions and processes like training for the Afghan National Army, a disarmament, demobilisation and reintegration program, counter-narcotics activities etc. But due to the lacking precision of the goal, these processes could not be prioritised or measured.¹¹⁸ SHAPE officers planned according to the scarce capabilities that member states were willing to contribute—not the other way around, defining necessary measures and then matching them with adequate resources or deciding that necessary measures could not be achieved with the resources at hand.¹¹⁹

As confirmed by several witnesses, this general constellation did not much change over the following years.¹²⁰ The strategic SHAPE operational plan (OPLAN) 10302, which was approved by the NAC on 14 April 2004¹²¹ and maintained until the end of ISAF, included almost the same vague and ambitious political end state plus sets of non-prioritised tasks.¹²² Canadian Lieutenant-General Rick Hillier, who served as ISAF commander in 2004, in retrospect aptly characterised the paradoxical practice of NATO representatives he witnessed:

People at the NATO headquarters were talking about all kinds of pie-in-the-sky ideas for Afghanistan, but they had no strategy, no clear articulation of what they wanted to achieve, no political guidance and no forces.¹²³

As they could not change the lack of political goals and unity among member states, the habitus of NATO representatives produced a common sense according to which strategy was not the problem; instead, their preoccupation was with resourcing and managing the mission. An article by a senior member of NATO’s administrative cadre from the late years of ISAF reflects this attitude. Instead of discussing the lack of political goals, he concluded: “What we got wrong was not the planning but the resources”.¹²⁴

By focusing internal NATO debates about ISAF on means instead of ends, NATO representatives contributed to spreading this means-centric common sense more widely among and within member states. This becomes clear by looking at the position-takings of the most influential seconded representatives and members of the administrative cadre on the issues that proved decisive for making NATO responsible for almost all security matters in Afghanistan: the formal expansion of ISAF to the whole country as well as the actual generation and deployment of the necessary forces. Their position-takings, in effect, strengthened NATO’s role. They mostly supported major representatives who, for other reasons, followed a similar policy on these issues.

Even before the initially limited ISAF expansion to northern Afghanistan, “key NATO leaders together with SHAPE and JFC [Joint Forces Command Brunssum] planners had waged a campaign […] to develop a comprehensive plan that would establish the basis for a broader expansion of the ISAF”.¹²⁵ General John Abizaid, who was responsible for operations in the greater Middle East (including Afghanistan) as Commander of US Central Command and sought to free US forces for Iraq, supported this effort and worked hand in hand with SACEUR Jones.¹²⁶ He asked US Secretary of Defense Rumsfeld for political support on the matter in November 2003.¹²⁷ Afterwards, senior US DoD officials lobbied accordingly among the member states.¹²⁸ Finally, as prepared by SHAPE planners, major representatives agreed to an expansion in five stages that was incorporated in OPLAN 10302.¹²⁹

Yet, during the coming years, non-US major representatives consistently remained reluctant to provide the military forces necessary to execute expansion.¹³⁰ In response, Secretary General Robertson applied “much pressure” on major representatives and publicly demanded that “expansion must be credible”.¹³¹ Similary, since January 2004, his successor Jaap de Hoop Scheffer publicly urged the member states that “they must now provide the military assets needed to carry it out”.¹³² Until the end of his service as SACEUR in 2006, General Jones also demanded more forces from member states.¹³³ Apparently to minimise the possibility that member state representatives could duck away by pointing to troop contributions for NATO missions in other parts of the world, his SHAPE planners organised the first Global Force Generation Conference in November 2004.¹³⁴ These measures as well as contributions and pressure from the US finally generated the necessary forces to execute expansion.¹³⁵ Yet, force generation remained a problem, leading Secretary General de Hoop Scheffer and SACEUR Jones in 2005 to “both call for common funding of operational expenses by Allies”.¹³⁶ While this would have dramatically strengthened NATO’s autonomy by giving it a much larger budget, the major national representatives rejected this breach of the principle of “costs lie where they fall”.

Conclusion

My detailed analysis of the initial decisions that paved the way for NATO in Afghanistan has demonstrated that some of the most common explanations in political science, those that argue from an aggregated state or international organisation perspective, miss the point in many respects. Looking at the practice of the very actors who constitute NATO, it becomes evident that it was not only the “unilateralist” major US representatives who were initially reluctant to invoke Article 5 of the Washington Treaty. Rather, this reluctance was shared by representatives of most member states. In fact, only the seconded representatives and NATO’s administrative cadre clearly pushed the issue. I also demonstrated that NATO’s assumption of ISAF command on the one hand, and ISAF expansion on the other, should not be conflated. They did not go hand in hand with a total revision of the DoD’s position. Major national representatives of the most significant member states were still very reluctant to vote for assumption and later expansion. It was not a rational cost–benefit assessment that guided their eventual decision in favour of NATO assumption and not all saw the need to reconcile the relationship with the US through ISAF expansion.¹³⁷ Only the interference of seconded representatives and the administrative cadre in favour of a NATO solution appears to have outbalanced the critics in member state capitals. My analysis of NATO policy-making after assumption of ISAF command shows that, even though there was no agreed upon political goal, seconded representatives and the administrative cadre pushed for an intensification of the mission.

These insights provide evidence to answer the research question of why NATO became so intensively engaged in Afghanistan without concrete political goals. Certainly, the immediate reason for this political deficit originated in major national representatives’ lack of political ideas for Afghanistan or disagreement among them on which ideas to follow; this is in line with the findings produced by most detailed studies on the national policies of ISAF contributors.¹³⁸ Yet one also has to consider an important prerequisite that made it possible to not perceive the lack of shared political goals as an impediment to ISAF. This prerequisite was the unquestioned common sense that evolved since “9/11”, according to which NATO had to be engaged in Afghanistan as intensely as possible − even without clearly defined political goals. This common sense was first and foremost created by the habitus of seconded representatives and NATO’s administrative cadre. It unconsciously guided their position-takings in a way that kept or increased the importance of their cultural capital as NATO experts, meaning that they worked towards getting the alliance as intensely involved as possible.

Only with the (temporary) support of major representatives of important member states were these position-takings successful in steering NATO’s course in the desired direction. Yet, while these major representatives were only occasionally interested in enhancing NATO’s role in Afghanistan, namely when this stance supported their national policies, the seconded representatives’ and administrative cadre’s position-takings created the necessary consistency to get the alliance more intensely involved. A preliminary look indicates that from 2006 until 2014, NATO debate on ISAF centred around means instead of ends: how the burden of fighting should be equally distributed among the member states; what operational concepts like the “comprehensive approach” or “counterinsurgency”—often wrongly termed “strategies”—should be followed, or how to “transition” to Afghan responsibility.¹³⁹ It therefore appears that in the long run, seconded representatives and administrative cadre de facto took over much of NATO’s Afghanistan policy making, as political guidance from major representatives became increasingly less needed and desired.

My empirical findings demonstrate the general value of practice theory. In particular, they demonstrate the explanatory power of a structuralist approach to practice theory that combines the usual focus on real world actors with closer attention to field characteristics and takes seriously the Bourdieusian assumption of a mutual constitution of actor and structure. This more structuralist approach also contributes to our theoretical understanding of the character and perseverance of NATO after the Cold War. A structuralist perspective suggests that the formation and endurance of NATO can be understood in analogy to the modern statebuilding process as interpreted by Bourdieu. He saw this process not as a conscious project but as an unintended result of the colluding interests of monarchs with those of their administrative cadres. Accordingly, especially the kings’ jurists developed an interest in defining the state as the superior authority in all fields of life. Through their judgements, they contributed to establishing a common sense that saw the state and its categories as an unquestionable natural thing.¹⁴⁰ The perseverance and transformation of NATO from a mere military alliance into a complex international organisation should be understood as the result of a similar constellation.