Sibylle Bauer
Search for other papers by Sibylle Bauer in
Current site
Google Scholar
Eric Remacle
Search for other papers by Eric Remacle in
Current site
Google Scholar
Theory and practice of multi-level foreign policy
The European union’s policy in the field of arms export controls

European foreign policy functions as a 'system' of multi-level policies structured in four levels (conflicting, cooperative and institutionalised intergovernmentalism, and supranationalism), each of which refers indirectly to one of the main integration theories. This chapter discusses the following three dimensions of European foreign policy-making with reference to the case of arms export controls: convergence or vertical coherence, consistency or horizontal coherence and variable geometry. The various levels of foreign policy-making inside the European Union (EU) allow for different speeds and degrees of integration in different policy areas of the three pillars as well as within the pillars, and also in various aspects of the same policy area. Each area of foreign policy decision-making in the EU seems to be inspired by different ideological approaches to integration favoured by the different foreign policy actors.

Theories of integration usually provide monocausal explanations of integrative processes. They cannot therefore be considered as general theories but rather as ideological models which reflect the state of the European construction, state the preferences and values of the actors in EU policy-making, and contribute to the inevitable compromises between them (Caporaso and Keeler 1995). From a constructivist perspective (Adler 1997b), one could say that integration theories are social and ideological constructs that contribute to shaping the reality of integration rather than to explaining it.

European foreign policy is a case in point. Each area of foreign policy decision-making in the EU seems to be inspired by different ideological approaches to integration favoured by the different foreign policy actors. Accordingly, European foreign policy functions as a ‘system’ (Monar 1997) of multi-level policies structured in four levels, each of which refers indirectly to one of the main integration theories (Remacle 1997). This approach will be applied in this chapter to the field of arms export policies.

Conflicting intergovernmentalism results from a perception of national policies as being the core of, and key to, explaining European policy-making. It is mainly inspired by the realist paradigm. Fundamental to this approach are the traditional application of the concept of sovereignty and the inter-state balance of power (Hill 1983; Pijpers 1991). The management of the Yugoslav crisis by EU countries best illustrates this aspect of EU foreign policy, as do national arms exports policies.

Cooperative intergovernmentalism corresponds to theories of functionalism (Mitrany 1966), of adaptation (Rosenau 1970; Petersen 1998) and of maximisation of the national interest and/or convergence of preferences (Milward 1992; Moravcsik 1993; Pfetsch 1994). Illustrations of this type of European foreign policy are reflected in various forms of ad hoc bi- and multilateral cooperation between EU member states in the field of foreign policy. These include: reference to common interests in Article 11 of the TEU, the policy process of European political cooperation 1970–93, EU consultations in the United Nations,1 and participation in export control regimes such as the ‘Wassenaar arrangement’.

Institutionalised regime-type intergovernmentalism is based on assumptions of regime and interdependence theories (Keohane and Nye 1977; Krasner 1981). Their application to the institutional system of the EU leads to the development of a security community among EU states (Deutsch 1954) and to a new type of polity called a ‘confederal system’ (Wallace 1994), a ‘consociational confederation’ (Taylor 1993; Chryssochoou 1996), an ‘intergovernmental federation’ (Quermonne 1998) or a ‘federation of states’ (Telò and Magnette 1996). The institutionalisation of the Common Foreign and Security Policy (CFSP) and its associated instruments established by the Maastricht and Amsterdam Treaties is an illustration of this regime-building approach. Examples of these instruments include the EU Code of Conduct on Arms Exports of 1998 and the EU Joint Action on Small Arms of 1998, both of which were agreed within the framework of the CFSP.

Supranationalism requires the existence of a foreign policy dimension based on first pillar instruments, which has also been called an integrated structural foreign policy (Keukeleire 1998). It relates to theories developed by federalists (Sidjanski 1992), neo-functionalists (Haas 1958; Lindberg 1963; Schmitter 1992) or neo-institutionalists (Regelsberger and Wessels 1996; Wessels 1997), all of whom predict the transformation of the Community into an international and global player which influences its partners both by its existence and its actions (Ginsberg 1999). With regard to the field of arms exports, while there is no first pillar EU competence, a grey zone of dual-use items2 exists to which Community legislation applies (European Council 1994a, b and 2000a).

The development of the EU’s foreign policy can be interpreted either as a combination or as a chronological/teleological succession of these four levels. Whichever interpretation is chosen, the following three concepts are key to understanding the mechanisms of European foreign policy: first, convergence (or vertical coherence) between national policies and EU policy; second, consistency (or horizontal coherence) between policies carried out in the different pillars of the EU institutional system; and third, variable geometry in policy-making and institutions. These three dimensions of European foreign policy-making will be discussed hereafter with reference to the case of arms export controls.


EPC in the 1970s and 1980s had already allowed the EC member states to build channels of communication and methods of socialisation among their diplomats in order to define common objectives and coordinate the first joint diplomatic actions (de Schoutheete de Tervarent 1986; Pijpers 1991; Nuttall 1997). The CFSP continued this process (Willaert and Marques-Ruiz 1995; Maury 1996; Holland 1997) and has given rise to parallel tendencies towards a Europeanisation of national policies as well as towards a maximisation of national positions through European channels (Hill 1983; Peterson and Sjursen 1998). Studies on the Spanish and Portuguese cases (Algieri and Regelsberger 1996), on the Nordic dimension (Jopp and Warjovaara 1998) and on the Europeanisation of national structures and policies (Hanf and Soetendorp 1998; Tonra 2000) provide solid empirical evidence of this double trend. The question of arms export policies can be added to these case studies.

Arms export policy falls simultaneously into the fields of economic, competition, foreign, security and defence policy with their different degrees of community and member state competence. At the same time, this policy area has traditionally been considered as inextricably linked to national sovereignty. This led to the inclusion of an ‘opt-out clause’ in the Treaty of Rome (Article 296 of the Treaty establishing the European Community, former Article 223), which gives EU member states the right to exclude arms exports from EU competence.3 Member states have been reluctant to open up a sector that has traditionally been protected from outside competition, highly subsidised, and subject to a high degree of state intervention and control. Article 133 (former Article 113) of the EC Treaty states that the ‘common commercial policy shall be based on . . . the achievements of uniformity in measures of trade liberalisation, export policy and measures to protect trade such as those to be taken in the event of dumping or subsidies’. Article 296 (Bauer and Küchenmeister 1996; Emiliou 1996) allows member states to invoke their security interests to exempt armaments from these provisions.

According to EU member states’ interpretation of Article 296, matters related to the production of and trade in armaments have therefore been dealt with within the framework of national jurisdiction of the individual member states (Anthony 1991; Costa and Pereira 1999). The current process of Europeanisation of the defence industry, however, has made it imperative for member states increasingly to coordinate export policies. The tradition of national decision-making with regard to arms exports was linked to the existence of national defence industries, which have gradually ceased to exist due to fundamental structural changes in the industry, such as mergers, joint ventures and increased dependence on components from abroad (De Vestel 1993, 1995; Gummet and Walker 1993; Schmitt 2000).

This process of defence industry Europeanisation has been determined to a large extent by the post-Cold War reduction in military expenditure; the fast pace of technological change and spiralling costs of modern weapons systems; shrinking domestic and export markets; and increased availability of surplus weapons. Other factors contributing to an increased co-ordination of export policies are the creation of a common market for civilian and, with some restrictions, also for dual-use goods, and the fear of ‘undercutting’4 in the face of shrinking markets. These developments have also led to a push towards harmonisation at EU level from the defence industry itself, to create a ‘level playing field’ (European Defence Industries Group 1995). Furthermore, the need for more restrictive regulations across the EU as a tool of conflict prevention has become increasingly apparent, particularly considering the boomerang effect illustrated by the Gulf War and in peacekeeping missions where European soldiers fight troops equipped with weapons and equipment exported by their own governments. Linked to this, the need for a coherent EU foreign policy has become all too obvious given cases where past EU arms deliveries have been counterproductive for the EU’s humanitarian efforts and have contributed to the very development of these humanitarian disasters.

Cooperation has therefore been increasing in this field, but arms export policies have nevertheless remained national preserves. Those aspects of policy have so far remained within the intergovernmental second pillar. At the European Councils of Luxembourg and Lisbon in 1991 and 1992, the Council agreed eight criteria for arms exports. On the basis of these, a more comprehensive EU Code of Conduct for Arms Exports was elaborated by COARM, the Council Working Party on Conventional Armaments, and adopted by the Council of Ministers on 8 June 1998 as a Council Declaration. The Code document consists of three parts: a preamble, which outlines aims and underlying principles of the Code; export guidelines in the form of an elaboration of the eight criteria of 1991/92; and a number of operative provisions, which set out mechanisms for EU member states to cooperate on certain clearly defined matters. Primarily, member states are required to provide each other with annual reports on implementation of the Code, to review implementation of the agreement5 and to exchange information on the denial of export licences. This consultation mechanism, unprecedented in this field, aims to prevent undercutting. Before undercutting can take place, the two governments involved are required to enter into consultations.

As a Council Declaration, the Code agreement is only politically, not legally binding. And since it was agreed within CFSP, on the basis of Title V of the TEU, its implementation cannot be enforced by the European Court of Justice (Article 46 TEU). Thus no Community legislation on arms exports exists to date. And while the Code includes an elaboration of the original eight criteria, it leaves ample room for subjective interpretation by national governments (Amnesty International et al. 1998; Bauer 1998), even with regard to the operative provisions, for example regarding the contents of annual reports. Implementation is up to individual member states and can only be verified if information is provided by governments through other transparency mechanisms (such as national reports on arms exports, the UN Register for Conventional Arms, etc.) or by non-governmental sources (such as the defence industry).

The review and further development of the Code since its adoption reflect an increased Europeanisation of export policies. For example, a number of clarifications and agreements for practical implementation were made during the first two years (European Council 1999b; 2000b). While a clear European dimension of national export policies has developed, national governments have also tried to maximise their national positions through European channels. During negotiations for the Code of Conduct, for example, it was primarily the French government that blocked the agreement of more transparent mechanisms. On the other hand, under the Finnish Presidency, with its tradition of public and parliamentary transparency, the first annual report on implementation of the Code was published, although this had not been provided for in the original document. The document is a compromise agreement which reflects national traditions and negotiating positions. The Code is therefore a result of a Europeanisation dynamic as well as of member states’ attempts to maximise their national positions.

Inter-pillar consistency and the tension between supranationalism and intergovernmentalism

Foreign policy remains a competence of member states, who are not ready to cede it to the supranational level. Therefore attempts to build a European foreign policy since the Fouchet Plan in the 1960s have centred on the distribution of competencies between intergovernmental cooperation and control by supranational institutions. The Maastricht and Amsterdam Treaties illustrate this tension (Petite 1997; Remacle 1997; Dehousse 1998; Franck 1998). In the preparation of the Maastricht Treaty, the supranationalists, especially the Commission and the Benelux states, insisted vehemently on the need to give to the Community the coordinating role of an integrated external policy (Remacle 1992). The question of consistency between the three pillars thus became the sacred cow of supranationalists (Krenzler and Schneider 1997; Tietje 1997).

The emphasis on the need for a legal personality of the Union and its capability to sign agreements with third parties in the field of CFSP represented another example of the same debate (Bribosia and Weyembergh 1999). In Amsterdam, the approach appeared to be less ambitious and more pragmatic. Instead of reviving the old tensions between supranationalists and intergovernmentalists, negotiators adopted institutional working procedures designed to overcome some of the deep-rooted tensions (Gourlay and Remacle 1998). First, no legal personality is given to the Union as a whole, but Article 24 of the TEU allows the Union to sign agreements with third countries or organisations in the second and third pillars. Second, Article 3 of the TEU emphasises the need for consistency between all external policies and joint responsibility of the Council and the Commission in their implementation. Co-management of CFSP and other external policies by the Council and the Commission remains effectively the only way for the Union to proceed in shaping its political international profile. The appointment of Javier Solana as Secretary-General of the Council/High Representative for CFSP according to Article 26 of the TEU (and as Secretary-General of the WEU) as well as the designation of Chris Patten as ‘Super-Commissioner’ for External Relations6 are structural measures towards this objective. Despite some criticisms of federalists against the creation of the post of ‘Mr CFSP’ (Dehousse 1998), this solution of a Council–Commission tandem seems to be the only one able to reflect the consociational/confederal nature of the second pillar which makes it necessary to base it jointly on states and supranational institutions.

The question of consistency confirms therefore our theoretical assumption that EU foreign policy has to be conceived of as a ‘system’ of multi-level foreign policies rather than a unified set of external relations. This has led scholars to study how policy-making in this field addresses the consistency question and how institutions cooperate or compete in the clarification of their respective roles in the three pillars. Again, the case of arms exports is informative. We have already mentioned, for example, that the Code of Conduct is based on a Council Declaration, not on Community law. The regulation on dual-use goods of 1994 is also often quoted as an interesting case because it was a mix of a Community regulation and a CFSP joint action (European Council 1994a, b; Holland 1997), while in 2000 competence for dual-use goods was transferred to the European Commission (European Council 2000a). These cases concern primarily the question of the relationship between the Council and the Commission. Since 1998, the issue of arms exports has also been addressed by the Court of Justice, thereby creating a new dynamic in the field.

According to the Treaties, the jurisdiction of the ECJ does not extend to the second pillar, the field of CFSP. Title V of the Treaty on European Union is expressly excluded from the Court’s jurisdiction (Article 46 TEU). Nevertheless, decisions on access to second pillar documents agreed under Title V can be subject to European jurisdiction. A precedent for this has been set by a transparency case against the Council on access to second pillar documents, which was brought before the Court of First Instance in 1998 (Case T-14/98).

Heidi Hautala, a Finnish Green Member of the European Parliament, put a written question to the Council on the implementation of the eight EU criteria for arms exports of 1991 and 1992 (Hautala 1996). Subsequently, she requested access to a report on this matter, which had been agreed at the Political Committee level in 1993 (Hautala 1997a, b, 1999). Having been refused access to the document, Ms Hautala sued the Council, asking the Court of First Instance to annul the Council decision made at the ministerial level and to order the Council to bear the costs of the case (European Court of Justice 1999, para. 32). The Court ruled positively regarding both aspects and requested the Council to consider partial access to the document (para. 87). The Council, supported by Spain, appealed against this ruling, while Heidi Hautala was supported by Denmark, France, Finland, Sweden and the United Kingdom. The following aspects of the case are particularly relevant with regard to inter-pillar consistency:

Parties before Court

The General Affairs Council’s contested decision to refuse access to the document was not made unanimously: four states voted in favour of granting access to the document (European Court of Justice 1999, para. 20). Two governments, those of Sweden and Finland, even intervened in the written and oral procedures in support of the plaintiff’s request to annul the contested decision and her position on the Court’s competence (paras 25–6). The French government intervened in favour of the Council and dismissed the option of partial access (para. 63). Member states defending different positions before the ECJ can clearly be qualified as a first pillar practice, since the second pillar is formally excluded from ECJ competence.

Competence of the ECJ

Access to the contested document is governed by Council Decision 93/731 on public access to documents (European Council 1993). Therefore the applicant, supported by the two Scandinavian governments, maintained that it fell under Article 173 of the EC Treaty (now Article 230), regarding ‘review by the Court of Justice of the legality of Council acts’. Thus, according to Hautala, access to the document was within the Court’s jurisdiction, even if the document fell within Title V (European Court of Justice 1999, paras 21–2).

In the written procedure, the Council, supported by the French government, challenged the Court’s competence in this matter on the grounds of the relevant Treaty provisions (Vesterdorf, Judge-Rapporteur 1999, paras 21–2). It did not, however, argue in favour of inadmissibility and ‘left it to the Court of First Instance to consider the question of its jurisdiction’ (European Court of Justice 1999, para. 36). During the oral hearing, it acknowledged the Court’s competence in this specific case, while it outlined certain limits.7

The Court argued, as in a similar third pillar case,8 that the decision on rules governing access to Council documents, as an act based on Article 151(3) of the EC Treaty (now Article 207(3)) applied to all Council documents (European Court of Justice 1999, para. 41). It further held that the Court not having ‘jurisdiction to assess the lawfulness of acts falling within Title V did not exclude its jurisdiction to rule on public access to those documents’ (ibid., para. 42). The Court, however, limited its own jurisdiction using a distinction between political and legal assessments. It stated that ruling on the question of whether partial access had to be considered fell within its competence, since it was a legal question (ibid., para. 76). However, it did not consider a ruling on the Council’s assessment of potential threats to the public interest and the EU’s external relations as falling within its jurisdiction (ibid., para. 71). It defined its own competence as ‘limited to verifying whether the procedural rules have been complied with, the contested decision is properly reasoned, and the facts have been accurately stated, and whether there has been a manifest error of assessment of the facts or a misuse of power’ (ibid., para. 72).

Interpretation of the principle of widest possible access to documents

A key document on the principles of transparency and citizens’ right of access to information is a Code of Conduct agreed by Council and Commission on public access to these institutions’ documents (European Commission and European Council 1993), which establishes a principle of ‘widest possible access to documents held by the Commission and the Council’ (p. 41). Public access to a document can be refused under specifically named circumstances,which are also confirmed in Council Decision 93/731 on public access to documents (European Commission and European Council 1993: Article 4).9 This adds to interpillar consistency, although the lack of an enforcement mechanism is a clear difference with regard to the Council. To some extent (taking into consideration the limited competence of the Court) this consistency gap is narrowed through the Court case in question.

In the case discussed here, the Council evoked public security as one of the limitations to widest public access provided for in Article 4(1) of Decision 93/731 (European Council 1997a) and potential harm to the EU’s relations with third countries (European Council 1997b). It further maintained that the type of document requested automatically led to a refusal of access.10

This court case was the first one at EU level in which the Court explicitly ruled that partial access had to be considered by one of the EU institutions (Hautala 1999). The Council interpreted the rules governing access to documents as only providing for access to complete documents and stated that fragmentary information could be misleading. It emphasised that a case where partial access was granted by the Council had been an exception (European Court of Justice 1999, paras 55–9). The Court interpreted access to documents as access to the information contained in them (para. 75). It further based its position on the principle of proportionality, maintaining that the aim of protecting ‘the public interest with regard to international relations’ could be achieved even when parts of a document are published (para. 85). While the Court emphasised that its competence did not extend to an assessment of the Council’s decision based on the contents of the report, it ruled that partial access had to be considered. The Court therefore assumed a role in the interpretation of the Council Decision and thus participated in its further legal development through the creation of case-law.

One can conclude that first pillar practices and principles were applied to a second pillar issue, thereby providing an example of a spillover effect from the first to the second pillar with regard to access to documents, and therefore transparency and accountability, through the admission of ECJ jurisdiction. European Court of Justice jurisdiction regarding public access to a second pillar document thus illustrates at the same time an aspect of Europeanisation of arms export policies (in the sense of a new dimension to national policies) as well as the problem of inter-pillar consistency.

Variable geometry

The trends towards convergence and consistency in EU foreign policy indicate that decision-making processes have been integrated to some extent. But they also show that this integration process is not similar to first pillar policy-making, but is rather more complex and multi-level. Our hypothesis about the peculiarity of foreign policy integration is again verified when we take into account a third dimension of the Europeanisation process; that is, variable geometry. The shaping of concentric circles, differentiated security dilemmas and interlocking regional security subsystems and organisations (Remacle and Seidelmann 1998) as aspects of variable geometry characterise the evolving nature of the European security system since the end of bipolarity. In addition, variable geometry has been gradually introduced into the institutional framework of the European Union with the development of the second pillar (Edwards and Philippart 1997; Missiroli 1998).

The Amsterdam Treaty does not formally include a provision for enhanced cooperation in the second pillar, as it does for the first and third ones. No agreement was possible on this point because of the reluctance of large member states which were interested in maintaining the option of establishing contact groups or ‘directoires’ for managing international questions or crises. Nevertheless, Article 23 introduces the option of constructive abstention of countries, if they represent at most one-third of votes in the Council, when adopting guidelines or common strategies. In addition, qualified majority voting was introduced for decisions on common positions and joint actions, although a country can still evoke national interests to veto such a majority decision. As far as the outside representation of the Union is concerned, foreign policy will generate a system of variable ‘troikas’: besides the traditional troika (preceding, present and succeeding Presidencies plus the Commission), a special troika is created for CFSP (present and next Presidencies plus the High Representative for CFSP plus the Commission). Yet another troika has been set up for the external representation of the euro (the Presidency of euro-,11 the President of the European Central Bank, and one of the three EMU participants in G-7 assisted technically by the Commission). This means the introduction of a kind of variable geometry in foreign policy. Furthermore, security policy is most affected by variable geometry, and again the field of armaments policy provides a clear example of this trend.

In the past, member states have interpreted Article 296 of the EC Treaty such that all matters relating to the production and trade in arms have been considered as falling within national jurisdiction. While attempts were made within the first and second pillar (Agstner 1998; European Commission 1996, 1997) to achieve some progress along the lines of coordination and harmonisation of armaments policies, few substantial results have been produced. This is due to dramatically diverging interests as a result of the different sizes of national defence industries, different defence doctrines, and membership in different defence alliances or even the status of neutrality. To date, there has been no integration of defence procurement and defence markets in the EU, although Article 17 of the TEU mentions the importance of coordination on armaments issues.

Since the beginning of the 1990s, economic, technological, strategic, political and industrial developments have pushed governments to seek progress in the field of armaments cooperation (Bauer and Winks 2001). Economic imperatives of cost reduction and the increased integration of the defence industry have arguably been the most important factors driving this process. This has led to a revitalisation of established bodies and to the creation of new mechanisms outside the existing institutions.

Armaments cooperation within NATO involves a complex structure of initiatives and bodies, including the Conference of National Directors of Armaments. The thirteen European member states of NATO (except Iceland) also meet in the West European Armaments Group, and can use the WEU’s continuing treaty-based legal personality through the West European Armament Organisation to commission studies and sign contracts. Third, the four main arms producers in the EU – that is, Britain, France, Germany and Italy – have founded a multilateral structure of co-ordination in the field of arms procurement called the Joint Armaments Cooperation Structure (JACS, mostly known by its French acronym OCCAR), which gained legal status in January 2001 (Agence Europe 2001).

In addition to cooperation within these bodies, there has been a recent trend to deal with armaments policy outside the established institutionalised framework. This is due to the failure of previous multilateral attempts to agree a common political and legal framework for the European defence industry. In July 2000, the defence ministers of the EU’s six biggest arms producers and exporters (Germany, France, Sweden, Spain, Italy and the UK) signed a Framework Agreement, ‘Concerning Measures to Facilitate the Restructuring and Operation of the European Defence Industry’11 (Schmitt 2000; Bauer and Winks 2001; House of Commons 2001). This initiative goes back to a ‘Letter of Intent’ signed by the aforementioned six countries in July 1998, in which the six defence ministers stated their desire to ‘establish a co-operative framework to facilitate the restructuring of the European Defence Industry’.12 Working groups consequently negotiated agreements within the following areas: security of supply; transfer and export procedures; security of classified information; defence-related research and technology; treatment of technical information; harmonisation of military requirements; and protection of commercially sensitive information. The document will be fleshed out through subsequent agreements in these specific areas. The Treaty has been ratified by all signatory states except Italy, and is in principle open for other states to join, provided there is consensus among the original signatories.

In the field of export policies, the Agreement established a mechanism to negotiate common ‘white lists’ of countries eligible to receive certain defence goods. Those countries involved in the joint production will negotiate these product-specific lists in advance. The mechanism contains a provision for adding or deleting potential recipients at the request of a contributing country. A rule of consensus through each participating state’s right to veto an export destination was thus integrated into the treaty. The current provision is a departure from the previous principle applied in Europe, which placed the responsibility for an export on the country in which the final assembly takes place. Export decisions will be based on the principles laid out in the EU Code of Conduct. The Agreement also facilitates transfers of equipment between the participating countries. This could be a first step towards a common market for defence goods within the EU.

Despite the impression of dispersion generated by variable geometry, previous experience has shown that variable geometry and enhanced cooperation are the best solutions for maintaining the momentum of integration and leading to an integration of most EU countries in new circles of cooperation (Telò and Remacle 1998). EMU and Schengen are two good examples of this. The transfer of the WEU ‘acquis’ into the EU, decided in 1999 and legalised by the Nice Treaty, seems to support this trend. Therefore, variable geometry is not only an indication of fragmentation of the Union but also an element of the continuing process of integration.


The delicate balance between integration and fragmentation is shaping a very peculiar model of foreign policy-making which cannot simply be described as a traditional international organisation nor as a state-like polity, but rather as a multi-level system. Evidence of all four analytical levels (conflicting, cooperative and institutionalised intergovernmentalism, and supranationalism) can be seen. There is clear evidence of spillover from the first to the second pillar. It has, however, been shown that functionalism, while an important driver for integration, is not the sole driving force behind the Europeanisation process. Institutional factors also play an important role. Various forms of ad-hoc cooperation between EU member states outside the established institutionalised system are driven by a convergence of preferences and attempts to maximise national interests. At the same time, political resistance to joint decision-making remains, arising from the traditional culture in this domain which is centred around an insistence of sovereignty, as traditionally defined, and nation-state competence for security and defence matters.

In looking at the aforementioned four levels of EU foreign policy through the aspects of consistency, convergence and variable geometry, this contribution shows parallel processes of integration and fragmentation. The field of arms export policies illustrates these tendencies. On the one hand there is a clear Europeanisation dynamic towards a harmonisation of national policies at the EU level. On the other hand, member states resist legal changes that would cede competence regarding arms exports to the supranational level. Therefore, decision-making on armaments and dual-use goods takes place at all four levels. The various levels of foreign policy-making inside the EU allow for different speeds and degrees of integration in different policy areas of the three pillars as well as within the pillars, and also in various aspects of the same policy area.


1 For example, in the framework of the UN Conference on the Illicit Trade in Small Arms and Light Weapons in All Its Aspects, held in July 2001.
2 Items with both military and civil applications.
3 Article 296(1) states: ‘(a) No Member State shall be obliged to supply information the disclosure of which it considers contrary to the essential interests of its security; (b) any Member State may take such measures as it considers necessary for the protection of the essential interests of its security which are connected with the production of or trade in arms, munitions and war material; such measures shall not adversely affect the conditions of competition in the common market regarding products which are not intended for specifically military purposes.’
4 ‘Undercutting’ refers to the delivery of weapons by a government after the export of essentially the same category of weapons was denied by a different government.
5 The first annual review took place on 21 September 1999, during the Finnish Presidency.
6 The Amsterdam negotiators had proposed to appoint a Vice-President of the Commission for External Relations. President Prodi was not able to fulfil this request but gave to Commissioner Patten a wide portfolio in external affairs which is almost equivalent to the power of a vice-president.
7 It stated that ‘assessment of the harm which might be caused to the public interest by the disclosure of one of its documents is within its sole discretion’ (European Court of Justice 1999, para. 57).
8 In case T-174/95 (Svenska Journalistförbundet v Council of the European Union), the Swedish Journalists’ Association challenged denial of access to a document and won the case, even though third pillar matters are also expressly excluded from the Court’s competence (European Court of Justice 1998). Thus, parallel to the case discussed here, the Court’s competence was also not evident from the Treaties. That case therefore established a precedent which is relevant for second pillar matters.
9 ‘Access to a Council document shall not be granted where its disclosure could undermine:
  • the protection of the public interest (public security, international relations, monetary stability, court proceedings, inspections and investigations),
  • the protection of the individual and of privacy,
  • the protection of commercial and industrial secrecy,
  • the protection of the Community’s financial interests,
  • the protection of confidentiality as requested by the natural or legal person who supplied any of the information contained in the document or as required by the legislation of the Member State which supplied any of that information.’
  Also, ‘[a]ccess to a Council document may be refused in order to protect the confidentiality of the Council’s proceedings’.
10 The Council argued that publication of an internal working document could impact the functioning of CFSP mechanisms and thereby harm public interest (European Court of Justice 1999: para. 60).
11 Available via
12 Also available via
  • Collapse
  • Expand

All of MUP's digital content including Open Access books and journals is now available on manchesterhive.



All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 2445 852 35
PDF Downloads 1002 131 23