Context and Process
The Treaty of Amsterdam (1997), officially known as “The Treaty of Amsterdam amending the Treaty on European Union, the treaties establishing the European Communities and certain related acts,” is the result of the 1996–1997 Intergovernmental Conference (IGC). As foreseen in the Treaty on European Union (Art. N, TEU, 1992) signed in Maastricht in February 1992, the IGC was convened in 1996.1 Part of the agenda including the further extension of the scope of qualified majority voting (QMV) and codecision, questions related to the security of the Union, and new provisions on energy, civil protection, and the hierarchy of norms that had already been defined in Maastricht. Other issues such as flexible integration and the integration of Schengen into the Treaty were new.
The negotiations took place against the context of the upcoming Eastern enlargement. The Madrid European Council of December 1995 confirmed that the decisions on launching the accession negotiations would be taken within six months of the conclusion of the IGC (Glenn, 2003).
Given its long-term “advance notice,” the IGC leading to the Treaty of Amsterdam(1997) has probably been one of the most well-prepared IGCs ever. Barely half a year after the entering into force of the Maastricht Treaty (November 1993), the European Council of Corfu already convened a Reflection Group on the Functioning of the Union (European Council, 1994). The group—chaired by the Spanish minister for European Affairs Carlos Westendorp and composed of representatives of the EU ministers for foreign affairs, the president of the European Commission, and two members of the European parliament (MEPs)—had the mandate to come up with ideas for revision and improvement, taking into account the future enlargement of the EU. It met 14 times between June and December 1995 (Stubb, 2000) and identified three core priorities: making Europe more relevant to its citizens, enabling it to work better and preparing it for enlargement, and giving it greater capacity for external action. All three objectives were taken over by the Turin European Council of March 29, launching the IGC (European Council, 1996).
The objectives proposed by the Reflection Group did not come as a surprise. The fierce ratification debates during the Danish and French referendums on the Maastricht Treaty had made it clear that public support for the European integration process could no longer be taken for granted (see “The Maastricht Treaty: Creating the European Union”). Citizens felt that decisions taken at the EU level were not sufficiently taking into account their views, and EU institutions were increasingly accused of lacking democratic legitimacy (see “Democratic Deficit in the European Union” in this volume). With the expected doubling of EU member states, a substantial reform of EU institutions and the decision making rules were seen as indispensable for the future survival of the Union. The EU’s incapacity to address the Yugoslav conflict had shown the limits of the EU as a civilian power and put the question of a European security role, including the development of military capabilities, again on the agenda.
As the name suggests, the key players in an IGC are the member states, meeting under the chairmanship of the rotating presidency and deciding by unanimity (Christiansen & Reh, 2009). The 1996–1997 IGC was the first IGC since the enlargement with Austria, Finland, and Sweden, who had joined the European Union (EU) in 1995. In the day-to-day management of the process, the national delegations and the presidency are supported by the Brussels-based Council general secretariat, which provides them with administrative backing and legal advice. Formally, all member states are equal in terms of representation and voting power (one vote per member state), but in practice bigger member states with more capacities at their disposal tend to be more often in the lead. France and Germany have traditionally fulfilled a motor role, especially when it comes to questions such as institutional reform and the development of European foreign policy. In interaction with their domestic constituencies, the member states determine their national preferences, which they translate into national position papers, presented either unilaterally or in cooperation with one or more like-minded delegations (Vanhoonacker, 2012).
The negotiations leading to Treaty of Amsterdam (1997) were conducted under the lead of three successive presidencies, that of Italy (first half of 1996), Ireland (second half of 1996), and The Netherlands (first half of 1996). In turn, each of them chaired the meetings at all levels, ranging from the European Council to the meeting of ministers of Foreign Affairs and their Group of Representatives in charge of the preparatory work.2 The country at the helm would manage the agenda and play an important role in forging agreements. Being at the center of the negotiation process gave it preferential access to information and a privileged position to influence the outcome. The role of the successive presidencies involved in the 1996–1997 IGC was very much influenced by the phase of the negotiations. The role of the Italian presidency consisted primarily in exploring the different positions. This led to the presentation of a Progress Report outlining the emerging trends, which was presented at the Florence European Council of June 21–22, 1996 (President, 1996). By the end of the Irish presidency, there was already a draft treaty text, which was presented to the European Council of December 13–14 in Dublin. Sensitive issues such as the integration of Schengen into the treaty, the communitarization of Justice and Home Affairs, the relations between the EU and the Western European Union (WEU), flexible integration and institutional reform could only be settled under the final Dutch presidency.
The role of the supranational institutions in an IGC is not as strong as in the EU legislative process, whereby the European Commission (EC) has the exclusive right of initiative and the European Parliament (EP) is a colegislator with the Council. Although the EC is present at the meetings at all three levels (European Council, ministerial level, preparatory group), it has no veto over the final results (Christiansen & Reh, 2009). At the time of the 1996–1997 IGC, the EC was headed by the former Luxembourg Prime Minister Jacques Santer (1995–1999), who had been appointed after the British had vetoed the nomination of the Belgian Prime Minister Jean-Luc Dehaene. While previous EC President Jacques Delors, through his active role in the preparations on the creation of an Economic and Monetary Union (EMU), had a prominent voice in the Maastricht Treaty negotiations, Santer took a more pragmatic and low-key approach. The role of the EC was geared toward closely cooperating with the rotating presidency and the Council General Secretariat (Gray, 2002).
The EP is only an associated member of the IGC and can try to influence the process through the adoption of resolutions. At the Reflection Group preparing the IGC, it was represented by the French socialist MEP Elisabeth Guigou and the German Christian Democrat Elmar Brok (Maurer, 2002). The member states also accepted the involvement of the EP representatives beyond the Reflection Group debates, but France and the United Kingdom blocked their presence in critical decision sessions (Moravcsik & Nicolaïdis, 1999).
The IGC negotiations were concluded in Amsterdam in the early hours of June 18. After further polishing by the legal and linguistic services of the Council General Secretariat, the treaty was signed on October 2, 1997. In 13 of the 15 member states, it sufficed to ratify the treaty through parliament. In Denmark and Ireland, there was a referendum. Compared to the Maastricht Treaty ratification, in which the Danes only approved the amendments after further concessions and a second referendum (see “The Maastricht Treaty: Creating the European Union”), the process was relatively smooth and hardly attracted any public attention. Contrary to the referendum 1992, the Danes ratified the treaty with one vote, with a majority of 55.1%. In Ireland, the voters approved the treaty with 61.7% of the vote (referendum of May 22, 1998) (Hug & Konig, 2000). The treaty entered into force on May 1, 1999.
Scholarship
Although the Treaty of Amsterdam (1997) does not carry the same weight as the Maastricht text, it nevertheless raised considerable scholarly attention. A first group of publications was prepared by players who were either directly or indirectly involved in the negotiations. One of the first studies, already published in 1997, was a full text of the treaty, accompanied by a clause-by-clause commentary. It had been drafted by Andrew Duff, who at the time of the Amsterdam negotiations was vice president of the Liberal Democrats in the United Kingdom (1994–1997). While giving specific attention to the British position and context, the text also gives interesting information on other key players in the process. In his account “Original Sin in a Brave New World,” Bobby McDonagh (1998), an Irish diplomat who participated in the Reflection Group and the IGC negotiations, gives an interesting insight into the positions of the member states and EU institutions and into how the final treaty amendments gradually took shape. Another insider’s view is that provided by Franklin Dehousse, the Belgian representative to the Reflection Group (Dehousse, 1999).
In the series of academic studies, one of the most comprehensive accounts is the volume The Amsterdam Treaty: National preference formation, interstate bargaining and outcome, edited by Laursen (2002a). This publication not only contains chapters about the preferences of the main actors in the IGC but also examines the substance of the treaty changes in key policy fields. It concludes with a theoretical chapter explaining the final outcome of the bargaining process. An edited volume by Neunreither and Wiener (2000) primarily focuses on institutional reform, questions of democracy and flexibility, and tries to make theoretical sense of the treaty. A study by the Research Institute for European Affairs in Vienna, published in the same year with contributions by Griller, Droutsas, Falkner, Forgó, and Nentwich (2000) mainly focuses on the substance of the treaty.
The Treaty of Amsterdam (1997) also caught the attention of legal experts. In a follow-up to an earlier study on the Legal Issues of the Maastricht Treaty (O’Keeffe, 1994), O’Keeffe and Twomey (1999), for instance, edited a collection of more than 20 essays on constitutional, institutional, and policy changes.
There are also authors choosing to focus on specific dimensions of the treaty. Stubb (1998), for instance, dedicated his PhD to the negotiations on the provisions on flexible integration. As a civil servant at the Finnish Permanent Representation at the time, he was well informed and had preferential access to sources. He particularly emphasizes the importance of incremental learning during the negotiation process, leading to some fluctuation in the national positions. He pays special attention to the role of the bureaucratic level and more particularly that of the civil servants of the rotating presidency and the Council general secretariat. Svensson (2000) builds on his work and dedicates a full study to the role of successive presidencies in the prenegotiation and negotiation phases, exploring their room for maneuver as central players in a complex and highly dynamic negotiation process.
Next to the above-mentioned extensive studies, the Treaty of Amsterdam (1997) also gave rise to a range of journal articles dealing with the process and/or the substance of the treaty. Most authors chose for an in-depth study of one particular dimension of the treaty such as the negotiation process (Devuyst, 1998), the institutional dimension (Dehousse, 1998; Sverdrup, 2002), flexible integration (Edwards & Philippart, 1999), and social policy and employment (Venturini, 1998). One of the most seminal studies is a contribution to the Journal of Common Market Studies by Moravcsik and Nicolaïdis (1999). The outcome of the negotiations can in their eyes best be explained by interstate bargaining with a marginal role for the supranational actors. The distribution of the gains is primarily determined by asymmetrical interdependence among governments (see “Evaluating and Understanding Amsterdam”).
Main Achievements
As mentioned above (see “Context and Process”), the IGC leading to the Treaty of Amsterdam (1997) aspired to make Europe more relevant to its citizens, to improve its functioning in light of the upcoming “big bang” enlargement toward the east, and to give it greater capacity for external action. The present section gives an overview of the main achievements in these three areas and dedicates a separate section of flexibility, one of the most heavily debate innovations of the new treaty.3
Bringing Europe Closer to Its Citizens
The fierce public debates around the ratification of the Maastricht Treaty were a brutal wake-up call for many national and European politicians. A recurring point of criticism was that the European integration process, with its strong focus on economic liberalization, had primarily benefited big companies while leaving the ordinary citizen in the cold. During the preparations of the IGC, it was therefore considered important to show results in areas of direct concern to the man in the street, such as employment, social policy, fundamental rights, migration, and the fight against crime.
When the IGC was launched in the first half of 1996, there were 18 million EU citizens who were out of work, and in Spain, unemployment rates had even risen to 22.2%. It is in that light that the treaty introduced a new title on employment (Title VI, TEC) (Duff, 1997). The member states agreed to work toward a coordinated European strategy, and a high level of employment was also to be taken into account in the development of other policies. One of the driving forces behind the new chapter was the Swedish social-democratic government, striving to counterbalance the strong focus on monetary stability under the EMU with a policy of growth and job creation. Sweden also pleaded for the creation of a new Employment Committee aimed to counterbalance the economic and financial committee that was established under the Maastricht Treaty (Duff, 1997). Important allies of the Swedish delegation were Austria, Belgium, Ireland, Denmark, and Finland. With the election of Lionel Jospin as French Prime Minister in June 1997, France also became an important voice in the debate, bringing the number of countries in which social-democratic parties were part of government up to 12 (Dehousse, 1999). In practice, however, the chapter was primality symbolic, since it did not provide the EU with substantial instruments to address the challenge of unemployment.
Social policy had already been on the agenda in Maastricht, but as a social chapter was inacceptable to the British Prime Minister John Major, the provisions on social policy had to be parked in a separate protocol, and Britain was given an opt-out. The change in government with the election of a new Labour government under Tony Blair in May 1997—shortly before the conclusion of the IGC—led to a change in position. The new U.K. government gave up its opt-out from the social chapter and the provisions of the Maastricht social protocol were integrated into the social-policy chapter of the Treaty (Best, 2002). Still, Labour remained reluctant to introduce QMV in this sensitive policy field but was ready to make an exception for the article on equal pay for equal work (Art. 199, TEC).
Migration was a further major preoccupation of many European citizens. At the time of the Amsterdam negotiations, the migratory pressures were mainly coming from central and eastern Europe. Germany was a popular destination and was pressuring for a European-level approach. In addition, the intergovernmental decision-making process agreed upon in Maastricht proved ineffective. In order to allow the EU to be more responsive to these urgent political challenges, all questions related to asylum and migration were removed from the intergovernmental pillar on Justice and Home Affairs to the first community pillar. This communitarization implied the exclusive right of initiative for the EC (after a five-year transition period) and judicial review by the European Court of Justice. The price to be paid, however, was opt-outs for Denmark, Ireland, and the United Kingdom. The communitarization was also extended to judicial cooperation in civil matters, customs fraud, and fraud against the community budget
Additionally, the heads of state or government, thanks to major efforts of the Dutch presidency, for whom this was a major priority, also managed to reach an agreement on the integration of the Schengen acquis into the treaty by means of a protocol (den Boer, 2002). The Schengen area—of which all EU member states, with the exception of the United Kingdom and Ireland, were part—was a zone where internal border controls had been abolished and a common visa policy applied. The Schengen agreement of 1985, which in 1990 had been supplemented with a Schengen Convention, had so far operated outside the treaty framework and without the involvement of the EU institutions. The Schengen Protocol consisted of a general agreement making it possible to incorporate all the decisions and declarations adopted under Schengen into the community legislation and ensuring that future cooperation would take place within the legal framework of the EU. The United Kingdom objected to the communitarization of Schengen and received an opt-out. As a result, Ireland, who was supportive of the integration of Schengen but wanted to safeguard its common travel area with its neighbor, had to follow the British opt-out. It was, however, foreseen that both countries could “at any time request to take part in some or all of the provisions of the Schengen acquis” (Protocol on the Schengen Acquis, 1997). Denmark, which had recently joined the Schengen area, decided to subscribe to the legislation adopted under the third intergovernmental pillar but received an exemption for the laws adopted under the supranational first pillar.
Finally, the new treaty firmly enshrined the fundamental rights of the EU citizens into the Treaty (Art. 6, Title I, TEU, 1992) by explicitly mentioning that “the Union is founded on the principles of liberty, democracy, respect for human rights and fundamental freedoms and the rule of law.” If a country is in breach of these principles, the European Council can suspend certain rights (including voting rights) of the member state in question. As a first step, however, the European Council, acting by unanimity upon a proposal by one third of the member states or the EC and with the assent of the EP has to determine the existence of “a serious and persistent breach” by a member state of the earlier mentioned principles (Art. 7 (1), TEU, 1992). The procedure was introduced in anticipation of the enlargement with several formerly communist countries. The EC triggered the article for the first time in 2017 in a case against Poland because the independence of the judiciary and the rule of law was considered to be under threat.
Institutional Reform
The debate on institutional reform was spurred on for various reasons. Next to concerns about questions of legitimacy and more efficient decision making, the prospect of further enlargement was a major factor. There were serious concerns that without a substantial reform of the existing rules of the game, the EU machinery would become paralyzed. The discussions revolved around a broad range of issues, including the extension of QMV, the reweighting of the votes in the Council, further generalization of the codecision procedure whereby the EP acts on equal footing with the Council, and the future composition of the EC (Laursen, 2002b).
Measures aimed at increased efficiency included the reduction of the number of legislative procedures by the abolition of the cooperation procedure, the reduction of the number of MEPS to 700, and the extension of QMV. The last of these proved extremely sensitive. As the proposal of the Reflection Group to make QMV the general rule was not acceptable to many delegations, the extension needed to be negotiated on a case-by-case basis. While the list of areas falling under QMV was extended by 14 new areas,4 especially in flanking areas of the internal market, many more-sensitive fields such as culture, social security, indirect taxation, and professional services continued to be decided by unanimity. An important factor in the poor results was the change in position of Germany, traditionally a strong advocate of the extension of QMV. Chancellor Helmut Kohl was under severe domestic pressure of the German Länder, which wanted to protect their sovereign rights in areas such as migration and the environment (Duff, 1997).
The legitimacy debate centered largely on a strengthened role of the EP. The scope of the codecision procedure introduced in Maastricht and putting the EP on an equal footing with the Council, was greatly extended to many new and old treaty provisions, but decisions on the Common Agricultural Policy, still representing more than half of the EU budget, remained excluded. Codecision itself was simplified by abolishing the third reading. Furthermore, the EP’s role in the nomination of the president of the EC was considerably strengthened by making its vote on the candidate for Commission president legally binding and giving the approved president the right to nominate the remaining commissioners together with the member states. In addition, at the instigation of France, there was also an attempt to enhance the role of national parliaments in European decision-making. Although their expertise and capacities are more limited than those of the EP, they are considered to be closer to the citizens and their concerns (Corbett, Jacobs, & Shackleton, 2011). In a declaration attached to the Maastricht Treaty, the member states had already committed themselves to inform national parliaments better by making sure that they would receive EC proposals in a timely fashion. In Amsterdam, this was followed up with a legally binding Protocol, which gave national parliaments a period of minimum six weeks to scrutinize a legislative proposal before it was put on the Council agenda. This was much less ambitious than a French proposal for a High Consultative Council of 30 national deputies scrutinizing European Affairs, which was rejected. The debate on the role of national parliaments would linger on beyond Amsterdam and come back during the Convention on the Future of Europe and the negotiations on the Lisbon Treaty (Corbett, Jacobs, & Shackleton, 2011).
Last, there were also two key issues on which the member states did not reach agreement in Amsterdam: the size of the EC and the reweighting of the Council voting system. At the time of Amsterdam, the treaty stipulated that the college of commissioners was composed of at least one national per member state but not more than two of the same country (Art.157, TEC). In practice, this meant that the big member states (France, Germany, Italy, Spain, and the United Kingdom) each had two members, and the smaller countries only one. With the upcoming enlargement, this practice was no longer tenable, as it would lead to a commission of more than 30 members. While the larger countries were ready to give up their second commissioner, many of the small member states were not ready to have no commissioner at all. The proposal by Spain, for instance, whereby all the large member states would have one commissioner in combination with a system of rotation among the small ones was rejected. Furthermore, the suggestion by France to work with a small commission with 10 to 12 members reflecting a regional balance could not count on sympathy. The negotiations were further complicated by the fact that the large member states that would be the obvious losers in this dossier linked it the reweighting of votes in the Council. With every enlargement, the large member states had seen a further decline of the population represented in QMV. All of them, with the exception of Germany, wanted to compensate for this with an increased number of votes. The German government favored a system of double majority, supplementing the existing scheme with the requirement that a majority would also represent the majority of EU citizens (Laursen, 2002a). Despite the efforts of the Dutch presidency, the dossier of the size of the EC and the weighting of votes did not prove ripe for decision. As there were still some years to go before the next enlargement, it was decided to postpone the issue to a later IGC to be convened before the EU membership would exceed 20 (Protocol on the institutions, 1997). Not surprisingly, an agreement on the size of the EC was again linked with that of the reweighting of votes (Protocol on the institutions, 1997).
Of the three core objectives of the IGC, the results on the field of institutional reform were undoubtedly the weakest. This does not come as a surprise: the questions on the agenda go to the core of the relative influence and power of member states in EU policy-making.
EU External Action
The end of the cold war not only had led to the unification of the European continent, but also raised fundamental questions about the future shape of the European security landscape. The initial reluctance of the United States to intervene in the Yugoslav crisis and the incapacity of the EU to bring peace in the successor states of the former Yugoslav Republic had given new impetus to the development of a fully fledged European foreign and security policy. The strenuous debates in Maastricht had made it clear that the member states had widely diverging views on the future European security architecture. While Atlanticist countries such as the United Kingdom, The Netherlands, and Portugal favored a strengthening of the European pillar of NATO, Europeanists as France pleaded for an independent European security role and its own capacities. For neutral Ireland, security was to stay outside the scope of EU competencies (see “The Maastricht Treaty: Creating the European Union”). The foreign-policy debate in the 1996–1997 IGC was centered around two issues: a more efficient decision-making process and the development of the security dimension of the Common Foreign and Security Policy (CFSP).
Although the Maastricht Treaty had opted for a single institutional framework for all EU policy fields, the decision-making process in CFSP remained organized on an intergovernmental basis, with the member states deciding by unanimity and only a coright of initiative for the EC. France and Germany, traditionally fulfilling a motor role in the European foreign-policy debate, found themselves at opposite sides. Paris firmly opposed the German plea for a communitarization of CFSP and the introduction of QMV in foreign-policy decisions. The compromise consisted in unanimity remaining the rule but allowing the use of QMV for implementing decisions. The same applied for decisions implementing common strategies, a newly introduced instrument used by the European Council to determine an overall policy frame toward a particular country or region (Art. 23, Title V, TEU, 1992). In addition, Amsterdam also introduced the use of constructive abstention: this new form flexibility made it possible for member states to abstain from a decision without blocking it (Art. 23, Title V, TEU, 1992).
The most important institutional innovation in CFSP was the creation of the post of a High Representative (HR) for the CFSP, an idea stemming from a French proposal. The institutional home of this new player was heavily debated, ranging from proposals to upgrade the role of the secretary general of the Council general secretariat, to integrate the HR into the EC or, as suggested by the French, to create a new position outside the existing institutions (McDonagh, 1998). In the end, the first option was agreed upon: the role of HR would be fulfilled by the secretary general of the Council general secretariat (Art. 26, Title V, TEU, 1992).The HR was not an autonomous player with the right of initiative but had as mandate to contribute to “the formulation, preparation and implementation of policy decisions” (Treaty of Amsterdam, Art. J.16). Until the entering into force of the Lisbon Treaty, the post would be occupied by Xavier Solana, former secretary general of NATO and former Spanish minister of foreign affairs. He was assisted by a small Policy Planning and Early Warning Unit (PPEWU), composed of civil servants and diplomats from the member states, the Council general secretariat, and the EC (Duke, 2002). The PPEWU also engaged in foreign-policy analyses and therefore went beyond the traditionally administrative role of the secretariat. Both the HR and the PPEWU were further steps in the gradual Brusselsization and institutionalization of European foreign policy. With the Lisbon Treaty, the HR would become a member of the college of European Commissioners, receive the right of initiative, and would receive a fully fledged administration through the European External Action Service (EEAS).
The development of a European security identity and capacity was even more sensitive. The dividing lines between Atlanticists and Europeanists mentioned earlier in this section were as present as ever, and the discussions were further complicated by the neutral and nonaligned countries such as Austria, Finland, Ireland, and Sweden. The Europeanists, such as France, Germany, Italy, Belgium, Luxembourg, and Spain, advocated a three-step integration of the WEU into the EU. The WEU was the European security organization, which the Maastricht Treaty had put in charge of the implementation of the security decisions of the EU. As in Maastricht, this was still unacceptable for the United Kingdom, which did not want to endanger the central role of NATO. It received the support of the then-enlarged group of neutral countries that wanted to limit the EU security role to that of a peacekeeper. It was only after the poor EU performance in the 1998–1999 Kosovo crisis that an agreement on this matter was reached during the Nice Treaty negotiations (Sbragia, 2014). A Finnish-Swedish proposal to integrate the WEU crisis management tasks as defined in the 1992 Petersberg declaration into the treaty was more successful. The new reference to “humanitarian and rescue tasks, peace-keeping tasks and tasks of combat forces in crisis management, including peacemaking” (Treaty of Amsterdam, Art. J.7) could also count on the support of the Atlanticist countries, since it was an implicit recognition that “hard security matters” and collective security would remain the responsibility of NATO.
Flexibility
Flexible integration had already been a topic of discussion since the 1970s (Thym, 2017), but it was in Amsterdam that the member states for the first time tried to elaborate rules about its scope and decision-making rules. With every enlargement, the EU was becoming more diverse economically, politically, and culturally. As had been well illustrated with the difficult discussions on EMU, social policy, and increased security integration, it was not always possible to come to an overall agreement. In order to avoid a deadlock, the heads of states and government had started to resort to the handy instrument of opt-outs in particularly sensitive areas. An enlargement to the east brought with it the expectation that the diversity in views would only increase. With the aim to reduce the risk of a total paralysis, it was agreed to explore in more detail the possibilities of differentiated integration whereby some member states would move ahead in certain policy areas, whereas other would (temporarily) stay on the sideline.
The debate on flexible integration had already taken off prior to the IGC with the launch of the Schäuble and Lamers (1994) paper and an interview with French Prime Minister Édouard Balladur with Le Figaro one month earlier. In the paper written by Schäuble and Lamers (1994), the two German deputies of the ruling Christian democratic party, the authors advocated for the possibility of flexibility and multiple speeds and the strengthening of the EU core of France, Germany, and the Benelux countries (Schäuble & Lamers, 1994). Balladur called for a three-tier Europe of concentric circles, with the strongest states forming the central core. Together, Berlin and Paris were the most active in trying to introduce a flexibility clause. Britain, traditionally one of the most reluctant partners toward further European integration, saw increased flexibility as a welcome instrument for dealing with diversity. However, in a speech in Leiden on September 7, 1994, Major (1994) warned against an EU that had inner and outer circles in which “some would be more equal than others.”
The institutionalization of flexible integration proved difficult, however, because there were divergent views with regard to its purpose, form, and possible effect. While countries such as France and Germany saw it as a way to strengthen the European integration process from within, other more Euroskeptic countries were keen to see an institutionalization of the opt-out mechanism while fearing that it could turn into an instrument of exclusion. It was only during the Dutch presidency that concrete progress was made (McDonagh, 1998), leading to a compromise on a general flexibility clause in combination with specific provisions on the first (European Community) and third (Justice and Home Affairs) pillars. The general enabling clause specified the overall objectives, scope, and procedures of what it labeled as “closer cooperation” (Title VII, Arts. 43–45, TEU, 1992). It for instance specified that it would only be an instrument of last resort and concern at least a majority of member states and would not affect the acquis communautaire. It also embedded closer cooperation within the existing legal and institutional frame of the EU. In addition, there were also more specific provisions on the scope and decision-making rules in the first and third pillar. In the first pillar (Art. 11, TEC), the authorization to establish cooperation is given by the Council, acting by QMV on a proposal from the EC and after consulting the EP. There was, however, also an emergency break that gave member states the opportunity to refer the matter to the European Council for a decision by unanimity. In the third pillar (Art.40, Title VI, TEU, 1992), the opinion of the EC is nonbinding.
As there was a considerable concern that “closer cooperation” could be a potential instrument of excluding certain member states or negatively affect the uniform application of EU law, the conditions for applying closer cooperation were defined in a restrictive way. The result was that they were unworkable and required a revision under the next IGC leading to the Nice Treaty.
Evaluating and Understanding Amsterdam
When the dust on the Amsterdam negotiations had settled, the reactions were mixed. Those hoping for a major leap forward in the European integration process considered the results to be rather meagre (Duff, 1997; Sutherland, 1997). It is true that when successively examining the three core objectives, the results were far from optimal. Regarding the objective of bringing the EU closer to its citizens, the member states managed to make progress in important public dossiers such as employment, social policy, asylum and migration, and organized crime. In response to the complaints about the democratic legitimacy of the EU, the IGC further enhanced the powers of the EP. Codecision was further expanded (not to the field of agriculture), and the EP vote on the candidate for the position of commission president became legally binding. In addition, there was also an attempt to further activate the involvement of national parliaments, traditionally considered to be closer to their electorate than the EP was. In the field of foreign policy, the results were meagre. There was now some form of flexibility with the introduction of constructive abstention, and through the creation of the position of an HR, there was potential for more continuity and visibility in European foreign affairs. CFSP, however, remained a primarily intergovernmental policy, jealously guarded by the national capitals. The integration of the WEU into the EU would only be realized in Nice.
Institutional reform proved to be the most sensitive dossier. The extension of QMV remained limited to many fewer fields than originally expected. The reweighting of votes in the Council and size of the EC, which touched upon the question of the future influence of the member states, including that of a unified Germany, were adjourned to a later IGC. In light of the growing diversity, the debate on flexibility or “closer cooperation” was an important one. Under pressure from those fearing that it would be used as an instrument of exclusion, however, its scope was defined in such a restrictive way that in practice its application would become impossible. It is therefore not surprising that in Nice, the text would again be amended. Furthermore, the member states also agreed to clean up the existing treaty texts by removing obsolete provisions. It was decided that the articles would no longer be designated by letters but by numbers.
Beyond the question of the substance of the treaty and its place in the long-term development of European integration, a key question of academic interest is how we can make sense of the process and its outcome. In a joint article, Moravcsik and Nicolaïdis (1999), examining the treaty through liberal intergovernmentalist lenses, argue that the process and the outcome can best be understood as the result of negotiations among rational state actors whose preferences are a function of issue-specific interdependence. The central argument is that countries with an effective unilateral policy will be reticent to delegate sovereignty, while for those with a high level of economic or political-military interdependence, the willingness to cooperate will be high. In other words, the preferences of the member states will vary according to their needs in a specific policy field. In their view, this is well illustrated by, for instance, the discussions on migration and asylum in Amsterdam. Germany, which faces a high number of central and eastern European migrants, was eager to come to a European-level approach, while member states as Ireland and the United Kingdom, both outside the Schengen zone, were reluctant to communitarize these policy fields. Another example is the cleavages in the field of European security. Countries with close links to the United States, such as the United Kingdom, were not keen on the development of a European security identity, whereas France, which had more strained relations with Washington, DC, and which had left the integrated military command structures of NATO in 1966, strongly favored the integration of the WEU in the EU. Other countries concerned about the future U.S. commitment to European security supported that position.
Neofunctionalists and institutionalists disagree with this strong focus on the role of the big member states that are emphasized by the liberal intergovernmentalist school, and they point to the independent role of supranational institutions. The EC was present at the negotiations at all levels, but the college under the lead of Jacques Santer clearly played a less prominent role than did the Delors Commission during the Maastricht IGC (Gray, 2002). According to Moravcsik and Nicolaïdis (1999), there is no proof that their initiatives or compromise proposals affected the outcome of the negotiations substantially. What may be interesting to examine in more detail is the role of the Council general secretariat, referred to by Beach (2004) as the unseen hand in treaty-reform negotiations. The highly qualified civil servants of the secretariat were an important factor of continuity during the negotiations and played an important role in the drafting process both during the Irish and Dutch presidencies. It remains difficult, however, to determine to what extent they had a real effect, since they always operated under the supervision of the rotating presidency.
However interesting it is to explore the role of the national delegations and the EU institutions, such actor-centered approach can never tell the full story. The protagonists of the IGC operate in a broader institutional and (geo)political context. In their study on the constitutionalization of the EU, Christiansen and Reh (2009) point to the legal-institutional structure behind EU constitutionalization, the ideational-discursive context, and the temporal-political conditions such as crises and deadlines as factors that also matter. They plead for an inclusive approach that takes into account the interaction between both structure and actors. Their analysis also includes some examples of structural factors of particular importance for the Amsterdam negotiations. In the legal-institutional sphere, they for instance point to how the increased emphasis on openness and transparency made it more difficult for national delegations to change their position or show flexibility during the negotiations. A good example of how an ideational-discursive context can matter is how in its discourse the IGC was strongly linked to the upcoming enlargement toward the east. Provisions for fundamental rights, migration, the reform of the institutions, and flexible integration were all framed as issues that needed to be urgently addressed in light of the doubling of EU membership. When there was no deal on the composition and the EC and the weighting of votes dossier, it was decided that a further IGC was needed prior to the next enlargement. When it comes to temporal-political factors, one can point to the fact that from the start, it had been agreed upon that the negotiations would be concluded under the Dutch presidency in June 1997. This not only meant that there was pressure to come to a deal but also that none of the member states would make major concessions on strategically key issues before the European Council in Amsterdam.
Trying to understand the process and the outcome of the Amsterdam negotiations and processes of constitutionalization in the EU more broadly is a complex matter. The use of different theoretical lenses helps to unveil different parts of the story and the diversity in the theoretical approaches should therefore be welcomed.
