OJPCR: The Online Journal of Peace and Conflict Resolution

Issue 3.1 | March 2000

ISSN 1522-211X

www.trinstitute.org/ojpcr/

The Impossibility of Schengen: A Multi-Level Game Analysis of the State of Refugees and Asylum in the European Union

By Daniel W. Boettcher

The European Union has entered a new, enthusiastic era of integration. The Union seems to be both broadening and deepening. The new era of integration can be traced to the 1985 Jacques Delors white paper calling for the completion of the internal market. This white paper became the SEA, which allowed for the unprecedented enforcement of the movement of labor, capital, goods, and services across Western Europe. During the same period, five nations(1) signed an agreement at Schengen, Luxembourg that eliminated border controls among those nations and standardized entry procedures. More nations have joined the Schengen Accord since 1990, and its addition to the Amsterdam Revision of the Treaty on European Union has become a symbolic and tangible notion of the coming federal Europe.

The political geography of the world underwent unprecedented change with the final peace treaty of World War II reunifying Germany and the wane of Soviet influence in Eastern Europe and, in fact, in the Soviet Union itself. Most Eastern European states were able to reclaim their identity of the pre-War era and make new, bold steps toward self-determination. Unfortunately, Yugoslavia was not to follow this road. For many reasons, Yugoslavia splintered and entered into a period of bloody conflict that is not yet over.

The integrationists in Brussels did not foresee the conflict in the former Yugoslavia. They were ill equipped to deal with the onslaught of refugees who came pouring into the Union. These displaced people, combined with a flagging economy and an abundance of immigrants from the Maghreb, Maqreb, Turkey, the Middle East, and the Philippines, created a socially explosive problem in the traditionally homogeneous European cities.

The Eurocrats also realized that, as internal borders fell, the area of immigration and asylum law was not necessarily going to be in the domain of the Member State any longer. Realizing this, they wrote the Dublin Accord to harmonize, and in turn severely restrict, the availability of asylum across the Union. Furthermore, they attempted to transfer authority in this important area away from the Member States. Then, the Union entered a new, restrictive course in terms of refugee management.

In this paper I will explore the institutional dynamic of the European Institutions in relation to the movement toward a broad European refugee policy. In doing so, I will take a rather systematic walk through the state of policy starting with the 1951 Convention on the status of refugees through the contemporary period. Inside each time period, I will focus both on the socio-historical aspects of the era and on the legal tools that Europe had, or did not have. After this historical detail, I will take a hard look at the goings-on in the Amsterdam Treaty. My aim, in this paper, is to question the European Union's current asylum regime. Through this examination, I will use the multi-level game theory to show why the current regime should be dissolved and reconstituted. Then I will look at whether the Amsterdam treaty provides the necessary framework from which to launch a new, tabula rasa effort at forming a European Union asylum policy.

1. The 50s and 60s: Forming the Instruments

Europe had no common asylum policy before the mid-1980s. Any discussion of European asylum policy in the period after World War II but before Schengen is merely an exploration of common themes that ran across the continent.(2) It is, however, from these themes that the current system has developed. It is because of the impossibility of completely understanding the implications and concepts of European asylum policy at the turn of the millennium without knowing the roots of the policies that I will discuss the early conventions.

Every member of the European Union is a signatory of the 1951 Convention Relating to the Status of Refugees (Geneva Convention).(3) The Geneva Convention, principally born of the massive refugee movements of World War II, was undertaken to ensure that refugees and asylum seekers are able to escape danger through flight.(4) Perhaps the most important provision of the Geneva Convention is found in Article 33. Article 33 spells out the principle of nonrefoulement, which prohibits the expulsion or return of a refugee "in any manner whatsoever to the frontiers of territories where his life or freedom would be threatened on account of his race, religion, nationality, membership of a particular social group or political opinion" unless there is a threat to the public safety or security by the refugee's presence in the country to which he relocated.(5)

Aside from the principle of nonrefoulement, on which the piers of asylum laws are set, Article 1A(2) of the Geneva Convention defined 'refugee' as a term of art, separate from its quotidian definition. Under the Geneva Convention, a 'refugee' is a person who, "[a]s a result of events prior to 1 January 1951 and owing to well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social groups or political opinion, is outside the country of his nationality and is unable, or owing to such fear, is unwilling to avail himself of the protection of that country and return to it."(6)

The initial interpretation of 'refugee' was even more restrictive than the plain language of the Convention appeared to be. Aside from the limitation of events by date, the Convention left it up to the European nations to decide whether "event" referred to events in the entire world or only to events in Europe. Most nations chose the more restrictive interpretation of the term. Because of the post-War turmoil, it was thought that the Convention would be in place only for three years, at which point the issue of refugees in Europe would be finally settled. Sadly, this was not to be even remotely near the actual European experience.(7)

As it became clear that the three-year mandate for the Commissioner was not going to be adequate to serve the purposes of resolving the refugee issue, the European nations met again in Bellagio, Italy in 1966. Article I(1) of the Protocol Relating to the Status of Refugees (Bellagio Protocol) extended the time limitation, saying, "the term 'refugee' shall…mean any person within the definition of Article 1 of the Convention as if the words '…before 1 January 1951'…were omitted."(8)

On the ground, there were few systems existing for the protection of refugees and asylum seekers in the 1950s and 1960s. What did exist was largely done on an ad hoc basis with a bent toward granting asylum as often as possible. This period of liberalism was ignorant of the supranational European Community organs. In fact, each nation created and maintained rules concerning asylum that suited them. The post-War reconstruction left plenty of work to go around, so there was little public sentiment against the refugees.

The Treaty Establishing the European Economic Communities envisioned a Single Market devoid of internal frontiers, including for citizens.(9) The original framers of the Treaty Establishing the European Economic Community set forth standards of national treatment early in the treaty.(10) Article 6 of the original treaty prohibits "any discrimination on grounds of nationality" within the scope of the treaty.(11) Relating to the movement of persons, the Treaty of Rome, in Article 48, guaranteed the free movement of workers across the Single Market and Article 48(3) allows those seeking work to establish themselves in any Member State for the purpose of searching for employment.(12) Although this Article has been much litigated, the concept of a Community definition of worker eventually emerged.(13) Article 52 of the Treaty of Rome created the right of establishment for the self-employed in any Member State in the Union under conditions of national treatment.(14) Article 52 seeks to prohibit discrimination on grounds of nationality resulting from national laws, regulations, or practices.(15)

Although all the members of the Community had signed the Geneva convention by 1960, the interpretation and administration of the asylum system was left up to national sovereignties with a few guidelines from the EEC treaty. Generally speaking, the protections set out in the international treaties were never seen as bright line obligations, but rather as being open to legal interpretation. The early period of asylum was thus characterized as one of coalescing state-based regimes.

2. Rumblings of Cooperation: TREVI

The period from the signature of the Geneva Convention by the Member States to the beginning years of the Terrorism, Radicalism, Extremism, and Violence International (TREVI) structure of the European Political Cooperation (EPC) has been described as "uncoordinated liberalism."(16) TREVI, although primarily aimed at fighting terrorism, became the prototype organization for the Third Pillar of TEU in 1993.(17)

Originally created in 1975, the TREVI group sought to increase practical cooperation in terms of security, policing, and border control. Interior ministers of the participating Member States, who met twice yearly, staffed the organization. It had no secretariat for its entire existence.(18) The resulting culture of wining and dining prevented it from ever becoming a real institution. Its most significant work was in its service, albeit dysfunctional, as the training ground for the future officials in the Community's K.4 structure.(19)

3. The Single European Act and Schengen

In 1986, Jacques Delors, as President of the Commission, disapproved of the period of atrophy he perceived in the European Economic Community. In a White Paper on the completion of the internal market, Delors called for nothing less than a re-launching of the Community.(20) The embodiment of these ideals is seen in the politics of the SEA (SEA) and the Single European Market (SEM) program.(21) At the end of this program, the Community found fortifications of the original 1958 plan for the creation of the Common Market and also was positioned to dismantle the system of internal borders in the Community.(22)

It was this new integrationist period, marked by the SEA and the reduction of the strength of borders inside the Community, that necessitated the coordination of asylum policies. The policies of the era are linked to the deepening of the Community and the entrenchment of the Union's authority over many new areas of competence. In order to complete the market for labor, it was argued, labor must be able to flow across the Community to the areas where it can be employed most efficiently. Economically, this was thought to bring about the leveling of wages across the community and the most rational and productive utilization of the workforce.(23)

Article 7a of the SEA confirms in law the desire to bring the EEC Treaty past the realm of economics and more in line with the triptych of widening, deepening, and completing the Internal market. According to Article 7a, "[t]he Community shall adopt measures with the aim of progressively establishing the internal market over a period expiring on 31 December 1992 …The internal market shall comprise an area without internal frontiers in which the free movement of goods, persons, services and capital is ensured in accordance with the provisions of the treaty."(24)

The main place of Article 7a in the European Union Acquis is its reinforcement of the Union's Founders contemplation of a Single Market and its validation of the Commission's 1985 White Paper, which, among other things, called for the free movement of people in Europe. Although this is seen as the further ratcheting down of the internal walls, none of the Union's provisions can be expected to have direct effect on any third-country nationals, who are the likely people seeking asylum in the European Union.(25) In reality, though, it is unlikely that it would be possible to enforce the allowance of some people in Europe to cross borders while others cannot. However, the ECJ has ruled that the police can check the identity of people crossing their borders, so long as they do not inquire as to the business and financial means of travelers for the purpose of determining their fitness to enter the country.(26)

Ultimately, the SEA stands as a transitional point in the development of the Union. Here, the Community took the step from guaranteeing the right to establishment for workers and the self-employed to granting the free flow of persons inside the Union.

4. Schengen and Dublin: Protecting the Union's Flank

The Community scrambled to institute flanking mechanisms to enforce the external frontiers communally while the internal borders disintegrated under the SEA. The Schengen Accord, which was signed intergovernmentally outside the machinery of the European Communities in 1985, was the principal agreement used to eliminate internal borders.(27) However, the Member States were not able to quickly agree on limiting their sovereignty in an area as sensitive as border controls, so this proceeded slowly. In fact, Schengen would hang over the Community, and then the Union, until the present day, compounded by the instability of the third pillar instruments, which are the result of intergovernmental cooperation governed by public international law, rather than by European Union law.(28)

The themes of Schengen are quite simple: In order to create the internal market while ensuring security against criminals and terrorists, the common external borders of the Community must be enforced while all vestiges of border controls among signatories disappear. Although the concept may seem simple enough, in Schengen the devil proved to be in the details.

The original 1985 Schengen Agreement, which is the conventional founding of the 1990 treaty, was allowed for signature without reservation among Belgium, France, Luxembourg, and the Netherlands. Because the Netherlands is the only country among the four that requires legislative approval of treaties, the Agreement stayed quiet and known only to small groups in the nations. In 1989, the French press divulged le grand secret de Schengen, startling even the French Ministry of Justice, which had never been informed by the Ministry of Foreign Affairs of its existence.(29)

The 1990 Schengen Agreement, designed to implement the 1985 agreement and also signed at Schengen, is a complete treaty of 142 articles.(30) Of these, eleven articles are explicitly under the heading "Responsibility for examining applications for asylum." These parts of the agreement deal with Police and Security, the Abolition of Controls at the Internal Borders and Movement of Persons, the creation of a Schengen Information System to track the movement of people, the Transport and Traffic of Goods, the Protection of Personal Data, and the creation of an Executive Committee.

The Dublin Convention is a second flanking measure undertaken by certain Member States in 1990, under the twelve minus one compromise, in order to secure the European Union's external borders.(31) As the interior borders came down, the Member States perceived a threat arising from the lack of security on their frontiers and also realized that, unless common external policies were enacted, a refugee would have the ability to apply in every Member State for asylum in the Union. Also, the Member States wished to end the problem of "refugees in orbit" because many refugees do not apply for asylum in the country that is their entry point into the Union. Dublin, then, guarantees that asylum seekers will have their application addressed, but only one time by one Member State.(32)

Justice and Home Affairs Council Decisions 1/97 and 2/97 fill gaps and provide details to the Dublin Convention. Decision 1/97 is important because it fills out the rules of the Dublin Convention by defining in which Member State an asylum seeker will appear to lodge his application. Decision 1/97 imposes on Member States duties to take charge of asylum applications and duties to take back asylum seekers in certain instances. In an attempt to organize and clarify the Dublin mandates, Decision 1/97 also offers general provisions on evidence and provides standard forms for determining the Member State responsible for the asylum application. Decision 2/97 provides the mechanics for an executive committee envisaged under Dublin's article 18.

The Dublin Convention has been attacked on numerous grounds since its inception. Often the commentary is focused on the inefficacy of its provisions and the lack of judicial control. However, some very powerful criticism has come against its conflicts with the Geneva Protocol, which states that it is incumbent on each state to determine autonomously whether an applicant for asylum is a refugee, and other human rights principles.(33)

5. The Maastricht Revision of the Treaties

Although the SEA was an attempt to bring the EEC beyond economics, it was still fashioned by economists. The economists did not predict the social implications of the completion of the Internal Market in the late-1980s. Nor were they able to predict the way in which those implications would be magnified if the Community fell into economic recession.(34) Furthermore, they could only have been able to dream about the post-Cold War world. Whether for better or worse, all these things did in fact happen upon the completion of the internal market. Europe's generous social welfare systems became taxed and jobs for Europeans became scarce. The political machine had not predicted the ability to deal with this crisis at the same time that citizens of the Warsaw Pact nations were doing everything in their power to get to the West, where they perceived their new fortunes to lie.

The public did not exactly greet all the curious and economically poor migrants with open arms. Although there was a general sense of euphoria over the victory and ending of the Cold War, most Europeans did not feel able to support the nations of the East. Furthermore, it became increasingly clear that the Community's largest economy, Germany, was going to be occupied for years to come with the reunification and redevelopment of the five Eastern Länder.

Although the average European felt the introduction of the SEA and noticed the completion of the internal market keenly, the Brussels elite was disappointed with what was done. There was a sense that more needed to be done in order to sufficiently advance the cause of integration.(35) As the policies of the SEA and SEM took effect in the Community, leaders began to realize the full benefit of these policies would not be felt without an Economic and Monetary Union (EMU).(36) Also on the minds of the leaders was the need to create a social dimension to the Community, so that the impacts of the open markets on individuals would be softened. Also, it was felt that there should be measures in place to prevent 'social dumping' or the location of business in the areas with the lowest and least expensive social standards.(37) In this same vein, demands to cure the Union's so-called democratic deficit were made.(38) Finally, the issue of security in the Community and the need to eliminate internal borders to complete the internal market had fermented to the point where the Member States knew there must be a Community-level scheme to enforce the external borders and deal with international crime, terrorism, drug trafficking, and the 'threat' of mass migrations from Eastern Europe, the Middle East, and North Africa.(39)

In 1992, the members of the European Community met in Maastricht, the Netherlands to set about the creation of a deeper confederation. At the end of the Maastricht Intergovernmental Conference (IGC), the European Community became the European Union under a broad revision of the Treaty of Rome. The Maastricht Treaty, as the instrument has become commonly called, gave Brussels authority in places it never had before and formalized ties with other organizations that had existed in Europe since the post-War period, but had never been included in the European Community sphere of competence.(40)

The foundation of the Maastricht treaty is the principle that the Community could be expanded into a Union of three distinct areas. The three pillars of the Union were set around major tracts of governmental competencies. The first pillar, crafted from the old Treaty of Rome, is made up of the Economic Community and its organizations, including the European Coal and Steel Community, the European Atomic Energy Commission, and the European Economic Community. The second pillar is the Common Foreign and Security Policy. This pillar is much less organized and is a loose confederation of international actors like NATO, the West European Union, and Europol. Also, the Eurocorps and the joint Dutch-Belgo Naval fleet find their authority under the second pillar of Maastricht. The third pillar, under Title IV of the Treaty, is known as Justice and Home Affairs.

The competence of the third pillar of Maastricht was barely specified, so an awkward system of intergovernmental action endorsed by the supranational European Union was created. In the realm of immigration law, EC Articles 100c and 100d specify that a common list of countries whose nationals need visas be established.(41) Ultimately, cooperation under the third pillar was intergovernmental.(42)

6. The Road to Amsterdam

The IGC that opened in Turin in 1996 was a long planned-for event.(43) A raft of leftover issues from the Treaty of Maastricht formed the general purpose of the IGC. Officially, the IGC was to revise the European Treaties "with the aim of ensuring the effectiveness of the mechanisms and the institutions of the Community." However, the governments realized the implications of reopening some of the matters and chose instead to focus officially on five points.(44) This official guide allowed the IGC to progress without touching the third-rail policies of EMU.(45)

At Turin, the European Council, under Italian politician Lamberto Dini, set the agenda around three realistic themes. First, a "Union closer to its citizens" was called for. Second, the theme of efficient and democratic institutions in the context of enlargement was introduced. Third, the IGC sought to develop critical mass in terms of the Union's ability to act externally in economic relations (under the first pillar or the EC) and to exercise more control over foreign and security policy (under the second pillar).

With relation to asylum policy, the first point of the IGC squarely demanded "adequate results" in one year's time on "developing coherent and effective visa, asylum and immigration policies." Other items in the first point are personal rights, including fighting international criminal syndicates, improving the readability of the treaties, and resolving the jurisdiction conflicts in the third pillar of the Maastricht treaty. It is impossible to ignore the tone that was set, especially in comparison to the Geneva Protocol and Bellagio Convention. The third pillar is Article IV and relates to Justice and Home Affairs.

The Union understood and wished to implement a common asylum policy. The reasons to do so seemed compelling and there was no denying the waves of refugees from the former Yugoslavia. It became clear, however, that Maastricht did not give the institutional tools necessary to truly complete the free movement of citizens. Furthermore, these institutions had been marginalized in respect to the third pillar of the Union, on which Maastricht mandated intergovernmental agreements exclusively. In ways, the Justice and Home Affairs pillar resembled the European Economic Community in the period before the Luxembourg Compromise.(46)

7. Adjusting to Amsterdam

The Amsterdam Treaty does no less than mandate the development of a common asylum policy among the European Union Member States.(47) Like any governing document, it does little to explain how this should happen. Since the treaty has been in effect, the Union has been fumbling inside the institutions and rules allowed by the treaty to unify asylum policy.

Title IIIa of the treaty sets forth the Union's policies toward the free movement of persons, and is considered an itemization of the key points of the Schengen Agreement. The first Article under title IIIa, Article 73i, establishes the goal of "an area of freedom, security and justice," including, in 73i(a), "measures aimed at the free movement of persons…in conjunction with directly related flanking measures with respect to external borders."(48) Article 73i(b) orders the union to take "other measures in fields of asylum, immigration and safeguarding the rights of third country nationals."(49)

These provisions become much more powerful when taken with the new Article 73k, which directs the Council to, within a period of five years after the entry in force of the Treaty of Amsterdam, adopt:

The treaty's mandate that the Union must establish these guidelines by 2003 is clear. Article 73o contains the procedure to be used to develop the measures specified in Article 73k. According to Article 73o(1), within the first five years of the Treaty, the Council must unanimously act on a recommendation from the Commission or a Member State after consulting with the European Parliament. After the period of five years, the procedure for developing Article 73k changes. According to Article 73o(2)(¶1), the Commission would then submit its proposals to the Council and the Member States would submit their proposals to the Commission. Creatively, Article 73o(2)(¶2) of the Treaty gives the Council the option of abandoning the unanimity of Article 73o(1) and 73o(2)(¶1) in favor of the Article 189b codecision procedure by a unanimous vote in consultation with the European Parliament. Since Article 73o spells out the decision-making framework for the whole of Title IIIa, the Council would be adopting the Article 189b procedure for all of the policies relating to visas, asylum, and the free movement of persons. Also, the Council can unanimously vote, in consultation with the European Parliament, to extend jurisdiction to the ECJ over all or part of Article 73o.

Amsterdam incorporates the Schengen Acquis into the European Union treaties for the first time.(53) The Schengen Acquis is specifically found in Articles J.7(1) and J.7(3) of the Treaty on European Union. The Schengen Protocol, which bifurcates the Schengen Acquis into its treaties and decisions, is annexed to the Treaty on European Union and the Treaty establishing the European Community as the incorporating instrument. Furthermore, Article 6 of the Schengen Protocol permits the unanimous Council to conclude agreements allowing the Republic of Iceland and the Kingdom of Norway to participate in the implementation of the Schengen Acquis. By incorporating Schengen on a community level, it allows the Union to control the external borders from a supranational position rather than an intergovernmental position. It is impossible that the institutionalization of Schengen will have any less than a dramatic impact on refugees and asylum seekers.

The Schengen Protocol gives the creation of the legal bases for the Schengen Acquis to the Council of the European Communities. Article 2(1)(¶2) creates a power in the Council when it allows the unanimous Council to "take any measure necessary for the implementation of [the Schengen Acquis]." Article 2(1) provides that the unanimous Council determines the exact legal basis for the Acquis.

Pursuant to the Schengen Protocol article 2(1), the Council issued its unanimous decision 1999/436/EC. That decision discussed the applicability of the Schengen Acquis to the European Union and set out the legal basis for the provisions of the Schengen Acquis. Because the Schengen Acquis is actually made up of many separate agreements, the Council has developed different legal bases concerning the Schengen Acquis, depending on the Member State at issue.(54)

The portions of the Schengen Acquis that are not given legal basis by the Commission become part of Title IV of the Treaty, with normal ECJ jurisdictional rules except as specified by Article 73p(2). The Council Decision 1999/436/EC allows the Council to take steps to adopt judicial and civil cooperation policies in this area.(55)

The Amsterdam treaty significantly increases the jurisdiction of the Court of Justice of the European Communities over the third pillar. In fact, normal jurisdiction rules apply to the area, with several exceptions. However, an important departure from the Court of Justice's traditional competencies in resolving other Treaty of Rome matters is that, in resolving third pillar matters, under article 73p(1), the obligation to request preliminary rulings under article 177 is maintained only for courts and tribunals against whose decisions there is no remedy.(56) The faculty for any other national body to use article 177 over the third pillar is abolished. This is dangerous in that it expects lower national courts and tribunals to understand the correct application of difficult tracts of European Union law. This may greatly reduce the safeguard of the article 177 proceeding.(57)

More important, however, is the carrying forward of Article 100c from the Maastricht Treaty, giving Member States the ability to trump the supremacy of European Union law.(58) Article 73p(2) provides generally that the ECJ has no jurisdiction on any measure or decision taken pursuant to Art 73j(1)(59) relating to the maintenance of law and order and to the safeguarding of internal security.(60) This is especially troubling when taken in conjunction with article K.7 on police cooperation, which gives much more authority to the state to enact police cooperative measures than it does to protect citizens from over-zealous enforcement.

To counteract the limitations placed on Article 177 reference proceedings, the Amsterdam Treaty includes a reference "in the interest of law."(61) Article 73p(3) allows the Council of the European Communities, the European Commission, and the Member States to request a ruling on an interpretation of the new title IIIa or on any of its institutions. These rulings are limited in that they cannot apply to the judgments of national courts or to any judgments previously judged in entirety.

8. Theoretical considerations of Amsterdam

a. Multi-level game theory

A recent move in the analysis of international law has been the use of international relations scholarship to help explain and understand the law. The use of international relations theory has allowed the international legal community to better understand the influences on law and has helped to offer concrete suggestions so that the law may be ameliorated.(62) This ability of international lawyers, to diagnose problems in the law and attempt to solve them, has become a powerful and popular tool.(63)

A useful scholarly model for analyzing the effects of different tiers of politics on treaty making is the multi-level game. The multi-level game was put forward when Glenn Snyder challenged the notion that the states in international negotiations act in a unified way.(64) Snyder noticed that nearly half the time top decision-makers are not unified. Robert Putnam built on Snyder's scholarship when he offered the two-level game as a model for understanding the phenomenon that Snyder had described.(65) The two-level game allows the researcher to analyze the impact of domestic politics on international negotiations. Ultimately, it allows the international negotiation process to be better understood by revealing the impact the domestic sphere has on the international organization.

Traditionally, Level I is seen as the negotiations that take place in the international organization among countries. Early theorists saw this merely in terms of the simple Nash game theory, but that notion was dispelled by Snyder's work. Putnam's scholarship introduced a second level to the analysis--that of the politics inside each nation-state.(66) Putnam's two-level game centers around the range of agreements for Level I, the international set, that are acceptable to a majority at Level II, the domestic constituency. This range is known as the win-set for the particular country on which the analysis is focused.(67) By analyzing the international negotiations from the perspective of one country through win-sets, Putnam argued that it is possible to estimate the impact of the domestic politics on the success of the international negotiation. From this model, then, Putnam hypothesized that larger win-sets make Level I agreement more likely and, conversely, the smaller the win-set the more likely the negotiations will break down. Also, the relative size of the Level II win-sets will affect the distribution of the joint gains at Level I. A smaller win-set at Level II can be a bargaining advantage for a country at Level I.(68) Along the same lines, the successful international agreement must fall inside the Level II win-set. The larger the Level II win-set, the more likely there will be a successful agreement reached. However, the larger Level II win-set increases the likelihood that the Level I negotiator will be challenged by other countries.(69)

Further work by Keisuke Iida expanded this model and challenged the notion that the size of the win-sets is important to the negotiators.(70) Iida claimed that the negotiations would take place only if there is an adequate amount of information from the outset that shows there to be an intersection of win-sets. Failure in international negotiations is caused then not by the misalignment of win-sets, but rather by involuntary defection precipitated by an incomplete understanding of the Level II political situation. Iida argued that the win-set hypothesis is correct only in that it describes what happens within international negotiations where there is some lack of information and a misunderstanding of the Level II politics and win-set.(71)

Jongryn Mo expanded the two-level game analysis by looking at the interaction of the international process and domestic coalitions.(72) Mo looked at the economic effects of coalition veto power and internal side-payments. Of particular interest to this paper is the veto power, which limits the Level II win-set. This limitation of the win-set may make it no longer possible for another Level I actor to negotiate the least costly win-set in relation to its domestic actors because the range of win-set possibilities becomes limited to those acceptable to the veto-bearing group.

b. Application of multi-level game theory to the problem

The question of European Union asylum law cannot be answered in one tranche. Typically, the negotiation of international agreements is easily boiled out of the multi-level game. On one side sit the national negotiators, on the other side sit their respective citizens' politics. Evaluating the win-sets of each individual country and matching them will tell the researcher what sort of outcome to expect from a negotiation. The question of European Asylum policy falls not in one game, but two. To complicate matters further neither game is on two levels. Rather, the games operate on three levels. The first game I will call the Schengen Game and the second I will call the Amsterdam Game.

The Schengen Game involves the states that negotiated the Schengen Agreement in 1985 and 1990. Those states, all of which are current European Union members, entered this purely intergovernmental exchange. The domestic win-sets were set on dealing with asylum policy in an international realm, and their confederation was mostly geographic. So, Level I is the nations negotiating the Schengen Agreement and Level II is the national politics in each state. There is a third level that comes in the Schengen Game, and that is the level at which the Schengen Acquis negotiates its terms with the European Union Council. I will call this Level I'. Thus, another win-set must be contemplated in the negotiation: first is the Level II win-set, which drives the Level I negotiations, second is the Level I' win-set, which limits the acceptable win-set at Level I. It also introduces another veto-bearing organ to the process, which ultimately constricts the possible outcomes of the agreement and likely constricts the win-sets to be only those which are acceptable to the Council of the European Union.

The Amsterdam Game is a concurrent game with almost all the same players. The Level I body is the Council of the European Union, where the Level II politics are announced and the corresponding win-set is determined. The third level in the Amsterdam Game is the Schengen Acquis, which stands above the Council of the European Union and consists of a ready-made formula of policies and decision. The inflexibility at this Level I' focuses the win-set possibilities into only those that are within the Schengen win-set.

The flow charts of the games are reasonable enough:

Schengen Game:

Amsterdam Game:

When the Schengen Game and the Amsterdam Game are laid one on the other, a graphical representation would appear thus:

When looking at the combined flowchart, the problem with competing loyalties becomes obvious. From the national perspective of the multi-level game, countries can be seen having different constraints to identical obligations.

To provide concrete examples to the dilemma, it is helpful to walk through the situation of a particular nation. Inside the European Union, the United Kingdom's domestic political constraint requires the Level I negotiators (in this game, the EU ministers) to want the most influence possible over the institutions. The United Kingdom's Level II win-set includes a veto over dramatic "continentalization," but allows the government to be active in the creation and guiding of the institutions so that the policies are not too offensive to the United Kingdom's less integrationist Level II politics. This often gives the UK a strong bargaining position and allows it to defuse issues that would be domestic deal-breakers. Thus, the UK is very involved in the European Union governance.

In the Schengen Game, the UK has attempted to scuttle the accord. The insular segment of the Level II politics basically prevents the Level I negotiators (in this game, the Schengen and Dublin negotiators) from allowing unfettered access to the UK from the Continent. The Level II veto is so strong that it has become a deal-breaker for the United Kingdom, which never became a party to Schengen. In keeping with Iida's scholarship, the UK and the other nations in the Schengen discussions knew that the UK could not become a member of the agreement. Because of this, the UK's position in the negotiations was limited and the other Level I negotiators did not let the UK have the power to scuttle the talks. The informational exchange was clear. In fact, the UK was not a party to the agreement at all.

Now that the Schengen Acquis is part of the Treaty of Amsterdam, it is wholly unsustainable for the UK, which is a Member State and net fiscal contributor to the EU, to attack collaterally the same EU policies it wishes to have a hand in guiding. By the incorporation of the Schengen Acquis into the Treaty of Amsterdam, the negotiators in fact created an unsustainable position.

1. Flow-chart of UK-Schengen win-set. The result is no deal.

2. Flow-chart of the UK-EU win-set. The win-set is inside the brackets.

3. Flow-chart of the EU-Schengen win-set with UK veto of EU policy overlay. The boxed-in area is precluded by the UK veto as shown in chart 2.

Combination of charts 1 and 3. Notice the competing UK win-sets for Schengen and the EU.

As we can see by the charts, the UK has two separate win-sets for the EU and for Schengen. Yet now Schengen is part of the EU by treaty. Since there is no clear domestic political will in relation to the EU after Schengen becomes part of the EU, the negotiators for the UK will find themselves deadlocked. By this, any attempt at creating any policy will be thwarted. It is clear that the UK has chosen to reject the Schengen Acquis, but it has not chosen to reject the EU. This is a dangerous political situation that creates win-sets that are actually in opposition to one another.

9. Conclusions

The attempt to construct a common European Union asylum policy has progressed with fits and starts over the past fifty years. Unfortunately, the current course is doomed. The European Union's attempt at creating asylum policy by introducing an international instrument foreign to the organization is not only doomed to failure, it is dangerous to the European Union itself. The conflict that is set up by the nonendogenous Schengen Treaty could threaten the stability of the European Union as nations choose to cling either to it or to the European Union's historic treaties.

The solution to the problem is simple. Schengen must be abandoned. The common asylum policy must be born inside the EU and it must be amenable to all parties. Until that happens, the schizoid policy that is now in place will continue. The creation of a system of standards built with the tools the EU already possesses, rather than trying to fabricate a system of harmonization from outside, is needed. This would restore the balance of the multi-level game and of the Level II politics inside the game.

Also, the development of standards coming from inside the EU itself will be governed from inside. Currently, the ECJ has no jurisdiction over Schengen; this is reserved for the national courts under a domestic security exemption. Furthermore, the Council can only dictate how Schengen fits inside the EU according to the treaty. It cannot shape the policy to make it a strong pan-European policy. A good starting point for the European Union on this process would be for it to sign the 1951 Geneva Convention and build its asylum policy framework on that, rather than seeking to build one that sits on the backbone of a 1980s open market agenda. The Amsterdam Treaty provides the mechanism to create a true common policy; in fact it commands it. The use of the Schengen Acquis, a foreign instrument, is unnecessary and destructive. In the future, the jurisdiction of the ECJ should be expanded to maintain the policy.

Ultimately, Dini reminds us that "our disappointment must not leave us without a rational and farsighted strategy."(73) Reiterating what German Enlightenment philosopher Moses Mendelsshon knew before, Dini reminds us that throughout Europe's history, even in the darkest moments, there has always been enough light to illuminate the next step. One hopes that for this the resolve of Europe glows brightly enough.

1. The Benelux Economic Union, France, and Germany were the original signatories of the Schengen Agreement.

2. Danièle Joly, Haven or Hell? Asylum Policies and Refugees in Europe 44 (1996).

3. Convention Relating to the Status of Refugees, July 29, 1951, 19 U.S.T. 6259, 606 U.N.T.S. 267 [hereinafter Geneva Convention].

4. Historically, the Geneva Convention is not the first attempt to deal with the problem of refugees. The entire compendium of instruments relating to the status of the refugee has its roots in the events immediately following World War I. Of the millions of war refugees, compounded by the millions of Russians fleeing their own October Revolution, nearly all found themselves in destitute and difficult situations. With a loss of de jure protection from their governments, they were unable to cross international frontiers (which had become newly fortified with the nascence of the modern states) to improve their lots. The League of Nations, in response to this problem, issued "Arrangements for Certificates of Identity" to the refugees so long as they were no longer in their home state.

In the run up to World War II, the Europeans also tried to establish instruments for the treatment of refugees seeking asylum. For example, the Evian Conference in 1938 attempted to solve the problem of refugees from Hitler's National Socialist government. It was the utter failure of these conventions to help those fleeing from genocide that caused the global community to take another long, hard look at the refugee policies. The 1948 Universal Declaration of Human Rights includes, in article 14(1), the right to seek and enjoy asylum in other countries. It was in the spirit of these, and other, post-War efforts that the Geneva Convention was signed.

5. Geneva Convention, supra note 3, at 6276.

6. Geneva Convention, supra note 3, at 6276.

7. See Danièle Joli et al., Refugees in Europe: The Hostile New Agenda 7 (1997).

8. Protocol Relating to the Status of Refugees, Jan. 31, 1967, 19 U.S.T. 6224, T.I.A.S. No. 6577 (1968) [hereinafter Bellagio Protocol].

9. Treaty Establishing the European Economic Community, Mar. 25, 1957, 298 U.N.T.S. 11 [Hereinafter Treaty of Rome].

10. See generally D Martin et al., Free Movement of Persons in the European Union (1996).

11. Article 6 EEC Treaty

12. Article 48 EEC Treaty states
  1. "Freedom of movement for workers shall be secured within the Community by the end of the transitional period at the latest.
  2. Such freedom of movement shall entail the abolition of any discrimination based on nationality between workers of the Member State as regards employment, remuneration and other conditions of work and employment.
  3. It shall entail the right, subject to limitations justified on grounds of public policy, public security or public health:
4. The provisions of this Article shall not apply to employment in the public service."

13. Lawrie-Blum: 66/85 [1986] ECR 2583, [1987] 3 cmlr 389 (1985).

14. Article 52 states "Within the framework of the provisions set out below, restrictions on the freedom of establishment of nationals of a Member State in the territory of another Member State shall be abolished by progressive stags in the course of the transitional period. . . . Freedom of establishment shall include the right to take up and pursue activities such as self-employed persons and to set up and manage undertakings . . . under the conditions laid down for own nationals by the law of the country where such establishment is effects. . ." Treaty Establishing the European Economic Community, Art. 52.

15. See Commission v. Luxembourg (childbirth and maternity allowances) 1993 I -817.

16. Danièle Joly, Haven or Hell? Asylum Policies and Refugees in Europe 46 (1996).

17. Carl Levy, European asylum and refugee policy after the Treaty of Amsterdam: the Birth of a new regime?, in Refugees, Citizenship and Social Policy in Europe 12, 23 (1999).

18. Levy paper at 23.

19. M. Den Boer, Justice and Home Affairs: Cooperation without Integration, in Policy-Making in the European Union 389, 395 (Wallace et al. Eds. 1996).

20. See generally Charles Grant, Delors: inside the house that Jacques built (1993).

21. Neill Nugent, The Government and Politics of the European Union 57 (3d ed. 1994).

22. Treaty Establishing the European Economic Community, Mar. 25, 1957, 298 U.N.T.S. 11 amended by Single European Act, 1987 (O.J. L 169/1). 2 C.M.L.R. 741 (1987) [hereinafter SEA]. See Pierre Mercier et al., La libre circulation des personnes et des services 43 (1991).

23. Molle at 201-222.

24. SEA, supra note11, art. 7A.

25. Cf. Meade: 238/83 [1984] ECR 2631 (provisions of the original treaty apply only to nationals of the Member States of the treaties).

26. Case 321/87, Commission v. Belgium (Border Controls) 1989 E.C.R. 997.

27. In this paper, the statements from the 1985 Schengen Agreement are the author's translations, as only the Dutch, German, and French texts are authentic. I have used the Dutch text, which can be found in Tractenblad [Netherlands Treaty News Series] 1985, No.102.

28. Muller-Graff, The Legal Bases of the Third Pillar and its Position in the Framework of the Union Treaty, 31 Common Mkt. L. Rev. 493, 495 (1994).

29. Le grand secret de Schengen, Le Figaro, May 14, 1989, at 1. See also, Valeurs Actuelles, May 16, 1989, at 2 (offering a corporate perspective on the removal of border crossing checkpoints).

30. Agreement for the Execution of the Agreement Concerning the Gradual Abolition of Controls at the Common Borders Concluded Between the Governments of the States of the Benelux Economic Union, the Federal Republic of Germany and the French Republic at Schengen on 14 June 1985, June 19, 1990, Tractenblad 1990, No. 145. Only the French, German, and Dutch texts are authentic; all three are published in the Tractenblad.

31. Convention determining the State responsible for examining application for asylum lodged in one of the Member States of the European Communities [hereinafter the Dublin Convention], June 15, 1990, 1997 O.J. (C 254) 1; 30 I.L.M. 425 (1991).

32. Neuman, Buffer Zones Against Refugees: Dublin, Schengen, and the German Asylum Amendment, 33 Va. J. Int'l L. 503, n. 23 (1993).

33. See Note, Implementing the Dublin Convention on Asylum, 4 Colum. J. Eur. L. 182, 191 (1998).

34. For evidence of the inability to see the modern consequences of the completion of the internal market, see generally E. Cerexhe, Le droit européen. La libre circulation des personnes et des entreprises (1982).

35. Nugent 57

36. For an excellent overview of the pre-Maastricht debate over the Economic and Monetary Union, which is outside the scope of this paper, see Willem Molle, The Economics of European Integration (1990).

37. Eleftherios Botsas, The Socioeconomic Structure of Europe, in The Impact of European Integration 39, 50-52 (George A. Kourvetaris et al. Eds. 1996).

38. The democratic deficit in the European Union refers to the lack of direct participation that European citizens have in the European Institutions. While the people do elect their Members of the European Parliament (MEPs), the European Parliament has never been allowed to do much more beyond comment on most legislation and amend certain types of legislation. The governments of the Member States appoint all the members of the Commission, where all legislation begins, and of the Council of the European Communities, which irons out the details of legislation. Juliet Lodge, EC policymaking: institutional dynamics, in The European Community and the Challenge of the Future 1, 22 (Juliet Lodge Ed. 1993).

39. Nugent at 58. See also Kostas Messas, Return to Decadent Europe, in The Impact of European Integration 299 (George A. Kourvetaris et al. Eds. 1996).

40. Treaty on European Union, Feb. 7, 1992, 1992 (O.J. C 224/1), 1 C.M.L.R. 719 (1992) [hereinafter Maastricht Treaty] amending Treaty Establishing the European Economic Community, Mar. 25, 1957, 298 U.N.T.S. 11.

41. See Ole Due, The Impact of the Amsterdam Treaty upon the Court of Justice, 22 Fordham Int'l L.J. S48 (1999).

42. 1 CMLR at 96 (provisions on JHA) FN 7

43. Piris, The Amsterdam Treaty: Overview and Institutional Aspects, 22 Fordham Int'l L.J. S32 (1999).

44. Andrew Duff, The Treaty of Amsterdam Text and Commentary xxx (1999).

45. The precarious state of the Economic and Monetary Union, especially in relation to EMU members in economic recession, made any discussion on the topic likely to collapse the institution.

46. The Luxembourg Compromise is that which allows for qualified majority voting rather than consensus on certain issues.

47. Treaty of Amsterdam amending the Treaty on European Union, the Treaties establishing the European Communities and certain related acts, Oct. 2, 1997, O.J. 1997 (C 340/1), 37 I.L.M. 56 (1998) [herein after Amsterdam Treaty].

48. Amsterdam Treaty Article 73i(a).

49. Amsterdam Treaty Article 73i(b)

50. See part 2, supra, for discussion of the Geneva Convention. Specifically, see note # 3.

51. See part 2, supra, for discussion of the Bellagio Protocol. See note #8.

52. Amsterdam Treaty Article 73k.

53. The Schengen Acquis is made up of the original agreement, the supplementary agreements signed by the other Member States except Ireland and the United Kingdom, and the appurtenant bureaucratic directives and national and European court cases. The entire Acquis is over 3000 pages. It is not to be confused with the Schengen Agreement, which is the original 1985 intergovernmental agreement, and the Schengen Protocol, which is the instrument that incorporates the Acquis into the Treaties. The entire text of the Schengen Acquis is spelled out in the Annex to the Schengen Protocol to the Treaty on European Union.

54. Council Decision 1999/436/EC of 20 May 199 Determining the Legal Basis for Each of the Provisions or Decisions which Constitute the Schengen Acquis, art. 2-3, 1999 O.J. (L 176) 17, 18.

55. Amsterdam Treaty art. 2(15), 1997 O.J. (C 340/1) at 31 (inserting art. 73p into the EC Treaty).

56. Amsterdam Treaty art. 73p(1).

57. Ole Due, supra note 57.

58. Maastricht Treaty art. 100c.

59. Amsterdam Treaty art. 73(j)1 governs the movement across borders.

60. Amsterdam Treaty art. 73p(2).

61. Amsterdam Treaty art. 73p(3).

62. Slaughter et. al., International Law and International Relations Theory: A New Generation of Interdisciplinary Scholarship, 92 Am. J. Int'l L. 367, 368-371 (1998).

63. Id. at 367.

64. Glenn Snyder, et al., Conflict Among Nations: Bargaining, Decision making, and System Structure in International Crises 12 (1977).

65. Robert Putnam, Diplomacy and Domestic Politics: The Logic of Two Level Games, 42 Int'l Org. 427 (1988).

66. Id. at 437

67. Id.

68. Id. at 438.

69. Id. at 440.

70. Keisuke Iida, When and How Do Domestic Constraints Matter?, 37 J. Conflict Resol. 403 (1993).

71. Id. at 404.

72. Jongren Mo, The Logic of Two Level Games with Endogenous Domestic Coalitions, 38 J. Conflict Resol. 402, 412 (1994).

73. Foreword by Lamberto Dini in Duff, at xxix

Daniel W. Boettcher will graduate with a Juris Doctor from the Georgetown University Law Center in May, 2000 and holds a BA from the School of International Service at American University. A former employee in trade advocacy at the United States Department of Commerce, Mr. Boettcher is a founding member of the Tabula Rasa Institute and is Vice-President of its Board of Directors.

OJPCR: The Online Journal of Peace and Conflict Resolution is published by the Tabula Rasa Institute, www.trinstitute.org.