International organization at war: NATO practices in the Afghan campaign

The emergence of practices as a research agenda

Adler and Pouliot define practices as ‘socially meaningful patterns of action which, in being performed more or less competently, simultaneously embody, act out, and possibly reify background knowledge and discourse in and on the material world’ (2011a: 6). Practices have four main characteristics: they are a performance, are patterned, are interpreted along similar standards by groups of individuals and weave together the discursive and material world. Surveys of this field identify a number of different traditions of practice-based research (Bueger and Gadinger, 2014; Pouliot and Cornut, 2015), but there is a common basic understanding of practices as socially meaningful patterns of activities. This article does not attempt to weigh in on the debate by following one school or another. Instead, it uses practices as a ‘sensitizing concept’ (Bueger and Gadinger, 2014: 78) for analyzing concrete empirical phenomena.

Practices can be considered as explanandum or as explanans – as the result of some driving force or as a mechanism of social transformation. In this article, I document cases of both situations: I refer to cases of practices resulting from previous social dynamics (explanandum), but I also illustrate how practices made possible some political outcomes that would have been otherwise impossible (explanans). As explanans, practices can trigger change in three areas: ‘subjectivities (e.g. preferences, dispositions, or intentionality), in practice themselves, or in social orders (e.g. structures, domination patterns, or discourse)’ (Adler and Pouliot, 2011b: 18). The empirical portion of the article distinguishes between practices considered as explanandum or as explanans.

Critics often claim that the causal power of practice is difficult to establish, as no single practice can single-handedly explain an event. In fact, ‘strictly speaking, diplomatic practices do not ‘cause’ specific outcomes’ (Pouliot and Cornut, 2015: 309). It is better to conceive of practices as creating a context in which things become possible, and others become impossible: ‘causality works on this terrain not in terms of clearly identifiable causes but in terms of conditions of possibility’ (Kuus, 2015: 380). As such, it is pointless to require practices to explain a specific event (which is ‘caused’ by a multiplicity of factors), but it is worth asking which practices made certain events possible, and how. The objective of the research is neither to give the ultimate explanation of all facets of the ISAF intervention, nor to favor a practice-based approach over power-based or norms-based explanations. It is clear that power, norms and institutions are all important factors that help to explain the ISAF mission. Yet, the analytical plane is different: the point is to illustrate how those practices gave the intervention its specific shape.

Studying practices within NATO

As an intergovernmental organization focused on military issues, NATO embodies a number of bureaucratic routines and procedures. In particular, specific decision-making processes, which involve diplomats, military representatives and NATO civil servants, are adopted and refined over time, and influence the conduct of any campaign. In their study of the Libyan intervention, Adler-Nissen and Pouliot (2014: 895) observe how power works in practice. They identify three different dynamics: claiming competence occurs ‘when players seek to establish their mastery of the game by framing particular issues and positioning themselves as leaders’; social negotiation refers to ‘other players challeng[ing] the skillfulness of a player’s moves as part of an ongoing struggle’; and wielding influence is when ‘players cash out their socially recognized mastery in the form of non-coercive influence’. Below, I show how actors claimed competence, in particular in the US, and how this claim was challenged through various practices.

Other types of practices are also studied. First, while some practices involve military officers, most decision-making processes within NATO rely on diplomats. As Iver B Neumann illustrates in his work, diplomats share a distinctive set of practices governing behavior: rituals, language, stylized hospitality, etc. In particular, the diplomatic discourse is of great importance: ‘diplomats are specialists in meaning. They create, manipulate, convey and interpret words and symbolic actions. Such work is incomparably more complex than the mere conveying of messages, or even the negotiation of specific issues, though it encompasses both’ (Gould-Davies, 2013: 1465). Hence, studying discourses and the use of language is particularly meaningful when studying diplomatic work, especially considering that the circulation of competing narratives about Afghanistan may have had a negative impact on the outcome of the intervention (Friis, 2012).

There is also a spatial dimension to social activities. Iver B Neumann has illustrated the idea that new diplomatic arenas beyond the traditional conference table or press conference shape the reconfiguration of global diplomatic activities. Neumann argues that ‘the ever-increasing density of global life keeps changing the old, familiar diplomatic sites and creating new ones by bringing in agents, bringing on new procedures and dismantling old ones’ (Neumann, 2013: 3). It is, then, appropriate to observe how the spatial dimension affects practices and thus contributes to the shaping of the social world when it comes to the Afghan War.

These three dimensions (routine procedures, language, space) are found simultaneously in any activity performed by the agents involved in the management of the Afghan War. Yet, for the sake of clarity, they will serve as organizing devices in order to illustrate how practices operate and are summarized in Table 1 above: while I am ‘slicing up’ different events according to these three dimensions in order to document a specific practice, it must be remembered that these dimensions are always simultaneously at play.

OPEN IN VIEWER

Table 1.

Practices within NATO HQ.

	Procedures	Language	Space
Practices as explanandum	Decision by minority US primacy	Bilingualism	Proper seating
Practices as explanans	Informing allies Accommodating partners	Surprise announcements Copy-pasting

The material for this research has been collected over six months of participant observation within the operations section at NATO HQ while I was working as a NATO civil servant with access to the Operations Policy Committee (OPC), a committee of junior diplomats preparing policy documents and tasked to adopt the same in relation to the Afghan campaign. Most of the time, policy documents were agreed upon at the OPC level but, on specific occasions, when a consensus could not be reached, the discussion would be brought up at the North Atlantic Council (NAC) level. The OPC can meet in NATO format (allies only) or in ISAF format (allies and partners). The OPC formulates recommendations, agrees on texts that are subsequently approved by the NAC and functions as the civilian oversight to the work of the Military Committee. While strategic decisions are made by the NAC, the OPC is responsible for the vast majority of routine policy-making decisions, and also has some shaping power regarding the strategic decisions. During these discussions, I was in charge of what is called the ‘live-editing’, which consists of displaying on a screen the modifications to policy documents suggested by diplomats, so that everyone around the table can see them. As such, I was not directly involved in the shaping of the documents themselves as I was not participating in the debate (which is not the role of a NATO civil servant), but I had the privilege to observe the social dynamics at play during these numerous meetings. Part of the daily work also included liaising with the NATO chain of command and with national delegations.

During this assignment and after it ended, I have been able to conduct more than 100 semi-structured interviews with NATO policy-makers, as well as with diplomats and military representatives from the US, the UK, France, Germany, Italy, the Netherlands, Canada and several other countries. The interview sample was comprised of individuals at all levels of the NATO chain of command: country capitals, NATO HQ, Supreme Headquarters Allied Powers Europe (SHAPE) in Mons, Allied Joint Force Command (JFC) in Brunssum and ISAF in Afghanistan. Interviews were conducted under the condition of anonymity between 2012 and 2014. I use a standardized anonym classification to reference the interviews. ¹

Dominating the military campaign

There is little doubt that the Afghan campaign was dominated by the US. The US was the major force provider on the battlefield, engaged in some of the toughest combat activities and was perceived by the local and regional actors as the key political interlocutor. This predominance translated into – and was the product of – a number of practices at the tactical/operational (ISAF, SHAPE) and strategic/political (NATO HQ) levels.

First, and foremost, US officers heavily dominated the chain of command. Since 2006, the commander of ISAF (COMISAF) had always been an American, with some of the most high-profile generals of the last decade, such as David Petraeus and Stanley McChrystal, commanding in Afghanistan. The COMISAF was designated by the president of the US and approved by the US congress, leaving the allies with no control over the process. This was accepted by other countries, who recognized that the US paid the greater price in Afghanistan and was, therefore, entitled to exert domination, as long as it did not appear as such to the public. The COMISAF also had a dual hat because he had been responsible for commanding US forces in the country (including Special Forces operating in OEF) since 2008. As such, COMISAF had two reporting chains of command that directly led to Washington: through the US Central Command (CENTCOM) and the US European Command (EUCOM). There was a degree of communication between CENTCOM and the ISAF that was far from negligible, and aspects of the conduct of the ISAF campaign escaped the control and oversight of the allies. ² This widespread feeling of US domination led to some resentment when McChrystal, and to a lesser degree Petraeus, held the position of COMISAF. McChrystal, for example, was famous for very reluctantly allowing non-US or non-British officers in his office. General Petraeus’ and General Allen’s tenures improved this situation, and were more positively welcomed by the allies. ³

This US domination was also reflected in the design of the Operational Plans (OPLANs) for the campaign. In theory, an OPLAN should be established by the COMISAF, receive inputs from SHAPE and finally go to the Military Committee (MC) for approval by the member nations. There should, therefore, be opportunities for planners at SHAPE and for member nations to modify the OPLANs designed by the COMISAF. In reality, the OPLANs were left practically unchanged by the Supreme Allied Commander in Europe (SACEUR, who heads SHAPE and is also a US officer), and arrived in Brussels for approval by the member nations loaded with the symbolic weight of already having been approved by the field commander and by the strategic commander at SHAPE. It was, therefore, extremely difficult for nations to contradict what was supposed to be the military expertise of NATO’s structure. Moreover, the OPLANs were usually already being implemented in Afghanistan before being officially validated at NATO HQ level, showcasing the COMISAF’s confidence that they would be left unchanged. Many nations’ representatives and desk officers within the International Military Staff (IMS) ⁴ complained that SHAPE did not want to alter the COMISAF’s plans and de facto restrained its role as an ‘enabler’ for ISAF; this choice had the perverse effect of increasing the COMISAF’s status and leaving little room for member nations to criticize and amend the OPLANs prepared in Kabul. ⁵ The US domination over the establishment of the OPLANs was an important factor in the US claim of competence, through which the US positioned itself as leader of the intervention.

And yet, the OPLANs were under more scrutiny from ISAF nations when General McChrystal served as COMISAF than during the tenures of General Petraeus and General Allen. McChrystal’s behavior was perceived as disrespectful and was resented by other countries, who wanted to scrutinize more closely what was produced in Kabul before endorsing it. This led to a delay in the endorsement of one particular OPLAN that was even sent back to SHAPE for improvement after some nations declined to endorse it. ⁶ In particular, they criticized the use of Special Forces and night raids by McChrystal, and succeeded in imposing a mechanism to oversee such missions. However, OPLANs produced by Petraeus or Allen never underwent such detailed scrutiny. This suggests that practices can be performed more or less competently: McChrystal’s poor performance of the practice of US primacy led to a counter-reaction by other countries that had a tangible result in delaying the adoption of an OPLAN and modifying it on an important issue. Conversely, those countries did not resent Petraeus’ and Allen’s performances, and the adoption of OPLANs went ahead smoothly under their tenures, which illustrates how the imbalance of power distribution favoring the US is not automatically translated into a desired outcome; it has to be performed in a competent way by the relevant actors to achieve its full effect.

The political hierarchy

In addition to US military domination, the US political leadership shaped, to a large extent, NATO’s policy-making process. Policy decisions regarding the ISAF campaign (as well as other NATO operations) were made by one of the junior NATO committees, the OPC. As in every diplomatic circle, there was a hierarchy of status within the OPC, depending on the nations diplomats represented, their personal skills or their seniority within the committee. At NATO, there was an established practice of ‘minority decision-making’, which left responsibility for important decisions to a small number of countries. This practice took several forms. All interviewees agreed that the key players in the OPC were the US, the UK, France and, to a lesser degree, Germany and Italy. The Netherlands, Canada, Denmark, Sweden, Turkey and Australia belonged to a second circle of important nations. This means that out of 50 nations comprising the ISAF, only 10 to 11 spoke regularly during the OPC meetings and actually shaped policy decisions. Some nations, be it because of a lack of diplomatic manpower or of interest, did not even attend most of the OPC meetings. There is, then, a clear hierarchical relationship at NATO HQ between nations, ranging from a small group of proactive countries effectively shaping policies and a larger group that remained silent. This observation is consistent with and reinforces the observation that ‘the alliance’s decision rules would seem to encourage deal-making by powerful and/or passionate members who want to get their way’ (Auerswald and Saideman, 2014: 35). Not only did decision-making rules favor powerful and/or passionate members, they also reified a social hierarchy within the OPC that was based on military contributions and being a ‘good ally’ (Pouliot, 2016). As was described by one diplomat from a small European country: ‘everyone knows his place: we do not expect a country to make an inflammatory statement when its military contribution is close to zero’. ⁷ Some countries were expected to shape the outcomes, while others were expected to remain silent. This indicates that the formal hierarchy within the OPC is a function of the social recognition that comes with significant contributions on the ground.

The NATO bureaucracy also contributed to this hierarchical management of the campaign. The real power of NATO’s bureaucratic structure lies in the fact that NATO civil servants draft the documents that will subsequently be discussed by the OPC and the NAC. Drafters are usually very careful to accommodate nations’ concerns before the document is circulated to them, and key nations then become part of the drafting process. However, civil servants know all nations’ preferences whereas an individual nation is typically uncertain about the others’ intentions; this asymmetry of information is in favor of the civil servant, who can exploit it to advance his/her own preferences while still being seen as a neutral expert figure. ⁸ There was a habit of coordinating with key allies before the OPC meetings and, as a result, the informal hierarchy of the decision-making process was then made explicit through this selective consultation of key nations. Interestingly, this consultation of key nations was not supposed to take place because it would acknowledge a hierarchy – NATO, like other international organizations, is based on the fiction of legal equality among member states. In fact, everybody knew that such meetings took place, but nobody mentioned them. Not mentioning these meetings is in itself a sign that keeping the fiction is important for NATO cohesion, hence showing that this cohesion is valued by the member states. There is, therefore, an established practice of leaving a minority of countries to handle policy decisions, either at the OPC or during the pre-OPC meetings, which all actors are aware of, but which is not openly acknowledged.

Among this minority, the US was widely considered the most influential; this is another example of the practice of US primacy. The US had been the driving force within the OPC and the NAC regarding policy developments on topics such as strategic partnerships with Afghanistan, the size and format of the ANSF, funding schemes for the government in Afghanistan and the post-ISAF operation. When drafting documents, the US proposal was routinely the first to be welcomed by the chairman, who then invited opinions from other countries. Implicitly, this prioritization of the US proposal over those of other nations both represented and reproduced the US position as a primus inter pares.

The ability of US policy-makers to perform their leadership role in the Afghan campaign, suggesting policy developments at the NAC or OPC level, facilitated the acceptance of this domination by other countries. Yet, US representatives can perform their role more or less competently. For example, after a round of negotiations during which the US delegate did not seem to pay much attention to the proposal of a small European country’s delegate (although this delegate obviously had high hopes about his suggestion), the same delegate, as a matter of exception, asked to speak first during the next meeting in order to reiterate the same proposal which then constituted the basis for further discussions. Here, the incompetent performance of the practice of US primacy at the negotiating table by the US delegate led to another round of negotiations, in which the US representative was in an unfavorable situation as her proposal was not the first to be discussed and was, in fact, substantially transformed.

A language practice reveals the underlying dynamics of coalition warfare. NATO is officially a bilingual institution, and work can be carried out in either French or English. In practice, since the arrival of Eastern European countries whose representatives usually don’t speak French, English has become the dominating working idiom. Yet, the French representatives (who are instructed to speak French) continue to use it during the OPC and NAC meetings. Other representatives, coming from countries such as Italy, Germany, Luxembourg, Belgium or Turkey usually have a good command of French. Interestingly, depending on the position they want to advocate, such representatives will switch from English to French or back to English. Everything is determined by the position of the US representative. If other countries agree with the US delegate, they will express their agreement (or suggestions for minor changes) in English. If they disagree strongly with him/her, they will express this in French, thus forcing the majority of diplomats around the table (including the US representatives whose command of French is usually minimal) to use the services of the live translation. What might appear at first glance to be purely anecdotal is, in fact, significant, because it is an example of the way the US domination of the campaign was either accepted or challenged. Speaking English is for those countries who do so an acceptance of the framing of an issue by the US, while speaking French is a way to challenge the US-preferred policy positions. Moreover, this practice of bilingualism also formalizes an informal hierarchy between countries capable of formulating propositions in a different language and countries only able to speak English. It is no surprise that representatives able to speak both French and English usually come from countries deemed important within NATO (France, Italy, Germany, Turkey).

In terms of spatial practices, the observation of the NAC meetings revealed interesting dynamics. Because the ISAF is comprised of more than 40 different nations, meetings cannot be held in the traditional NAC meeting room, which would be too small. Thus, the meetings were located in a second meeting room, large enough to accommodate everyone. In addition to space at the table, it was also necessary to accommodate staffers (junior diplomats, etc.), but also any NATO civil servants who had an interest in attending the meeting. A number of seats behind each ambassador were reserved for staff from his/her own delegation, and NATO civil servants were supposed to sit in the non-reserved places. Yet, the sitting location embodied the hierarchized conduct of the Afghan campaign. First, not all nations had an equal number of seats assigned for staffers, although this should officially have been the case. Major nations (US, UK, France, Germany, etc.) received more assigned seats than others, thus physically showing the unofficial hierarchy at play. This practice derived from the pragmatic observation made by NATO staff that some countries had more staffers than others, and thus needed more seats. The NATO staff in charge of preparing the NAC meeting room are purely logistical and administrative assistants. Hence, it is particularly telling that they decided on their own initiative to adapt the layout in the NAC room to what they perceived was needed in order to facilitate the meetings, without realizing they were breaking the NATO regulation stipulating that each country must have an equal number of seats; ⁹ thus, the reification of the social hierarchy at play in the conduct of the Afghan War has a spatial dimension that is both perceived and reproduced even by the NATO logistical staff. It is even more interesting to observe that state representatives and NATO civil servants noticed this progressive rearrangement without anyone complaining about it; the hierarchical nature of the policy-making process was translated into spatial practices which, in turn, helped reinforce this social stratification.

Considering practices as explanandum helps in understanding how the imbalance of the distribution of power operates by reifying a hierarchy of social status, but also illustrates that the translation of the imbalance of power into outcome is far from automatic: US representatives have to competently perform their roles for their power advantage to achieve its full effect, and this unequal distribution of power can be challenged and circumvented in original ways.

Going to war

As was previously noticed, NATO became officially involved in Afghanistan in 2003 after a decision taken in April to assume command of ISAF by August. This decision was the result of two parallel sets of events: the first being the gradual acceptance by NATO’s members of the potential necessity of intervening outside of NATO’s traditional geographical comfort zone; the second being a sense of urgency as the ISAF was running out of volunteers to command it. The 9/11 attacks triggered a new mindset within the alliance, institutionalized in the 2002 Prague Summit, during which heads of state and government launched a transformation process for the alliance, including its enlargement to accommodate new members and the creation of an expeditionary warfare capability called the NATO Response Force (NRF) (Moore, 2007). NATO’s bureaucratic structure adjusted to these new incentives, and the operations division started to grow both in terms of numbers of employees and in proactivity: more and more proposals for change and reform (especially regarding expeditionary capabilities) began to circulate within NATO HQ. ¹⁰

In the meantime, in 2003, governments suddenly realized that the ISAF was running out of nations willing to take the lead. The first two nations to have assumed command of ISAF, the UK and Turkey, had deployed their own national divisional headquarters. However, Germany and the Netherlands, who took over ISAF command from Turkey in February 2003 and were supposed to assume it until August, wanted to deploy their NATO-assigned multinational corps headquarters (some headquarters are placed under the Alliance’s disposal on a permanent or temporary basis under specified readiness criteria). According to Rynning (2012: 86–87), ‘Unless NATO stepped in, force generation and coordination, logistics, and operational command would all fall on the two host nations. They thus asked for NATO’s help’. This political desire by Germany and the Netherlands to involve NATO crossed paths with an established practice within NATO HQ: the sharing of information about the ISAF mission.

Only NATO countries had assumed command of ISAF, and it had become usual for the UK and Turkey to report to their NATO colleagues in formal settings (such as the NAC) or in informal discussions on their own initiative – ISAF not being a NATO operation at the time, there was no formal procedure. This practice of informing allies created a habit of discussing ISAF issues within NATO, thus facilitating the transition to a NATO-led ISAF. The practice of reporting to the NAC and informing other NATO members of the mechanics of managing the ISAF mission reified background knowledge on the opportunity of ‘out-of-area’ operations, showing that they were, indeed, possible. Several interviews with diplomats from key NATO nations confirmed that they regularly conveyed back to their capitals the positive impression of ISAF they had gained from their Turkish and British counterparts. Representatives of two of these countries showed me the internal confidential documents advocating the takeover of ISAF by NATO that were addressed to their ministries of defense and foreign affairs. Both of them mention the positive portrayal of the ISAF mission at the NAC level by the UK and Turkey as an argument in favor of NATO taking over, and it seems likely that other countries’ internal documents mention a similar argument. What matters here is the fact that the practice of reporting to the NAC (although NATO was not yet engaged in Afghanistan) wove together the discursive and the material world in relation to out-of-area operations; while all NATO member states were debating out-of-area operations, Turkey and the UK were actually conducting one and gave other allies an idea of what such operations would look like. In this case, the practice of reporting created a condition of possibility: it seems very likely that the NAC would have been much more reluctant to take command of ISAF without the flattering picture displayed by the UK and Turkey. ¹¹

This finding is important, because it complements the purely power-based explanation of the decision to engage NATO in Afghanistan. The decision is, therefore, the result of a combination of political incentives (reconciling both sides of the Atlantic after the Iraq crisis and using NATO’s help in the case of Germany and the Netherlands) and a favorable context. The context consisted of a more intervention-prone organization that was already familiar with ISAF through the practice of reporting to the NAC. This does not dismiss the political incentives that drove NATO to intervene in Afghanistan, but practices explain the establishment of the context that made possible the adoption of such a decision.

Defining a strategy for NATO

Summits are important events in NATO’s organization. They are an opportunity for heads of states and governments to meet and to provide strategic guidance to the alliance. Ministers of defense or ministers of foreign affairs meet on a more regular basis (‘ministerials’, in NATO parlance), but summits are the real pulse of the alliance: a relatively short space of time (usually two days), prepared months in advance by members and NATO’s bureaucracy, during which important decisions are made. A summit has all the elements of a classical tragedy (unity of time, space and action) as well as a touch of drama, and key issues that cannot be resolved at the level of diplomat or minister are held in reserve for a summit. Summits guide the alliance’s work and direct energies within NATO HQ for at least six months before the event. ¹²

I am focusing here on the effect and importance of language in the practice of surprise announcements. In pushing for the ratification of a carefully crafted protocol, the president of Afghanistan, Hamid Karzai, was the first to speak during the sessions devoted to ISAF at the Bucharest summit of 2008. His introductory remarks were followed by the NATO Secretary-General’s speech, before presentations by the Senior Civilian Representative (SCR) and the COMISAF. Hamid Karzai surprised everyone by announcing that Afghan forces would be in charge of the Kabul Security Zone within a month, following the end of the French tour in the Afghan capital and France’s redeployment to the Kapisa and Surobi regions. This surprise announcement triggered the drafting of a transition strategy that was designed primarily by Mark Sedwill during his tenure as SCR. ¹³ However, there is more to this story than meets the eye. After having made such a strong impression, President Karzai became accustomed to making surprise announcements, and did so again during his opening remarks in 2009, 2010 and 2012, although none were as surprising as that of 2008. ¹⁴ The net effect was a realization by NATO countries that the Afghans now had agency and would no longer be treated as faithful executants of orders over which they had no control. Of course, this must be put into the context of a gradual and mutual disappointment in the relationship between Hamid Karzai and the US, which culminated in the crisis over the results of the 2009 ¹⁵ presidential election. Nevertheless, the opening remarks by Karzai became a moment both expected and feared in national capitals in 2009, 2010 and 2012, as nobody could be sure that the Afghan president would not make the headlines again with a new, and never previously agreed-upon announcement. As such, ‘accommodating the Afghans’ became a mantra for NATO diplomats when negotiating key policy documents such as the joint declarations with Afghanistan at the Chicago and Wales summits, or the Status of Force Agreement (SOFA) for the Resolute Support Mission (RSM), which replaced ISAF as of January 2015. This is a clear case of practices triggering change in social orders, with the Afghans gradually being recognized as having agency, which implied wholly different negotiation dynamics (Cowper-Coles, 2012; D’Amécourt, 2013).

Negotiating ISAF

As mentioned, NATO decision-making procedures emphasize a hierarchy of states, and reward the most proactive countries, as well as those more committed on the ground. This had an impact on the role Australia, New Zealand and Sweden started to play in both the OPC and NAC. Through a recognized commitment on the ground and a valued contribution to policy developments, these countries managed to get to a position where they were considered as almost equals to the NATO allies in terms of prerogatives and status. For example, while in 2006 it was standard procedure for NATO allies to discuss and agree on policy papers in a ‘NATO members only’ format first, and then discuss it with partners, this gradually changed and partners were routinely involved with the very early phases of a document on an equal footing with the allies. This new practice was institutionalized in the rewriting of the Political–Military Framework (PMF) in 2011, a guidebook detailing NATO procedures for dealing with crisis situations that now involve NATO partners at every stage of the planning process. This important shift is a clear example of how practices of policy-making circumvent the official legal hierarchy between states created by the division between NATO members and partners. As mentioned above, another hierarchy based on contributions, skills and proactivity was actually at play. As one interviewee explains: ‘back in 2006, nobody would have thought that Australia would be sitting at the NAC table on an equal footing with NATO countries. And yet, here they are.’ ¹⁶ This practice of gradually including partners within what should have been purely NATO decision-making mechanisms influenced the conduct of the Libyan crisis, as Qatar and Jordan were quickly involved on a status similar to the one Australia or Sweden could enjoy in the ISAF context (Schmitt, 2015). This illustrates how practices established in the ISAF context were carried on in another intervention by those already enacting them (OPC and NAC members), thus facilitating the swift inclusion of Qatar and Jordan in the NATO machinery. However, this practice also had an effect on the partners, and is probably related to the close links Sweden and Australia currently enjoy with NATO, Sweden even discussing membership in the light of current Russian behavior.

The way issues were framed and negotiations conducted also had an impact on the final political outcome. One of the most defining features of any policy document agreed upon by the OPC or NAC is the extent to which specific text is ‘recycled’ from other policy documents. Because multilateral diplomacy with more than 40 countries is difficult, it is standard practice to use sections of already agreed-upon policy documents in order to avoid endlessly reopening contentious issues. As a result, diplomats can often be heard explaining: ‘this is already agreed-upon text’ when they are suggesting modifications to a document under negotiation. ‘Already agreed-upon text’ is supposed to be the magic formula avoiding endless and repetitive debates. Because the Afghan War has produced several hundred policy documents over the years, it is not unusual to find a specific formulation that suits a specific state’s preference. The practice is not limited to NATO policy documents either; diplomats routinely refer to texts adopted under the framework of the United Nations, for example. The craft of the diplomat is, therefore, his/her capability to master a large body of policy documents and extract the specific formulation that would favor his/her state’s political preferences while in the meantime presenting this formulation as consensual, since it is supposed to be ‘already agreed-upon language’. Of course, other diplomats are engaged in the same activity, and the OPC meetings are often a battle for the perpetual re-definition of the meaning of past documents to suit the needs of the present.

This practice of ‘copy-pasting’ from other documents has consequences. Documents are re-interpreted to suit a diplomat’s purpose, but the frame for re-interpretation is limited (not every meaning can be attached to any word), which thus constrains the way issues are framed and, therefore, had consequences for the political management of the Afghan campaign. For example, the final communiqué of the 2012 summit includes the following sentence:

We also underscore the importance of our shared understanding with the Government of the Islamic Republic of Afghanistan regarding the full participation of all Afghan women in the reconstruction, political, peace and reconciliation processes in Afghanistan and the need to respect the institutional arrangements protecting their rights (NATO, 2012: 9).

The mention of full participation of all women in the reconstruction process was opposed by the Afghan government, with the support of the US and the UK, neither of which wanted another bone of contention with Kabul while their priority was the transition process. Yet, the Norwegian representative insisted on adhering to the fact that this text was drawn from the United Nations Security Council Resolution 1325 and, as such, was ‘already agreed-upon language’ with considerable political weight. This is how para. 9 made its way into the final communiqué, and is now used by women’s rights organizations in their campaigns. Similarly, para. 6 includes the following sentence: ‘ISAF shifts from focusing primarily on combat increasingly to the provision of training, advice and assistance to the ANSF’ (NATO, 2012: 6). ISAF’s role during the transition to Afghan-led security was discussed at length and no consensus could be found on the degree of assistance it should provide: some countries wanted to keep an active role, including combat missions, while others were more reluctant and wanted to use the transition as an opportunity to withdraw their troops. The compromise was found in recycling the text that read ‘provision of training, advice and assistance to the ANSF’ from a policy document related to the NATO Training Mission-Afghanistan (NTM-A), which was deemed vague enough to accommodate everyone. However, this language suggests a continuous involvement in the ISAF in one way or another and, for example, played a role in France’s decision to keep a number of trainers in Afghanistan despite the withdrawal of the combat forces after President Hollande’s election in 2012; ¹⁷ the recycled language was more committing than initially thought by some nations. Numerous classified documents are negotiated in similar ways, and it is, therefore, critical to take into account the ‘shadow of the past’ when analyzing documents produced by NATO, as the practice of recycling ‘already agreed-upon language’ has political consequences.