No common position! European arms export control in crisis

The European arms export control system is facing a dual crisis: An ongoing crisis of effectiveness has led to a crisis of legitimacy. In many ways, that crisis is a permanent one, where collective efforts to regulate a policy field and implement agreed-upon norms and rules fail to succeed. To explain the different layers of the European arms export control crisis, we draw on global governance research of International Relations and legalisation approaches of international law. The lack of effectiveness of the European Union Common Position on arms exports has become particularly visible in the two review processes where member states failed to agree on meaningful revisions. The Europeanisation of arms production adds additional pressure to the system, as member states differ in their national arms export licensing policy practices. In sum, these developments have led non-governmental organisations and civil society to protest and file successful court cases against arms export practices violating international humanitarian law and human rights law. The article also focuses on possible means of alleviating the crisis of European arms export control.

the main addition was the reference to respect international humanitarian law (IHL) in article 2, criterion 2 (Cops and Duquet 2019, p. 4). Meanwhile, two review processes and a User's Guide manual have further institutionalised the EU Common Position. However, we argue that despite such efforts, EU member states' attempts to harmonise their arms export policies through common global governance efforts remain insufficient and that the entire system of European arms export control remains in deep crisis.
While existent research does not explicitly focus on crisis, authors mostly agree on the principal defectiveness of the Common Standards of European arms export control. However, they differ in their explanations for the lack of communitisation. Stavrianakis (2019Stavrianakis ( , 2016 perceives different standards of arms export control as an attempt of Western states to legitimise liberal forms of militarism. Looking at the Arms Trade Treaty (ATT), which entered into force in 2014, she identifies a careful balancing between states' security interests and human rights concerns. In consequence, she believes that the risk assessment scheme of the EU Common Position should rather be perceived as a "regime of recklessness" when European arms end up fuelling the war in Yemen despite such foreseeable risks (Stavrianakis 2020). Other authors point to the material interests of different stakeholders who remain hesitant to conform to common standards of arms export control. Countries such as France and the United Kingdom (UK), but also the arms industry, would rather pursue their economic and security interests than adhere to the EU Common Position. This results in a general weakening and ambiguity of the global governance regulation and limits any attempts at strengthening common standards (Hansen 2016, Erickson 2011. We agree with many of these authors' findings but want to introduce new insights into the debate by relying on another theoretical framework. We draw on a synergetic definition that not only perceives crisis as an observation object where different stakeholders are socially constructing a crisis, but also sees it as an observation instrument to identify the visible dimensions and mechanisms of crisis (Bösch et al. 2020, p. 5). We agree with Milstein (2015, p. 143) that while crisis is often tangible, we also have to take responsibility for our social world and take action. This element of action is why we write this article, which aims to identify and possibly tackle the deficiencies of European arms export control. We believe that the diverse forms of civil society protest within many EU member states have been successful forms of action. Moreover, the mere existence of the EU Common Position has raised accountability standards. In several EU countries, judges have relied on the eight criteria of the EU Common Position to hold governments accountable for failing to adhere to these rules and revise existing arms export decisions. Civil society also submitted a case to the International Criminal Court, in order to sue defence companies claiming corporate legal responsibility. However, the European arms industry remains a powerful counterpart. European defence cooperation might intensify the crisis, as its general framework is likely to shift the final decision on arms exports to those member states with the least restrictive standards.
In contrast to existent theoretical approaches to explaining European arms export control policy, we draw on global governance research of International Relations (IR) and legalisation approaches of international law (IL). We argue that the current European arms export control system faces a dual crisis, combining a crisis of effectiveness with an ensuing crisis of legitimacy. In the remainder of this article, we look at the various aspects of the European arms export policy crisis, focussing on the EU Common Position and its application. The crisis is in many ways a permanent one where collective efforts to regulate a policy field and implement agreed-upon norms fail to be successful. The following section lays out our theoretical approach. Section 3 concentrates on the lack of effectiveness of the EU Common Position: this lack became particularly visible in the two review processes. Section 4 sheds light on the different layers of protests in various EU member states where nongovernmental organisation (NGO) networks contested the legitimacy of the EU Common Position's application and European arms export control practices through protests or filing court cases. Section 5 elaborates on the Europeanisation of arms production and how this will most likely further add to the crisis. Section 6 focuses on possible means to alleviate the crisis of European arms export control. Section 7 sums up our theoretical and analytical arguments.

Deficiencies of global governance and legalisation: the case of EU arms export policy
While the Common Position on arms exports provides the norms and rules, the established implementation practices are the centrepiece of the attempt to harmonise EU member states' arms export policies. Theories from IR and from IL help to understand the phenomenon. Global governance can be perceived as "the exercise of authority across national borders as well as consented norms and rules beyond the nation state, both of them justified with reference to common goods or transnational problems" (Zürn 2018, p. 3-4). Zürn identifies three layers of global governance: First, a double constituency consisting of states and societal actors; second, issuespecific institutions; and third, the interplay of different spheres of authority. Multilevel global governance contains norms and authority which produce conflicts, contestation and resistance (Zürn 2012). Contrary to norms, global governance research focuses on the sum of stakeholders and institutions that produce regulations in different policy fields and concentrates "on the interplay and the relationship between these institutions and their embedding in a normative order" (Zürn 2018, p. 6). Evaluating legitimacy and effectiveness plays a crucial role in global governance research. Legitimacy has myriad meanings in both IR and IL. However, for the purpose of this article, we rely on a sociological understanding of legitimacy, which concentrates on actors' subjective understanding of legitimacy (Steffek 2007(Steffek , p. 178, 2003. In general, legitimacy concentrates on "an actor's normative belief that a rule or institution ought to be obeyed" (Hurd 2008(Hurd , p. 7, 1999. A crucial aspect of legitimacy is what actors themselves would regard as the subjective requirement for a legitimate order (Peters 2013, p. 11). As a global governance approach concentrates on different stakeholders, such as states, industry, civil society and international institutions, legitimacy in the sense of insights into the rightfulness of negotiated outcomes might vary significantly and easily becomes a matter of contestation.
Research on global governance also concentrates on the effectiveness of international institutions. Wolf (2010, p. 4) identifies three core criteria for such an evaluation: Output, outcomes and impact. While output can be defined as (self-)commitments of states, outcomes are behavioural changes that are based on such commitments. Impact relates to the governance effort and whether it contributes to problem-solving (Wolf 2010, p. 4). Legalisation research in IL adds to this perspective. In their rationalist approach, Abbott and Snidal (2000a, p. 401) differentiate institutions in terms of three core characteristics: Obligation, precision and delegation. They define obligation as a "set of commitments" by which states are legally bound along the dimension of precision. Precision addresses how concise or how broad the rules and norms actually are. A third dimension, delegation, concentrates on the role of third parties authorised to implement, interpret and apply the rules, resolve disputes, generate new rules and supervise compliance.
With their interactionist approach of IL, Toope (2010, p. 100, 2013) perceive the compliance pull of institutions as depending on the congruence of shared understandings, commonly accepted criteria of legality and, decisively, the interactive practices of legality. Kratochwil (2014Kratochwil ( , p. 54/67, 1989 shares this social constructivist view of international law-making as a "world of practice" where actors change and reproduce the normative structures when they "share meanings, communicate intentions, criticize claims, and justify choices". He underlines the relevance of the actors' practices and the "concomitant process of interstitial lawmaking". Legal practitioners/lawyers such as Lowe (2000, p. 213) also stress the role of practices of international law-making when concentrating on the fine-tuning of norms through regular meetings and review processes. These forms of interstitial law-making occur at the micro-level of institutionalising new agreements by elaborating on the interstices of norms. We argue that the rational and social constructivist legalisation approaches of international law add additional perspectives to global governance research. Together, these approaches help to illuminate the dual crisis of legitimacy and effectiveness of EU arms export control.

A crisis of effectiveness: failing to harmonise EU arms export policy
Many scholars and non-governmental organisations view the EU Common Position as an ineffective instrument to regulate arms export policy and prevent European weapons from violating international humanitarian law and human rights law in many recipient countries (Cops and Duquet 2019;Tricolt O' Farrell and Isbister 2018;Amnesty 2016;Lustgarten 2013). While we agree on the evaluation of the instrument, we also want to shed some new light on the issue by drawing on the global governance and legalisation approach. We argue that European arms exports during the Arab Spring and the wars in Libya and Yemen generated disunity between EU member states in their arms export policy. While this did not result in a more fruitful outcome of the second review of the EU Common Position in 2019, it provided leverage for NGOs to scandalise the failure to adhere to IHL and international human rights law (IHRL) and prompted them to initiate legal proceedings.
The scandalous arms export revelations in the aftermath of the 1990 invasion of Kuwait by Saddam Hussein and the First Gulf War led European states to make concerted efforts in coordinating arms export policy (Joachim and Dembinski 2011;Stavrianakis 2010;Broszka 1995). The outcome of this consultation process was the Declaration on Non-proliferation and Arms Exports containing some very basic arms export criteria, which left much room for ambiguous interpretation (Hansen 2016, p. 203). 2 In June 1998, the EU member states agreed on the politically binding Code of Conduct on Arms Export. While NGOs lobbied for such an agreement, the UK and France also perceived the Code as a useful instrument for levelling the playing field and curbing US efforts to control the arms export policies of its allies (Joachim and Dembinski 2011). As before, agreement on the Code was only reached on the lowest common denominator to get France in particular on board, while countries such as Germany, Ireland, the Netherlands and the Nordic countries favoured stronger language, particularly on the criteria. 3 As a result of continuous ambiguity and non-compliance, the EU countries finally agreed on the legally binding EU Common Position 2008/944/CFSP in 2008 Even though the EU Common Position is formally legally binding for member states, it is also part of the EU Common Security and Defence Policy (CSDP), which is otherwise under national jurisdiction. And Article 346 (1)(b) of the Treaty on the Functioning of the European Union grants member states the right to act independently with regard to arms production and trade. As long as there is only a very rudimentary EU foreign and security policy, arms export policy still remains a matter of national sovereignty (Hansen 2016). No cut-through law exists which allows any EU institution to interfere with member states' national rules and provisions on arms exports. All EU countries have different systems of arms export control in which the Common Position has to be included as part of their legal bindingness. In spite of this, the obligations resulting from the EU Common Positions remain very limited depending on the respective member states' national laws. Defence is still considered a matter of national sovereignty, implying that the Common Position is legally binding but not enforced at the EU institution level, leaving it without a sanction instrument.
The initial purpose of the Code of Conduct and later of the Common Position is the institutionalisation of a more restrictive and convergent European arms export policy. At the core of the EU Common Position is a set of eight common criteria that member states committed themselves to applying in their assessment of arms export licenses. 5 The criteria encompass, for example, the respect for IHL and human rights of the recipient country or the preservation of regional peace and stability.
2 Other criteria such as the "development" criterion did not meet the acceptance of France and the UK, who feared a negative impact on arms exports to MENA countries. In contrast, Germany and the Netherlands strongly favoured such a criterion. See Hansen (2016, p. 203-204). 3 Bauer (2003, p. 135) stresses that the criteria left ample room for interpretation and that the interpretation of the criteria remained at the national domain. 4 The EU member states needed three and a half years to agree on the Common Position. France in particular lobbied hard for lifting an arms embargo on China. The deteriorating domestic politics situation in China at this time and intense NGO lobbying led France to seek a compromise (Depauw 2010). 5 See https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32008E0944&from=EN.
The eight criteria of the EU Common Position could have a considerable impact on the outcome of the global governance of arms exports and could change existing practices if implemented rigorously. But the opposite is the case: The interpretation of the criteria varies greatly among EU member states, frequently resulting in controversial debates and special national paths. In the words of Abbott and Snidal (2000a, p. 401), the eight criteria of the EU Common Position lack precision. In more than 30 years of efforts to institutionalise a common European arms export policy, there have been several attempts to strengthen the EU Common Position. Legal practitioners/lawyers such as Lowe (2000, p. 213) point to the importance of law-making at the interstices of norms or global governance agreements in order to make them policy-relevant. However, when it comes to the practices of fine-tuning, the norms of the EU Common Position indicate a lack of precision.
In the case of the Common Position, the interstitial law-making practices encompass different measures. Already in the early 1990s, the European Council initiated the Working Group on Conventional Arms Exports (COARM), where member states could explore the opportunities to harmonise export controls by discussing best practices of national legislation. COARM has remained the monthly meeting space for officials of all EU member states to discuss arms export policies and to aim at the further convergence of these policies. Part of the European External Action Service (EEAS) is a small COARM secretariat that provides organisational structures for the meetings and has developed an online information-sharing system. Since 2010, a permanent chair from the EEAS has replaced the rotating chair of member states' presidency to better guarantee consistency and expertise (Bromley 2012, p. 5). However, the role of the EEAS regarding the Abbott/Snidal perspective of delegation remains very limited. While it does play a role in facilitating the meetings, it has no authority to implement, interpret or apply the rules, neither does it generate new rules or supervise compliance.
Member states are also asked to provide the others with a confidential annual report on their arms exports. As a consequence of pressure from civil society, the annual reports were made public in 1998, and their number has increased considerably (Bromley 2011, p. 43). With the Common Position of 2008, a European consolidated annual report was added, which is based on these annual national reports. The working at the interstices of the eight criteria, i.e. their operationalisation, has also led to the development of the User's Guide, which provides guidance for national authorities on how to interpret and apply the eight criteria. Since 2000, a Common EU Military List, closely related to the Munitions List of the Wassenaar Agreement, has been agreed. Member states are required to control the exports of the goods on the list. Last but not least, a denial notification process has been put in place as part of the procedures around the Common Position where member states could inform each other about their denials. This procedure aims to relinquish "undercutting", implying that a denial should not lead to an export licence being granted by other EU member states (Cops and Duquet 2019, p. 2-3).
Since the adoption of the 2008 Common Position, Article 15 of which stated that the instrument should be reviewed three years after its adoption, there have been two review processes. They both aimed to assess the existing framework's adequacy and sought to elaborate measures for improvement. To use the terminology of Snidal (2013, 2010b), the reviews also intended to make the Common Position more precise. The first review process was initiated in 2011 and took three years. It coincided with the Arab Spring, which generated conflicts among EU member states on conventional arms exports to Middle East and North African (MENA) countries. Two concrete initiatives followed from the first review: The User's Guide was adapted and further elaborated to help member states in their effort to operationalise the existing criteria. An IT system was to support the sharing of information of export license denials. However, proposals to elaborate on the eight criteria and insert substantive changes such as including democratic governance and adding surveillance technologies to the military control list did not achieve consensus among member states. Both issues had been controversially discussed in the light of the Arab Spring movements.
Similar things can be said regarding the second review process, which took 18 months until September 2019. Again, only minor improvements to the existing Common Position could be achieved. Similar to the first review process, public debates on different EU member states' arms export policies concerning MENA states involved in conflicts accompanied the review process. The text of the Common Position was updated as new instruments such as the Arms Trade Treaty had to be considered. An additional chapter on transparency requirements was added to the User's Guide. The establishment of a searchable database was decided, and a deadline for submitting the annual reports was agreed upon. In sum, both review processes could have been milestones in improving the effort to harmonise the EU member states' arms export policies but failed to match the expectations of NGOs and many EU parliamentarians. The lack of precision of the term and concept of "clear risk" remains a central problem. The User's Guide does provide some hints about information sources on the various risks, but at the end of the day, a risk assessment is completely in the hands of the national governments, and there is no accepted European benchmark against which it could be assessed (Grebe 2013). Stavrianakis (2020, p. 233) points to the flaws in the principle of the risk assessment scheme, which, in her view, facilitates arms trade and serves as "a mode of domination". While we sympathise with this argument in principle, we also want to point out that a risk assessment in the Yemen case resulted in a denial strategy, at least by some countries such as Germany, the Netherlands and Sweden. Moreover, officials stress that access to licensing decisions of other EU member states helps to justify their own decisions vis-à-vis their government, parliamentarians and the informed public (Bromley 2011, p. 45).
In sum, while the attempts at harmonising European arms export control have led to some output in terms of a general self-commitment of EU member states to regulate their arms export policies along commonly agreed standards, the outcomes in terms of sustainable behavioural changes have been limited. In general, there is still a gaping chasm between the more export-friendly countries and more restrictive states. 6 Therefore, harmonisation remains an urgent necessity as there is a growing discrepancy between the still nationalised arms export policy and an increasing Europeanisation of the arms industry (which we will discuss below in Section 5). Moreover, the lack of effectiveness of the EU Common Position became visible once more when EU member states continued to sell arms to MENA states who actively participated in the Yemen war. When civil society openly challenged the legitimacy of the EU member states' arms export practices to third party states waging war, this added another layer to the existing crisis of the EU Common Position. . The Yemen war might be a "hard" case for global governance, given the economic and security interest involved, but for NGOs lobbying for changes in EU arms control policy, it has also been a crucial case with some noteworthy developments. Numerous studies on the relevance of NGO activities point to the fact that the institutionalisation of EU arms export policy through the Code of Conduct and the Common Position has increased NGOs' access to the decision-making process (Dembinski 2009;Greenwood 2009).
While direct influence appears rare, the regulatory framework served as a moral standard for rhetorical entrapment and holding governments accountable for wrong licensing decisions. In the case of the Yemen war, civil society repeatedly contested the legitimacy of the arms export control system as practised by the EU and its member states. While legitimacy might indeed be subjective, some NGOs managed to objectify the EU Common Position by initiating legal proceedings in some member states. Criterion 2 (IHL violations) of the EU Common Position in particular became an objective reference ground for an increasing number of court decisions in member states, implying yet another path of rhetorical entrapment and a form of accountability for EU member states (Joachim andDembinski 2011, p. 1160;Stavrianakis 2012). 7 At first, the divergence in national arms export licensing and in interpreting the eight criteria of the EU Common Position became visible when a few European states decided to reduce or refrain from arms exports to Saudi Arabia, such as the Netherlands, Denmark, Norway and Germany. In the case of Germany, this policy Uturn came only after the assassination of the Saudi journalist Jamal Khashoggi in October 2018 at the Saudi embassy in Istanbul, Turkey. Media and NGO reports revealed the magnitude of European arms being used in the Yemen war and the IHL violations committed with them. NGOs issued legal complaints in different EU countries such as France, Italy and the UK, challenging their governments' decision to export arms and military equipment to Saudi Arabia that may be used to commit serious violations against IHL (Ferro 2019). In Italy and the UK, the claimants argued that the government had failed to comply with national and international law regulations citing the EU Common Position and the ATT (Maletta 2019).
The discrepancy between the eight criteria of the EU Common Position and EU member states' arms export practice already became visible during the Arab Spring, but the gap has considerably widened since the Yemen war starting in 2015. Several MENA countries are regular recipients of European arms, a practice which has continued despite IHL violations by the warring parties recognised by the UN Panel of Experts (UN Security Council 2020). Article 1, common to all Geneva Conventions, obliges all states to "respect and ensure respect" for IHL. This is considered a common law provision for states to ensure that arms exports will not be used for IHL violations. The EU Common Position requires EU member states to deny an export licence if there is a "clear risk" that the arms and goods "might be used" to commit serious violations of IHL. The recipient's attitude towards relevant IHL principles should also be taken into account (Maletta 2019).
In several countries, peace and human rights groups have initiated lawsuits against their governments based on the licence practices towards recipients involved in the Yemen war, such as Saudi Arabia and the United Arab Emirates (UAE) (Azarova et al. 2021). These collective efforts of NGOs in several EU countries are a visible sign of how and to what extent civil society relies on the criteria of the Common Position to hold governments accountable. Yet, all EU countries have different national arms export control systems, which include the Common Position. Whether NGOs can sue their governments or not is dependent on national prerogatives, such as whether an association has the right to legal action or not. The following three empirical cases highlight the importance of civil society taking critical arms exports decisions to a national court. In a fourth case, corporate responsibility of the arms industry for their arms sales in respect to the Yemen conflict is under investigation at the International Criminal Court (ICC). 7 The role and influence of NGOs in EU arms export policy is very differently assessed. While some authors (Joachim and Dembinski 2011;Greenwood 2009;Dembinski 2009;Bauer and Bromley 2004) perceive the Europeanisation of arms export control as leverage for NGOs, others (Stavrianakis 2020(Stavrianakis , 2012 evaluate NGO "technocratic" activism more critically. From our multilevel governance perspective, NGO activism in regard to the Yemen war facilitated the legal proceedings. In the UK, the Campaign Against Arms Trade (CAAT) successfully sued the government for breaching EU export laws with arms exports to Saudi Arabia (Beckerle and Minogue 2019; Isbister 2019). CAAT formally launched its judicial review in March 2016. More than three years later, the Court of Appeal ruled in June 2019 that the British government had acted unlawfully in granting arms export licenses to Saudi Arabia and not undertaking a proper risk assessment as to whether violations of IHL had taken place. The government was ordered to review all licenses, to conduct a proper risk assessment and stop issuing new licenses for arms exports to Saudi Arabia (CAAT 2020). Although the UK government refused to accept the judgment and appealed to the Supreme Court, the Secretary of State for International Trade, Liz Truss, issued a written statement to Parliament on 7 July 2020 explaining that the government had completed the review ordered by the Court of Appeal (Truss 2020). The government had concluded that any violations of international law were "isolated incidents" and announced its intention to resume arms exports to Saudi Arabia and its coalition partners. CAAT initiated new legal proceedings arguing that the "small numbers" argument was irrational and that even "isolated incidents" could involve a clear risk of future IHL violations. The recent court case is still pending at the time of writing (CAAT 2021).
In Belgium, an NGO network achieved two successes when it sued the regional government of Wallonia for selling arms to Saudi Arabia. 8 In March 2020, the Council of State ordered the immediate suspension of 27 export licenses issued by the Walloon government. The Council of State ruled that the licenses were not sufficiently supported by a government declaration regarding the respect for human rights by the Saudi Arabian government, which is required by a 2012 regional law. Moreover, the Council of State stated that there is a "clear risk" that the military equipment will be used "in serious breach of international humanitarian law in Yemen" (Hope 2020). In August, the Council of State again blocked the approval of one out of six arms export licenses to Saudi Arabia. It ruled to overturn the approval of an arms export licence between the small arms manufacturer FN Herstal and the Saudi National Guard, pointing to their involvement in the Yemen war. While the court ruling can be perceived as a success in holding governments accountable to IHL and existing global government agreements, the Belgian government has still not enacted an export moratorium towards Saudi Arabia. The NGO network heavily criticised the Walloon government's decision: "The resumption of arms exports towards a country which is suspected of committing war crimes in Yemen, therefore done in total violation of international humanitarian law (...) is all the more shocking since it comes barely two months after the Council of State already suspended the licenses authorising these exports" (The Brussels Times 2020).
The court case in Italy seems to be unique. Similar to the other European cases, an NGO coalition submitted a criminal complaint to the Public Prosecutor's Office in Rome in April 2018. 9 Three years earlier, a family of six, including a pregnant woman and four young children, had been killed in an airstrike by the Saudi-led military coalition in northwestern Yemen. Bomb remnants were discovered at the site of the attack, including a suspension lug manufactured by RWM Italia, a subsidiary of the German arms manufacturer Rheinmetall. After almost two years of investigations, the Italian Public Prosecutor's Office dismissed the case. The NGOs decided to appeal the Prosecutor's decision. They argued that the investigations undertaken were incomplete and that Italy's potential responsibility for IHL crimes had not been properly investigated. The NGOs pointed to the fact that Italy had approved arms exports to Saudi Arabia in November 2017 despite the fact that the warring country had gravely violated IHL. Accordingly, approving arms export licenses in such a case runs against Italian law, but also against the criteria of the EU Common Position and the Arms Trade Treaty (ECCHR 2019b). In December 2020, the Foreign Committee of the Italian Parliament extended a one-year-old moratorium denying licenses for aerial bombs and missiles used by Saudi Arabia and the UAE in the Yemen war.
Finally, a coalition of European and Yemenite NGOs submitted a case to the International Criminal Court's Prosecutor's Office, suing European defence companies for their involvement in war crimes in Yemen as part of their corporate legal responsibility (ECCHR 2019a). The case description seeks to challenge the European economic and political actors' role in the Yemen war and also asks the ICC for its judgment on IHL accountability. To summarise: European NGOs manage to contest the effectiveness of the EU Common Position and the legitimacy of its application by EU member states. In light of crisis evaluation, the four court cases provide interesting examples of multi-level (global) governance where states find themselves rhetorically entrapped and held accountable with the help of the EU Common Position as an important reference. However, the UK case also demonstrates some of the limitations of the court rulings. Holding the defence industry accountable for its continuous arms supply to warring parties might be another way of raising public legitimacy of the EU Common Position.

European arms cooperation as an accelerant of the crisis
The plans for more and closer cooperation of EU member states in armaments has the potential to add an additional layer to the crisis, as it will lead to more constellations where member states with different approaches to arms export control will have to harmonise their practices of applying the rules of the Common Position. This pressure could lead to interstitial law-making, which would improve the effectiveness of European arms export control. However, this improvement would be only superficial, as it would come at the cost of lower overall control standards 9 The criminal complaint was submitted by the European Center for Constitutional and Human Rights (ECCHR), together with Rete Italiana per il Disarmo and the Yemen-based partner Mwatana for Human Rights. and increasing arms exports to problematic third countries. In turn, this would very likely increase the already existent crisis of legitimacy.
From an economic perspective, European arms cooperation makes sense. With rising costs for developing new, sophisticated military technologies and comparatively slower-growing defence budgets, it is not surprising that governments seek to lower the costs for arms procurement via joint efforts. A centrepiece of these efforts in the framework of the EU is the so-called Permanent Structured Cooperation on security and defence (PESCO), formally established by the Council of the European Union in December 2017. The central aim of PESCO, in which 25 member states are participating, is the joint development of defence capabilities. By November 2019, governments had agreed on 47 projects in that framework. Italy, Greece and Slovakia are jointly developing and building a prototype armoured infantry fighting vehicle, while France, Germany and Spain joined forces to upgrade the capabilities of the attack helicopter Tiger, to give only two examples here. 10 To provide additional financial incentives for arms cooperation, EU member states established the European Defence Fund (EDF) in 2017. Its funds are meant to finance research and development of military technologies, and they could be used to co-fund PESCO projects in this regard. While the fund consisted of 590 million euro in its first phase from 2017 to 2020 (of which 575 million were spent), over seven billion euro will be allocated from 2021 to 2027.
Two large cooperation projects that are not (yet) part of PESCO but which have produced much interest in recent years are the planned Future Combat Air System (FCAS) and the planned future main battle tank, referred to as Main Ground Combat System (MGCS). FCAS is a cooperation project between France, Germany and Spain, coordinated by Dassault Aviation, Airbus and Indra Sistemas, that aims to develop a sixth-generation fighter aircraft accompanied by a remote carrier vehicle. Entry into service is planned around 2040. For the MGCS, Germany and France are co-operating to build a new main battle tank that could replace the German Leopard 2 and the French Leclerc in around 2035. Certainly, much of the success of European arms cooperation will depend upon the collaboration between France and Germany-and their arms industries. The organisation of the defence sector in France and Germany is quite different: While in the case of France, the majority of the defence industry is state-owned, and arms exports are understood as part of French foreign and security policy, the defence industry in Germany is private (Béraud-Sudreau 2018). In the broader framework of the Treaty of Aachen (Treaty on Franco-German Cooperation and Integration)-signed by President Emmanuel Macron and Federal Chancellor Angela Merkel in January 2019-the French and the German government signed an agreement on defence export controls in November 2019. In particular, they agreed on a so-called de minimis clause, which means that one government does not need to ask the other for consent when it wants to license an arms export if the share of parts and components for the respective weapons system that stems from the other country is below a certain level-in this case, a share of 20 per cent (Ministère de l'Europe et des Affaires Etrangères 2019). This could mean, for example, that FCAS, if the final assembly of the weapons system took place in France (which is a plausible scenario), could be exported under a French export license without the consent of the German government if the value of the parts and components that come from the German arms industry is below this 20 per cent threshold. If this approach forms a precedent and an increasing European arms cooperation leads to a multitude of bi-or trilateral agreements with similar de minimis regulations, this would reduce the pressure for harmonising export decisions (as countries with a contribution below the de minimis threshold would relinquish their decision-making power). However, such a process of interstitial law-making would be a disaster from an arms control export perspective, as it would very likely lead to an increase in arms exports to problematic third countries because the final assembly of arms and thereby export decisions could be transferred to the least restrictive states.
However, if combined with meaningful reforms of the European arms export control system, arms cooperation could even support a stricter and more coherent arms export policy. It could lead to a reduction in different types of weapons systems in Europe, which would reduce the costs for development and production and allow lot sizes of the produced arms that would reduce the pressure to export arms to third countries to reduce their unit costs. But this would only work if reducing arms exports to (problematic) third countries were the explicit goal behind a strategy of European arms cooperation and if rules and regulations were in place to ensure this. Otherwise, the European arms industry would use the additional competitiveness gained by the cost reduction to increase its shares in the global arms market (Mutschler 2019). If no political decision is taken to change the current arms export control system, this will be the most likely scenario. And this scenario would include a trend, as outlined above, according to which the final assembly of weapons systems is shifted to those EU member states that have the weakest arms export controls. In the worst case, this leads to a race to the bottom, where states try to undercut each other with regard to their arms export controls.
Certainly, the whole idea of closer and better European arms cooperation is not new. Closer cooperation in the defence sector was discussed in the 1980s and 1990s, and it was one of the main goals of the European Defence Agency (EDA) when it was established in 2004. National arms procurement policies and the wish of governments to protect their national arms industry have led to rather poor results in this field, and the EU has not become a relevant player in or framework for arms cooperation (Mölling 2015). This, however, could be about to change with the above-mentioned initiatives of PESCO, EDF and closer bilateral cooperation between France and Germany. So, while the history of European arms cooperation teaches us to be sceptical with regard to the future outcome of these initiatives, there is a real potential that they intensify the observed crisis of European arms export control because the need for harmonisation and arms export control at the European level increases, while at the same time, there is no improvement concerning the level of obligation, precision or delegation of the rules of the Common Position, as illustrated above. The cases of the Eurofighter combat aircraft illustrate that defence cooperation among European states has already started to undermine arms export control. When Germany enacted export restrictions against Saudi Arabia in 2018 (see above), France and the UK, whose defence industry is in the lead regarding the K production of Eurofighter aircraft for Saudi Arabia, successfully pressured the German government into easing the restrictions on the export of parts and components that were to be integrated into the Eurofighters (and other jointly produced arms) for Saudi Arabia (Reuters 2019).

European integration of arms export control as a way out of the crisis?
A classic solution to avoid a scenario where EU member states fail to coherently implement rather abstract norms and rules, like those of the EU Common Position, and risk undercutting each other would be the full-fledged communitisation of arms exports control at the EU level. This would mean that the competencies for taking decisions on arms exports would be shifted from the national to the supranational level. This could be done, for example, by dissolving Article 346 (1)(b) of the Treaty on the functioning of the European Union that grants member states general rights to act independently with regard to arms production and trade. The Common Position could then be transformed into an EU directive or regulation (similar to the EU Dual-Use Regulation), and arms export controls would be integrated into the competence of the Commission, which could either establish an own European licensing authority-a kind of EU arms export control agency-or leave this task to the national authorities but sanction violations through infringement proceedings. While such a communitisation, with the delegation of competencies to the Commission, could potentially improve the coherence of arms export decisions, there are significant downsides to such an approach. First, it is extremely unlikely that member states would agree to transfer their competencies in this field to the Commission. And second, even if they did, this might lead to more coherence but most likely at the detriment of a strict application of the criteria, as the Commission perceives arms export policy rather from a competitive perspective than a primary instrument of arms control (Kehne 2019, p. 3). However, several improvements below the level of full-scale delegation to the EU level can be made. The criteria set out in the Common Position would need to be improved by including the factors 'good governance' and 'democracy' (Bromley 2012, p. 15), but most importantly, by specifying the concept of "clear risk". For each of the eight criteria, the Common Position should include references to potential evidence for the existence of the identified risks (the remarks in the User's Guide to the Common Position are too imprecise and non-committal). On this basis, national representatives at COARM should come to a joint risk assessment before taking national decisions on export licences. Of course, this would mean more and better information-sharing among states. To support such a joint risk assessment at the EU level, the European External Action Service (EEAS) could play a role in collecting and disseminating information on specific end users and the human rights situation in potentially problematic countries (Cops et al. 2017, p. 191-194).
However, as our look at the poor result of both reviews of the Common Position illustrates, not much can be expected from member states' governments regarding interstitial law-making and procedural innovation. It seems that there is no interest in changing the status quo. However, while governments might be the most powerful actors in this policy field, they are not the only stakeholders. As we have shown, civil society organisations were at least partly successful in contesting the legitimacy of arms export decisions by initiating lawsuits against governments based on the regulatory framework of the EU Common Position. In our view, these legal contestations are also interstitial law-making practices, as they contribute to the finetuning of norms and the clarification of ambiguous concepts like "clear risk".
From this perspective, the best hope for overcoming the EU arms export control crisis is civil society-driven interstitial law-making. However, in all likelihood, NGOs will need to build alliances with stakeholders from the EU institutions if they want to institutionalise their norm interpretation. The most promising institution here-also as a political corrective towards member states (von Boemcken and Grebe 2014, p. 152)-seems to be the European Parliament (EP). The EP itself has repeatedly called for more competencies in this policy field. In its resolution of 17 September 2020 on the implementation of the Common Position, it calls for "meaningful parliamentary oversight" and announces that it sees a need for an annual European Parliament report in response to the annual report of COARM "in order to ensure a minimum of parliamentary scrutiny" (European Parliament 2020). The fact that with the EDF, money from the Common Budget of the EU is spent on research and the development of arms, lends additional legitimacy to such claims. At least, a regular exchange between COARM and the EP on arms export policy should be established. In October 2021, members of the Greens/EFA group in the EP presented a draft EU regulation for the control of arms exports which provides for a stronger role of the Commission and the EP in this field. While the EP has no formal powers to enforce any concessions from member states in the field of arms exports, it does have the power to set up a subcommittee on arms export controls (to the Committee on Foreign Affairs). This subcommittee would-together with civil society organisations-develop the Parliament's interpretation of the criteria of the Common Position and produce a detailed annual report that scrutinises member state's arms exports and evaluates them in accordance with the criteria of the Common Position. 11 From a social constructivist view of international law-making, such practices by NGOs and the EP can be considered part of interstitial law-making, as they change (or reproduce) the normative structures by criticising claims, sharing meanings, communicating intentions and justifying choices.

Conclusion
The system of European arms export control is facing a dual crisis-the ongoing crisis of effectiveness has also evolved into a crisis of legitimacy. We agree with findings of crisis research that crisis definitions should contain both elements, some visible and objective dimensions, but also socially constructed perceptions at-tributed by different stakeholders. To fully understand the complexities of European arms export policy, we choose a global governance approach that concentrates on stakeholders such as states, societies and issue-specific institutions. The legalisation perspective adds further interesting insights into the relevance of practices and interstitial law-making. While other authors have also frequently criticised the lack of effectiveness of the EU Common Position, the legitimacy perspective adds some new insights into the nature of the crisis and offers some possible ways out.
When looking from a global governance perspective and concentrating on states, the effective implementation of the EU Common Position is infringed by states' policy stressing national sovereignty rights. Weak compliance procedures-a lack of obligation, precision and delegation-have furthered the incoherent implementation of the common rules on arms exports and have allowed arms exports to countries actively engaging in wars and committing grave violations of international humanitarian law. Moreover, EU member states remain at odds regarding the application of criteria, as the case of the Yemen war and the different European arms exports decisions regarding the warring parties demonstrated.
With a rudimentary EU foreign and security policy, the institutionalisation of the EU Common Position remains dependent on national laws and decision-making. Instead of a common arms export policy, licensing remains at the discretion of its member states-an institutionalised mechanism of compliance is lacking. As concerns global governance research, the output-the normative self-commitment of the actors-of the Common Position can be described as meaningful. The outcome, in turn, seems to be far more limited when it comes to behavioural changes related to arms export control practices. While many EU member states have adopted the criteria and put them into national law, the lack of obligation becomes visible by the fact that violations of the eight criteria cannot be sanctioned at the supranational level. Also, common standards and practices of "risk assessment" regarding specific countries are missing.
However, the social constructivist perspective on law-making focussing on the interstitial fine-tuning of norms and the role of practised norm-contestation, which we applied here, directs our look to potential ways to overcome the crisis. We have shown that civil society and NGOs as another crucial stakeholder, in particular, have been quite effective in holding governments accountable for not complying with existing arms export control regulations. Also, the networking among European NGOs has increased, facilitated by the influence of the global instrument, the Arms Trade Treaty, which leads to increased cooperation and information-sharing. Taking the route via legal proceedings might provide another way to hold governments accountable and advance interstitial law-making. One promising strategy to mitigate and overcome the European arms export control crisis, in the long run, seems to be to continue and even intensify this pressure on governments, combined with efforts to strengthen the European Parliament as a counterpart to the Council and the member states. The Parliament resolution of 17 September 2020 indicates that the EP itself is willing to claim a more active role, while the peace research and arms export control community could help support this development with independent analyses of problematic export practices and recommendations on how to overcome them.
Last but not least, the arms industry remains an ambiguous but potentially "promising" stakeholder. Intensified arms cooperation will most likely come at the cost of increasing arms exports to problematic third countries. However, the avenue of suing the defence industry for corporate legal responsibility could lead to some forms of governance standards in which the European arms industry might have a vested interest. We do not claim that all those activities will (in the long run) lead to a change of preferences, in particular by those member states with the lowest control standards. However, we offer an analytical perspective that draws attention to the potential of gradual, incremental change by interstitial law-making involving stakeholders beyond the intergovernmental realm. This could pave the way for overcoming both the crisis of effectiveness and the crisis of legitimacy.
Funding Open Access funding enabled and organized by Projekt DEAL.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4. 0/.