Adjusting the Convention of Dublin

How to develop a system regulating the possibilities to seek asylum in the European Union in such a way that it does not only provide an efficient and fair procedure for asylum applications but also a realistic basis for burden sharing?

Whether the required consensus for the operation of refugee law can be achieved depends on a balance between legitimate interests of states to control entry and the legitimate interests of refugees to be protected against human rights violations. Taking into consideration this complex set of interests the starting point for adopting a balanced approach is the human rights obligations of the states and thus the international protection of refugees. The interest of states is to preserve their prerogative to control immigration by applying orderly resettlement programmes and to combat the irregular arrival of spontaneous non-citizens no matter whether they have left their former place of habitul residence voluntarily or involuntarily .

Hence, asylum is characterized by a paradox: conceptually, it transcends national boundaries but it can only be set in motion within national boundaries. This illustrates the fifth dilemma of this conference: It is the dilemma that bringing international protection into operation presupposes the existence of a national legal system which is politically based on the doctrine of national sovereignty but the principle of universality, which is the basis of human rights law as well as refugee law, precedes national sovereignty.

II.The meaning of international protection

When rethinking asylum and refugee law the starting point is the concept of international protection, which protects refugees and asylum seekers against refoulement and offers them durable solutions, be it by integrating them into the society of the asylum country, resettlement into a third state or by voluntarily returning them in safety and dignity to their country of origin. However, as long as persecution continues the last option is neither conceivable nor acceptable. Who is it then who is to offer international protection? UNHCR? The first state in which the refugee arrives? The state in which he or she wants to stay?

From the beginning, refugee law has avoided answering the question of state responsibilty. The doctrine of state sovereignty has always been a severe hindrance to human rights protection in the refugee field. Each state is free to deny access as long as it does not breach the principle of non-refoulement. This merely negative concept of refugee law has led to a somewhat of a jungle of law allowing the sending of refugees to any state except the state of origin. It is well known that this practice has been a dominant feature in Europe in the seventies and eighties and to a certain degree still is today.

Time and again states have insisted on their sovereign right to control immigration and upheld discretionary admission policies. If states are going to grant protection, they wish it to be operationalised through orderly resettlement programmes. Conversely, asylum, i. e. international protection for spontaneously arriving refugees challenges state sovereignty. If the endorsement of the right to asylum is more than mere lip service the doctrine of state sovereignty is restricted by the human rights of those refugees. Human rights precede states‘ interests.

One of the most challenging questions today is thus how to devise elaborate a system of state responsibility premised on human rights considerations. The concept of international protection cannot be implemented by individual states in isolation but requires a co-operative system of interstate action. Not the international community as such, neither UNHCR nor any other international organisation, can operationalise the function of protection.

From a human rights based perception of international law it is the sending state which is responsible for determining that the risk of refoulement by the third state is effectively excluded. If the sending state cannot exclude the risk of further deportation, it loses control of the claimant and, therefore, cannot act in compliance with its international obligations.

Hence, the best means to avoid further deportation is a proper processing of the substance of the claim. If the state feels, however, that it can refer the claimant to a third state it must at least see to it that it maintains control of the removal unless effective protection is granted or the claim is processed in an efficient and fair procedure in the third state.

The Dublin system was one of the first attempts in international law to solve the intricate and highly controversial question of state responsibility by elaborating a co-operative system of interstate action. Within the context of the EU states have overcome the negative approach of refugee law which only forbids refoulement but leaves open the question of who should offer protection. Although the Dublin system suffers from severe insufficiencies it is, conceptually, still a reasonable basis for the elaboration of strategies, policies and legal doctrines in the field of asylum.

The implementation of Article 63 paragraph 1 no. 1 lit. a) of the Treaty of the European Union may thus be based on the principles of the Dublin system but some of these need improvement before becoming part of Community law. Particularly, the question has to be answered whether a common right of asylum should be recognized as part of Community law. If this questions is answered it has to discussed if and under which conditions claimants may be sent to non-EU states (cf. Article 3 paragraph 5 of the Dublin Convention). In this context the relationship between the concept of reception in the region and the common right of asylum within the EU needs clarification. This leads to a general strategic debate on the question whether political solutions should precede the concept of international protection. Additionally, the criteria of establishing the state responsible for determining claims require further reflection, and in this context the criteria of family reunion is of particular importance. Finally, the question of burden sharing and how it has been applied through the Dublin system of state responsibility remains to be solved.

In its 1991 Communication the Commission emphasized that there should be a "common right of asylum" within the Community. Although an applicant cannot derive from the Dublin Convention an individual right to a material examination of his or her claim, it is the explicit objective of the convention as set out in the Preamble to provide asylum seekers with safeguards that their claims are examined in one member state. Hence, the Dublin Convention could be seen as a tool to implement a right of asylum. Thus the Dublin system reduces the phenomenon of refugees in orbit.

However, Article 3 paragraph 5 of the Dublin Convention retains the right of the contracting states to refer a claimant to a non-EU state instead of sending him or her to the state responsible. In their resolution on host third states the ministers in charge of immigration agreed that a claimant can be sent to a third state if he "has already been granted protection in the third country or has had an opportunity, at the border or within the territory of the third country, to make contact with that country‘s authorities in order to seek protection, before approaching the member state in which he is applying for asylum, or that there is clear evidence of his admissibility to a third contry".

Whereas the first alternative can be integrated into a human rights based system of international protection, provided effective protection is still available in the third state and the competent authorities will be asked for approval to readmit the claimant, the second and third option fall short of the standard of international protection. A former stay of the claimant by which some close links have been established in the third country brings interstate action into play. Furthermore, since consent is essential for shifting responsibility from a EU-state to a non-EU state, the sending EU-state must ask for appoval by the host third state.

If however the claimant has no links with the third state and, particularly, has not travelled through this state on his way to the member state nothing justifies sending him to that state. From a normative point of view, there must be a reason to deem a given state as the more appropriate to afford protection to the claimant, that is, a previous stay in that country.

State practice as it has developed takes this to be the starting point for bringing the third host country concept into operation. Mostly, the mere transit or even an airline stop-over is not sufficient ground to presume adequate protection in a third country. Excluding asylum seekers from access to asylum procedures simply on the basis of transit or the mere possibility of seeking protection in a third state during a stop-over is generally not considered useful in resolving refugee issues, since it implies shifting the whole task of providing protection to those countries which just happen to be the first asylum countries.

From a legal point of view, an evaluation of the real qualities of contact and connections that an applicant has with a given third country is relevant to the existence of adequate protection there. Otherwise it cannot be presumed that the refugee will be afforded protection by the third state with which he has no links other than mere transit. Accordingly, the concept of adequate protection is premised on an actual link to the third country. This is also the view of the US Supreme Court which states that adequate protection means that the claimant must have firmly resettled in the other country. Hence, refugees do not lose their right to protection simply because they make ‚stops along the way‘ in their search for a country of protection.

A common right of asylum means that international protection normally is operationalised within the Community. Only in those cases in which adequate protection is still available in a third state and its competent authorities consent to the transfer of responsibility may international protection be granted outside the Community.

Conceptually, it seems practical and to be in accordance with human rights obligations to define and implement a common right of asylum along these principles. Thus, the 1992 resolution on host third states may only be integrated into Community law if the second and third option is dropped and, additionally, if it is explicitly stipulated that transfer of responsibility from the EU to a third state is based on consent and that it is provided that the claimant will be granted access to a proper and fair procedure there. Thus the policies of some member states to make general assessments on the safety in third host states and remove claimants to such states without a prior detailed examination of their safety in such states should strictly be prohibited by Community law. This is also the view of UNHCR.

If transfer of responsibility takes place the sending EU-state must at least see to it that it maintains control of the removal of the claimant and his further treatment. Interstate action does not mean that the sending state is relieved of its responsibility after the removal. It must rather see to it that the claimant will be granted protection or access to the procedure, and that the third state will not remove the applicant to a further "third host state". Thus, the perception of the German Constitutional Court, that member states of the Refugee Convention are also allowed to send claimants to a third state if this state will refer him to a further third state ("Viertstaat"), is not acceptable. It encourages the policies of neighbouring states to connect bilateral readmission agreements with each other and by this remove the basic concept of state responsibility.

In its note on guidelines for an European migration and asylum strategy the Presidency endorsed that an European strategy on migration and asylum should be comprehensive and cover all aspects of migration and asylum, in particular questions of foreign and development policies. Such a strategy should eradicate or at least reduce root causes of involuntary movements, prevent conflicts, promote democratic structures and apply an active human rights policy. Already in 1994 the Commission had stressed the need for the development of a comprehensive approach encompassing a combination of policies: it requires the coordination of traditional areas of activity, such as social policy, aspects of common foreign and security policy and trade, cooperation and development instruments as well as migration management policies.

Strengthening preventive strategies to combat human rights violations and promote human rights is crucial to tackling the root causes of enforced migration and refugee flows. However, it should not precede strategies which aim at affording international protection to refugees. In the past critics had stressed that the trend towards preventing refugee flows may jeopardize the most fundamental principle of refugee protection, i.e. the right of a refugee to flee his country. Particularly, the question whether promotion of the right of people to remain in their places of origin may serve to expose them in an unjustified fashion to risk of serious harm.

Implementing the concept of a comprehensive approach is not without risks. In the beginning it was combined with the concept of "preventive strategies" which was developed to ensure that people did not flee. Hence, it was no longer protection that was conceived as preventing refugee flows, but containment which has taken many forms and serves many purposes. Unlike the EU High Level Working Group which has elaborated a comprehensive approach aiming primarily at the containment of migratory movements resolution 80 (XLXVII) (1996) of the Executive Committee of the Programme of UNHCR is protection-based and calls into question the responsibility of states "to ensure conditions which do not compel people to flee in fear, to uphold the institution of asylum, to create conditions conducive to voluntary repatriation, to take steps to meet essential humanitarian needs and to cooperate with countries on whom the large-scale presence of refugees weighs most heavily."

To summarize: from a strategic point of view a comprehensive approach should be developed which covers the variety of relevant aspects but may not weaken the concept of international protection. It is in the light of these considerations that the concept of reception in the region, which may entitle European states to send asylum claimants to available and internationally agreed and monitored reception facilities in their region of origin, cannot be endorsed. Firstly, it may encourage European states to continue to apply policies which are not premised on the concept of adequate protection. Secondly, the overwhelming majority of today‘s refugees reside in their region. Hence, the elaboration and implementation of the concept of reception in the region would run counter to the principle of burden-sharing as related to the universal context of the refugee problem. Thirdly, sending spontaneously arriving claimants against their will to countries to which they have no actual links would jeopardize the concept of adequate protection. Finally, such a concept would be a step backwards as regards the standard of ExCom Conclusion No 15 (XXX) on "Refugeres without an Asylum Country" which considers as one element in this context the intentions of the claimant as to the country in which he wishes to request asylum.

V. Burden-Sharing within the Community

The Convention of Dublin is perceived to serve the function of burden-sharing within the Community. It suffers from many insufficiencies but can still be seen as a tool to put a common right of asylum into operation. It has already been explained that the application of Article 3 paragraph 5 of the Convention should be in line with the concept of adequate protection. Furthermore, the understanding that asylum can be claimed only in one member state provides that all states have relatively common procedures and apply uniform criteria to determine refugee claims.

Such an uniform approach may be seen as an appropriate method for burden-sharing. Analyses of how and why asylum claimants choose particular destinations are scarce. Although, in the majority of cases the choice of country of asylum is not a conscious, rational choice by asylum claimants, the harmonization of so many divergent national systems as to procedural, substantive and aspects of reception may facilitate burden-sharing. A harmonized approach of asylum within the Community may reduce the phenomenon that claimants choose countries which they perceive as more attractive than other ones.

1. Procedural rules

In its document "Towards common standards on asylum procedures" the Commission has proposed that to implement Article 63 paragraph 1 lit. d) of the EU Treaty minimum procedural rules should be developed. Taking into consideration the different traditions, legal systems and administrative practices of the EU-states a common procedural approach can hardly be achieved. The Union is not a federation of states but an association of democratic states to achieve gradually the objective of unification. This must leave room for different ways to implement common rules according to the national system.

Procedural rules should be premised on international law and standard setting. Hence, the following principles should be binding and become part of Community law:

• All asylum claimants, arriving within the jurisdiction of a state in whatever manner , must be referred to the authority responsible for deciding on asylum claims.

• This authority must be an independent and specialized one whose sole and exclusive responsiblity is examining and making decisions on asylum.

• Claims should be examined in the first instance through a personal appearance by every applicant before the decision-maker of the responsible authority.

• Claimants must receive written reasons if their asylum claim is rejected, and have the right to appeal against a negative decision. The appeal authority must normally be of a judicial nature and be independent from the first instance body.

- The appeal must in all cases have suspensive effect on expulsion.

• Accelerated procedures may significantly shorten the length of determination procedures and by this serve a legitimate purpose but should nonetheless satisfy these procedural principles.

2. Determination Criteria

Whereas the Presidency in its draft Action Plan of 18 November 1998 (12028/2/98) considers both measures regarding minimum procedural rules and minimum standards as to the refugee criteria should to be taken within five years the Commission in its working document on common rules for asylum procedures of 3 March 1999 (SEK(1999)271) recommends that the Union should first concentrate on a binding instrument as to procedures within the next two years. To achieve a balanced process of harmonisation and an improvement of burden-sharing procedures as well as determination criteria should be developed within the same period. If both goals may not be achieved at the same time, a coherent approach requires that common standards for the application of substantive law be developed first, followed by measures for the harmonisation of asylum procedures, complementary protection schemes and temporary protection arrangements. An agreed level of determination criteria must form the basis for processing asylum claims; conversely, procedural rules serve the purpose of distinguishing between refugees and immigrants.

With respect to determination criteria it is self-evident that the refugee definition of Article 1 of the 1951 Refugee Convention must be the central basis. Under the Austrian Presidency, however, the relevance of the Convention with regard to major flight reasons was questioned. It was stated that the Convention cannot encompass ethnic conflicts or cope with illegal migration from specific areas of crisis. Hence, the system of international protection should be re-formulated taking into consideration new trends which perceive protection primarily as a political offer of asylum states instead of an individual right.

This perception of the linkage between protection and political solutions is anything but an adequate expression of current refugee law. Rather the five reasons of the refugee definition illustrate that ethnic, racial and religious conflicts were envisaged as prominent reasons for flight by the authors of the convention. According to comparative law civil wars and internal strifes do not automatically exclude people from the scope of the convention, provided the claimant is able to establish an individual fear. The 1996 Joint Position on Article 1 of the 1951 Refugee Definition, which is ambigious so far, should thus be reformulated. Hence, Community law should consider comparative state practice and not exclude civil wars and similar situations, unless the claimant fails to establish an individualized fear.

If the claimant cannot establish such a fear but according to the prevailing circumstances in his country cannot be returned the question of supplemantary protection has to be reflected. The required solution to problems concerned is primarily of a political nature. Neither Article 3 ECHR nor Article 3 of the UN Convention against Torture can serve as a legal basis in this context. The long-standing jurisprudence of the ECHR requires that the claimant can establish a "concrete risk" to be subjected to torture or inhuman treatment. The same individualstic approach is taken by the Committee against Torture which brings Article 3 CAT into operation if the claimant can establish that he would be exposed to a real risk or would be personally at risk of being subjected to torture.

Notwithstanding this narrow scope of these instruments they should be incorporated in Community law as a basis for determining asylum claims. Although Article 63 paragraph 1 of the EU Treaty refers only to the Geneva Refugee Convention a uniformed asylum policy should be premised also on Article 3 ECHR and Article 3 CAT. Whereas it is true that these norms cannot form the basis for a supplementary protection approach which goes beyond the individualistic concept of refugee law the question whether the claimant may satisfy the individual test of Article 3 ECHR or Article 3 CAT cannot be easily answered at first. A proper and fair processing is required to identify those claimants who may establish a concrete risk of being personally subjected to torture. Furthermore, both Article 3 ECHR and Article 3 CAT are applicable also in situations of failed states as the responsible treaty organs have explicitly ruled. As long as the question whether the claimant would be subjected personally to torture or inhuman treatment cannot finally be answered he may benefit thus from the refoulement component of these norms.

3. Supervisory Mechanism

A major structural deficit which characterizes the area of intergovernmental co-operation before the Treaty of the EU came into force was that there was no independent international supervisory mechanism. Although access to the ECJ is restricted by article 68 of the Amsterdam Treaty controversial questions of harmonisation of asylum may be decided by the ECJ in the near future. Compared with the past the positive aspects of this supervisory mechanism cannot be underestimated. Additionally, binding interpretation of Community law for all member states in the field of asylum may serve the purpose of burden-sharing within the Community.

The ECJ should interpret asylum related provisions of Community law in accordance with international human rights standards. This aim could be achieved at best by incorporating the protection provided by the ECHR and by CAT into Community law. Additionally, a clear link to the 1951 Refugee Concention and the 1967 Protocol concerned is required to assure that the ECJ will take rulings in the light of refugee law standards, primarily as developed by the executive authority and the office of UNHCR. The court may also consider the Handbook of the UNHCR office on determination criteria when it takes decisions.

In this context the proposal to establish an European Asylum Agency seems to be attractive because it is deemed to discourage secondary movements of asylum claimants on the territory of member states by monitoring uniform determination criteria, procedures and reception facilities. Such an agency should, however, not take determination decisions as is proposed by the academic group. First, considering the lengthy proecess of giving only restricted competence to the ECJ it is hardly conceivable that member states will agree on relieving themselves of the competence to decide on asylum claims. Second, an European administrative decision-making body seems not compatible with the character of the Union as an association of national states. Third, practical problems of monitoring and managing such a system stand against its implementation. Rather, the agency should be given a strong advisory role, particularly, with regard to decision making. UNHCR and NGOs should be included.

4. Determining the responsible state

The burden-sharing mechanism of the Dublin Convention is premised on the concept that only one state is responsible for determination of refugee claims. So far this system should be preserved and integrated into Community law. It suffers, however, from severe weaknesses which should be removed before a binding instrument is developed. The Council and the Commission in its joint action plan consider the effectiveness of the Dublin Convention also a matter of high priority.

Firstly, from a conceptual point of view the agreed set of criteria to allocate responsibility to a member state seem to be acceptable but uncertain procedures for establishing responsibility prejudice member states in the Eastern part of the Community, in particular Germany. Secondly, lengthy procedures to identify the responsible state in cases of applicants illegally crossing national borders run counter to the agreed objective to accelerate procedures and take decisions at the earliest time possible.

Therefore consideration should be given to implementing the proposal to establish a system of financial compensation for member states receiving high numbers of asylum seekers and refrain from applying time-consuming and often useless procedures in cases of illegally arriving applicants to identify the state responible. Although, statistics about the actual outcome of transfers within the Dublin system are rare, the observation seems to be realistic that in most instances the member states where the first application is made is in the long run also the state responsible for the processing of the claim.

Hence, when reflecting on the Dublin system the principle that only one member state should be responsible for determining asylum claims should be maintained but in cases of illegally arriving claimants lengthy procedures to establish the responsible state should be avoided, unless there is clear evidence as regards the way the claimant reached the territory of the Community. When transfer of the claimant is undertaken the sending member state must inform the responsible member state that the claim was only found inadmissible and the responsibility to take a decision on the merits of the claim rests with the responsible member state.

Additionally, Article 9 of the Convention should be redefined and provide for a humanitarian clause in order to avoid separation of family members or other situations impacting negatively the protection needs of asylum seekers. Generally, it should be ruled that family reunion should take place if the claim cannot be dealt with by means of accelerated procedures.

Finally, to upheld the common right of asylum readmission agreements must satisfy the concept of adequate protection. They must also contain specific safeguards against refoulement. In the long run transfer of reponsibility to non-EU states should only be based on a multilateral agreement between the Union and specific third states, provided the determination criteria and procedural rules in the receiving states meet the level already established within the Union. It is in this context that the Commission endorsed the active interest the Union should take in neighbouring countries of Central and Eastern Europe, since the successful application of the third host country will depend on how these countries can cope.

The history of intergovernmental cooperation had generated the adverse impact that the required distinction between illegal immigration and involuntary movements has been blurred and consensus on the need to protect refugees has weakened in European societies. Particularly, the Schengen context has played a dominant role so far, in which the focus is primarily on the combat against illegal entry and sojourn as set out in Article 9 of the 1985 Schengen Agreement. Hereby legitimate reasons to leave the country of origin have not caught public eye and awareness of the need to grant protection to those in need has lost ground.

This has led to a situation in which public support for the setting in operation of costly and complicated determination procedures is diminishing. To continue to grant protection to refugees a proper distinction between refugees and voluntary migrants is, however, crucial. The High Commissioner also endorses that although it is sometimes difficult to establish the distinction between refugees and economic migrants in practical terms this distinction must be preserved. States must respect refugees right while they seek solutions to the problem of irregular migration.

Speaking in strategical terms, the political will of states to give place to alternative solutions cannot be promoted by relinquishing the case by case approach of the convention. Today, refugees are no longer welcomed by host states, primarily by Western states. In almost all of the industrialized states, there has been a flurry of official activity over the last decade to obstruct or deter the arrival of asylum seekers. As the High Commissioner rightly points out in her statement to the Executive Committee of UNHCR‘s Programme in 1997 there is an increasing trend towards restrictive asylum policies in many Western countries, including rejection of asylum seekers at borders, interdiction at sea and the narrow interpretation of the refugee definition, including limiting this definition to persecution by state actors.

From the beginning, the Commission did not contribute to this specific way to harmonise justice and home affairs. Although it endorsed the need to prevent abuse of the right of asylum it has always made clear that at the same time rights of asylum seekers must be protected. With the Amsterdam Treaty coming into force the Union has overcome the area of harmonisation behind closed doors. The view that this Treaty is only a "dummy" ("Mogelpackung") is thus not a correct observation. An illustrative example of the change in attitudes and policies is the note of the Presidency on Guidelines for an European Strategy on Migration and Asylum which endorses the obligations deriving from the Refugee Convention and the European Convention of Human Rights, underscores that the implementation of a Strategy on Migration and Asylum provides consensus by the society and requires thus an offensive und comprehensive campaign to achieve public support.

There seems to be a realistic chance that the human rights tradition of Europe may remove the inward looking characterizing the past. In the coming years the implementation of the provisions of Title IV of the Amsterdam Treaty will become the test whether this tradition prevails. Challenges ahead are not easy to respond to but it seems to be a realistic expectationnow that the problems may be solved in a spirit of solidaritywith those in need of protection and that consensus of the European societies could be reachieved.

To summarize these considerations the following points need to be stressed: