European Agencies for Criminal Justice and Shared Enforcement

European agencies are playing an increasing role in the area of freedom, security and justice (AFSJ). The European Police College (CEPOL), the European Asylum Support Office (EASO), the European Institute for Gender Equality (EIGE), the European Monitoring Centre for Drugs and Drug Addiction (EMCCDA), the European Agency for large-scale IT systems (eu-LISA), Europol, Eurojust, the Fundamental Rights Agency (FRA) and the External Action Service (EEAS), that includes EU SITCEN, are the new players in the field and they operate even as a Justice and Home Affairs (JHA) network.1 This network proposes new measures and concrete actions, both in relation to their cooperation as well as to their functioning in the AFSJ, and they have reporting duties to the Standing Committee on Operational Cooperation on Internal Security (COSI). In 2011 at COSI they presented a final report on enhanced cooperation between the JHA agencies and a final scoreboard on JHA cooperation.2 Compared to classic EU agencies in the field of the internal market, the AFSJ agencies, also indicated as Justice and Home Affairs agencies (the JHA nomination of the Treaty of Maastricht’s third pillar), certainly have less regulatory and more operational powers. They also have as their distinctive feature that their operational activity is strongly interlinked with the national law enforcement communities. If we concentrate on the criminal law dimension and judicial cooperation in the AFSJ, we can limit our analysis to Eurojust and the proposed European Public Prosecutor’s Office (EPPO).3 There is no doubt that the supranational designs of these European judicial agencies, even if they would dispose of European-wide investigative powers in the common European territoriality of the AFSJ, would still need to be embedded in the national justice systems. The reasons for this shared enforcement design are multiple. In criminal justice, adjudication is the exclusive competence of the Member States. This means that all prosecutions and trial procedures are national. The investigative and prosecutorial efforts of European agencies have as their final destination: criminal evidence in national procedures. Even during the investigations, the European judicial authorities will therefore have to continually interact


Introduction
European agencies are playing an increasing role in the area of freedom, security and justice (AFSJ). The European Police College (CEPOL), the European Asylum Support Office (EASO), the European Institute for Gender Equality (EIGE), the European Monitoring Centre for Drugs and Drug Addiction (EMCCDA), the European Agency for large-scale IT systems (eu-LISA), Europol, Eurojust, the Fundamental Rights Agency (FRA) and the External Action Service (EEAS), that includes EU SITCEN, are the new players in the field and they operate even as a Justice and Home Affairs (JHA) network. 1 This network proposes new measures and concrete actions, both in relation to their cooperation as well as to their functioning in the AFSJ, and they have reporting duties to the Standing Committee on Operational Cooperation on Internal Security (COSI). In 2011 at COSI they presented a final report on enhanced cooperation between the JHA agencies and a final scoreboard on JHA cooperation. 2 Compared to classic EU agencies in the field of the internal market, the AFSJ agencies, also indicated as Justice and Home Affairs agencies (the JHA nomination of the Treaty of Maastricht's third pillar), certainly have less regulatory and more operational powers. They also have as their distinctive feature that their operational activity is strongly interlinked with the national law enforcement communities.
If we concentrate on the criminal law dimension and judicial cooperation in the AFSJ, we can limit our analysis to Eurojust and the proposed European Public Prosecutor's Office (EPPO). 3 There is no doubt that the supranational designs of these European judicial agencies, even if they would dispose of European-wide investigative powers in the common European territoriality of the AFSJ, would still need to be embedded in the national justice systems. The reasons for this shared enforcement design are multiple. In criminal justice, adjudication is the exclusive competence of the Member States. This means that all prosecutions and trial procedures are national. The investigative and prosecutorial efforts of European agencies have as their final destination: criminal evidence in national procedures. Even during the investigations, the European judicial authorities will therefore have to continually interact Member States, the proposed decentralized structures of the EPPO and Eurojust will result in a mixture of applicable European and national law with respect to: a) the applicable law for the definition and scope of the powers, including civil rights and liberties and procedural rights (substantive dimension); b) the applicable law for judicial control (institutional dimension); c) the applicable law for the admissibility of evidence (procedural dimension).
We will proceed in the following by focusing, first, on the institutional basis for the establishment of Eurojust and the EPPO (Section 2), followed by an initial analysis of their proposed institutional design in light of the three dimensions just mentioned (Section 3). From thereon, we will focus in more detail on the position of the citizen vis-à-vis these new players at the European level (Section 4).
One final caveat: obviously, our ambition cannot be to offer an all-encompassing answer to the foregoing questions, if only because the developments that we describe are in a state of flux, particularly with regard to the EPPO. The original proposal for the EPPO has meanwhile undergone some significant changes by the Council, 8 partly due to the yellow card that was aimed against the Commission's proposal by a series of national parliaments. At this point in time, it is therefore of no use to offer detailed analyses of the proposals themselves, as these will most likely undergo further modifications. Rather, we aim to focus on the bigger picture. Our ambition is to offer a first analysis of the problems that we anticipate. The relevance of the debates within criminal law are also of concern to similar developments in the areas of financial supervision (ECB and ESMA), and, possibly, other areas of EU law, such as consumer and data protection.

Shared enforcement by Eurojust and the EPPO: the institutional framework
Under the Lisbon Treaty, the Union has set as a key objective to maintain and develop the Union as an area of freedom, security and justice in which both the free movement of persons and the prevention and combating of crime are assured (Article 3(2) TEU). In Title V of the TFEU, on the AFSJ, Article 67(3) elaborates the tools: 'The Union shall endeavour to ensure a high level of security through measures to prevent and combat crime, racism and xenophobia, and through measures for coordination and cooperation between police and judicial authorities and other competent authorities, as well as through the mutual recognition of judgments in criminal matters and, if necessary, through the approximation of criminal laws. ' Under Chapter 4 of Title V, on judicial cooperation in criminal matters, the missions and competences of Eurojust and the EPPO are an integral part of the AFSJ.

Eurojust and the 2013 Eurojust reform proposal
Eurojust has been entrusted in Article 85 TFEU with a European mission and with supranational powers to strengthen judicial cooperation in criminal matters and to coordinate investigation and prosecutions when it comes to the surrender of suspects or convicted persons, the transnational gathering of evidence, the transnational execution of sanctions, et cetera. 9 In its July 2013 reform proposal, 10 Eurojust has been clearly labelled in the title as an European Union Agency for Criminal Justice Cooperation. One has to take into account the history of Eurojust in order to understand the more than symbolic meaning of this label. The discussion on the establishment of a judicial cooperation unit was first held at the special 1999 European Council Meeting in Tampere on the AFSJ and the European Council agreed in its Conclusion 46 that Eurojust should be set up. However, it was at the initiative of Portugal, France, Sweden and Belgium that a provisional judicial cooperation unit was formed in 2000, under the name of Pro-Eurojust. It was clearly a governmental body that was operating from the Council buildings in Brussels. In 2002, Eurojust was established formally by a Council decision. 11 Eurojust became a Union body with legal personality, but the national members of Eurojust (its arms and legs, so to speak) remained financed and governed by their Member States (Preamble, Recital 4). In the July 2013 reform proposal, the Commission underlined in the Preamble that the institutional European dimension of Eurojust will be enhanced through the Commission's participation in the management of the agency and the involvement of the European Parliament and national parliaments in the evaluation of its activities (Recital 6). The Commission will also be represented in the Eurojust College when it exercises its management functions and in the Executive Board to ensure the non-operational supervision and strategic guidance of Eurojust (Recital 15). Even if the Member States remain responsible for the salaries of their national members, their activity within Eurojust shall be considered as the operational expenditure of Eurojust.

The proposal on the European Public Prosecutor's Office (EPPO)
Article 86 TFEU provides a legal basis to establish the European Public Prosecutor's Office (EPPO) and also elaborates on the regulatory specifications when it comes to the investigation, prosecution and adjudication of crimes affecting the financial interest of the European Union (PIF crimes) or, if extended by the European Council, serious crimes having a cross-border dimension: '1. In order to combat crimes affecting the financial interests of the Union, the Council (…) may establish a European Public Prosecutor's Office from Eurojust (…) 2. The European Public Prosecutor's Office shall be responsible for investigating, prosecuting and bringing to judgment, where appropriate in liaison with Europol, the perpetrators of, and accomplices in, offences against the Union's financial interests, as determined by the regulation provided for in paragraph 1. It shall exercise the functions of prosecutor in the competent courts of the Member States in relation to such offences. 3. The regulations referred to in paragraph 1 shall determine the general rules applicable to the European Public Prosecutor's Office, the conditions governing the performance of its functions, the rules of procedure applicable to its activities, as well as those governing the admissibility of evidence, and the rules applicable to the judicial review of procedural measures taken by it in the performance of its functions. 4. The European Council may, at the same time or subsequently, adopt a decision amending paragraph 1 in order to extend the powers of the European Public Prosecutor's Office to include serious crime having a cross-border dimension and amending accordingly paragraph 2 as regards the perpetrators of, and accomplices in, serious crimes affecting more than one Member State. The European Council shall act unanimously after obtaining the consent of the European Parliament and after consulting the Commission. ' In June 2013, the Commission submitted to the Council its proposal for a regulation. 12 The EPPO will not deal with the horizontal judicial cooperation between national judicial authorities, as Eurojust does, but its essence is that it will act through a vertical setting as an investigative and prosecutorial office in the combined territory of the Union's Member States, being a single legal area (Article 25(1) of the proposed regulation). Fourteen chambers of eleven national parliaments 13

Applicable law for the definition and scope of the enforcement powers
Concerning the proper judicial powers of Eurojust, both in the dimension of the College or through its national members, there is a great deal of new potential in Article 85 TFEU, including the initiation of criminal investigations or requesting or ordering such an investigation and the binding resolution of (negative or positive) conflicts of jurisdiction in the AFSJ: '1. Eurojust's mission shall be to support and strengthen coordination and cooperation between national investigating and prosecuting authorities in relation to serious crime affecting two or more Member States or requiring a prosecution on common bases, on the basis of operations conducted and information supplied by the Member States' authorities and by Europol (…) These tasks may include: (a) the initiation of criminal investigations, as well as proposing the initiation of prosecutions conducted by competent national authorities, particularly those relating to offences against the financial interests of the Union; (b) the coordination of investigations and prosecutions referred to in point (a); (c) the strengthening of judicial cooperation, including by resolution of conflicts of jurisdiction and by close cooperation with the European Judicial Network. (…)' In the July 2013 reform proposal relevant articles deal with the tasks and competences of Eurojust (Articles 2-3) and with the operational functions of Eurojust (Articles 4 and 8). The reach of the articles is much more limited than the potential of Article 85 TFEU, when it comes to the judicial powers of Eurojust, but it does contain some elements that could be used to develop the European dimension of Eurojust. Article 2(1) on the tasks states that Eurojust shall support and strengthen coordination and cooperation between national investigating and prosecuting authorities in relation to serious crimes affecting two or more Member States, or requiring a prosecution on a common basis. The latter is new and can refer to the common interests of the AFSJ. In Article 2(3), it is also stipulated that Eurojust shall exercise its tasks at the request of the competent authorities of the Member States or on its own initiative. This 'own initiative' can be used to fill in the European dimension in the AFSJ. Member States are aware of this and are therefore questioning this phrasing during the negotiations. 18 Article 4 of the reform proposal, that deals with the operational functions of Eurojust, stipulates in Paragraphs 2 and 4: '2. In the exercise of its tasks, Eurojust may ask the competent authorities of the Member States concerned, giving its reasons, to: a) undertake an investigation or prosecution of specific acts; b) accept that one of them may be in a better position to undertake an investigation or to prosecute specific acts; c) coordinate between the competent authorities of the Member States concerned; d) set up a joint investigation team in accordance with the relevant cooperation instruments; e) provide it with any information that is necessary to carry out its tasks; f) take special investigative measures; g) take any other measure justified for the investigation or prosecution. (…) 4. Where two or more Member States cannot agree on which of them should undertake an investigation or prosecution following a request made under point (b) of paragraph 2, Eurojust shall issue a written opinion on the case. The opinion shall be promptly forwarded to the Member States concerned. ' Article 4(2) thus does not include any proper powers to initiate criminal investigations at all. Requests by Eurojust cannot be considered as binding mutual legal assistance (MLA) requests or mutual recognition (MR) orders. Finally, Article 4(4) does not include binding decisions on conflicts of jurisdiction either. However, Article 8, dealing with the powers of the national members, stipulates: '2. In agreement with the competent national authority the national members shall: a) order investigative measures; b) authorise and coordinate controlled deliveries in the Member State in accordance with national legislation. 3. In urgent cases when timely agreement cannot be reached, the national members shall be competent to take the measures referred to in paragraph 2, informing as soon as possible the national competent authority. ' To sum up, the proposal thus contains far-reaching ordering powers for the national member in his own jurisdiction, that in a case of urgency can be exercised without prior agreement by the competent national authorities. This means that a vertical chain of command is introduced between the national member of Eurojust, being part of an European agency, and the national law enforcement authorities. The vertical chain of command includes all investigative measures, also the most coercive ones. Even if the reform proposal remains silent on the precise interaction between the order of the national member and the execution by the national law enforcement community, there is no doubt that this power of the national member of Eurojust penetrates national systems and will challenge statist prerogatives. 19

Legal consequences for safeguards, judicial control and the admissibility of evidence
From the point of view of the judicial authorities, Eurojust does not currently take binding decisions on the choice of jurisdiction and is not sending out binding orders concerning investigative measures. For that reason the operational powers do not directly change the legal position of the defendants in criminal proceedings. Their rights and duties are ultimately and mainly affected by the national dimension of criminal justice. 20 If Article 8(2) on the operational powers of the national member would survive the negotiations, the question can be raised whether this reasoning can be upheld. The authorizations and orders are decisions of a national member in its legal order, but the national member is fully part of the European Agency for Criminal Justice Cooperation (Eurojust) and thus acting on behalf of Eurojust. The thesis could thus be defended that the decisions of the national member are Eurojust decisions with legal consequences in the national legal order of the national member concerned. We will come back to this in Section 4.3.

Applicable law for the definition and scope of the enforcement powers
When it comes to the opening and closing of the investigation and the decision to prosecute or to drop the case, there is no doubt that these decisions are taken by the EPPO on the basis of European law. 21 However, when it comes to the operational powers of the EPPO, the situation is rather unclear and complex. Several recitals in the Preamble of the original proposal dealing with this. 22 Recital 7 states that the mandate of the EPPO requires autonomous powers of investigation and prosecution, including the ability to carry out investigations in cross-border or complex cases. However, Recital 9 underlines that, as a rule, the investigations of the EPPO should be carried out by European Delegated Prosecutors in the Member States, albeit that in cases involving several Member States or cases which are of a particular complexity, the efficient investigation and prosecution may require that the European Public Prosecutor also exercises his powers by instructing national law enforcement authorities. Recital 14 goes in the same direction: the operational activities of the EPPO should be carried out under the instruction and on behalf of the European Public Prosecutor by the designated European Delegated Prosecutors or their national staff in the Member States. The Commission is fully aware of the importance of determining the rules of procedure which are applicable to the activities of the EPPO (Recital 14). However, it considers that detailed provisions would be disproportionate and can be left to national law (Recital 14).
The articles in the original proposal do not offer a clearer picture. Article 25(1), dealing with the EPPO's authority to investigate, clearly takes as a starting point the combined territory of the Union's Member States as a single legal area in which the EPPO may exercise its investigative and prosecutorial competence. The article does not use the concept of European territoriality, which would mean that for the purpose of conducting investigations and prosecutions the EPPO would operate in one single legal area and would not need instruments of mutual legal assistance or mutual recognition. However, the combined territory of the Union's Member States is equivalent to the common territory of the AFSJ and could thus be qualified as a single territory. It is not without importance that Article 4(2) makes a clear distinction between different types of investigative tasks of the EPPO. In some situations the EPPO will only supervise the investigations; in others it will direct them itself. The same distinction is reflected in Article 6(4), when stipulating the structure and organisation of the EPPO: 'The investigations and prosecutions of the European Public Prosecutor's Office shall be carried out by the European Delegated Prosecutors under the direction and supervision of the European Public Prosecutor. Where it is deemed necessary in the interest of the investigation or prosecution, the European Public Prosecutor may also exercise his/her authority directly in accordance with Article 18(5). ' However, in cases where the European Public Prosecutor exercises his/her authority directly and is allowed to use his/her autonomous investigative powers, Paragraph 6 clearly imposes new conditions: 'Where the investigation is undertaken by the European Public Prosecutor directly, he/she shall inform the European Delegated Prosecutor in the Member State where the investigation measures need to be carried out. Any investigation measure conducted by the European Public Prosecutor shall be carried out in liaison with the authorities of the Member State whose territory is concerned. Coercive measures shall be carried out by the competent national authorities. ' Article 18, dealing with the conduct of the investigations and thus with the use of investigative powers, is mostly adopting an approach based on national jurisdiction. The authority conducting the investigation will be the European Delegated Prosecutor, under the supervision of the EPPO. In the case of cross-border cases, the article even refers in Paragraph 3 to the setting up of joint teams. The powers of the EPPO are summarized in Paragraph 4 as monitoring, coordinating and instructing. It is only in Paragraph 5 that the autonomous investigative powers of the EPPO are again stipulated and are submitted to a set of criteria.
If we combine this with Article 26, that lists the non-coercive and coercive investigative measures, it becomes very clear that the tasks and powers of the EPPO, in its autonomous investigative function, are limited to non-coercive measures and that even in those cases the EPPO may be obliged to ask for judicial ex-ante authorization if national law prescribes this: '4. Member States shall ensure that the investigative measures referred to in points (a)-(j) of paragraph 1 are subject to authorisation by the competent judicial authority of the Member State where they are to be carried out. 5. The investigative measures referred to in points (k)-(u) of paragraph 1 shall be subject to judicial authorisation if required by the national law of the Member State where the investigation measure is to be carried out. 6. If the conditions set out in this Article as well as those applicable under national law for authorising the measure subject to the request are met, the authorisation shall be given within 48 hours in the form of a written and reasoned decision by the competent judicial authority. ' The final picture is that (already under the original proposal, but certainly after its modifications) the EPPO will be dependent on the delegates in the national legal order for coercive measures. Those delegates will apply the (partially harmonized) national criminal procedure. Moreover, the Delegated Prosecutors will wear a double-sided hat. Based on Article 6 they have exclusive competence for EPPO offences, but can also exercise their function as national prosecutors. This can result in an investigative and prosecutorial outcome that is not very different from the actual situation, which has been qualified as not achieving effective enforcement. Moreover, there is the risk that the national authorities will face an unclear division of labour, allowing them to hide behind national priorities and obstacles, or not preventing an overlap of enforcement efforts. It is also astonishing that the relation between the investigative powers of the EPPO and the mutual legal assistance and mutual recognition instruments remains unclarified in the text. The revision by the Greek Presidency does not only include a new governing structure within the EPPO (College, Permanent Chambers, European Prosecutors, European Delegated Prosecutors), but seems 23 to deduce from the fact that all investigations and prosecutions will be conducted in jurisdictions of the Member States that the proposal no longer has to deal with rules of procedure on investigations and related procedural safeguards. In that case, the applicable law for the definition and scope of the enforcement powers would be exclusively national.

Legal consequences for safeguards, judicial control and the admissibility of evidence
When it comes to the applicable procedural safeguards and human rights standards the recitals of the original EPPO proposal contain several references. Both Recitals 17 and 33 enshrine the full application of the CFR, especially Articles 47-48 and 50. Moreover, according to Recital 34, the harmonized procedural safeguards under Article 82 TFEU also apply. Thirdly, the recital refers to extra guarantees in the proposed regulation. Finally, Recital 11 refers to the ex-ante judicial control over coercive investigation powers and the ex-post judicial control by the trial court as to the compliance of the gathered evidence with the CFR.
Concerning human rights standards, Article 11(1) of the original proposal leaves no doubt as to the application of the CFR. Article 32 deals specifically with the rights of suspects and accused persons and imposes on the EPPO the obligation of fully complying with the rights of the defence and the right to a fair trial under the CFR and with the harmonized procedural safeguards under Article 82 TFEU, plus the right to remain silent and the right to be presumed innocent as well as the right to legal aid.
The original proposal does not deal explicitly with the transnational application of these standards. As most of the human rights standards will apply when applying coercive measures or at the trial setting, they will inevitably apply through national applicable law. Neither does it deal with criteria for the choice of the investigative jurisdiction. 24 The EPPO is thus free to call in or activate delegates in the national jurisdictions, and can thus 'optimalise' enforcement opportunities (forum shopping) in the AFSJ.
In line with the decentralized exercise of the coercive measures, the EPPO will need ex-ante authorization for all listed coercive measures under Article 26(1) (a-j) and for other measures if required by national law, by the competent judicial authority of the Member State where the investigation is to be carried out. This means that there is a 'géometrie variable' when it comes to the need for ex-ante authorization. The requirements and procedures of ex-ante judicial authorization are not harmonized and thus fully depend on national law. To give an example: can a judicial authorization be challenged on appeal (when it is not secret); does the appeal have a suspending effect? Finally, the original proposal does not apply the mutual recognition regime to the ex-ante judicial authorization. 25 This means that the EPPO has to seek judicial authorization in every single jurisdiction where it wants to trigger coercive measures.
From the point of view of the EPPO's investigative and prosecutorial interests this is without any doubt a complex patchwork. However, the EPPO can also optimalise enforcement through its choices of the territorial jurisdictions. Digital evidence is for instance easy to obtain from several jurisdictions. As a result of free movement, the same is increasingly true for evidence to be provided by European citizens (witness evidence, for instance). It can therefore be interesting to select a certain jurisdiction (and exante judicial authorization) because of its favorable requirements and proceedings from a prosecutor's perspective. The free choice of jurisdiction and the inherent possibility of forum shopping is not at all conditioned or compounded by legal consequences when it comes to the admissibility of evidence, as Article 30(1) stipulates: 'Evidence presented by the European Public Prosecutor's Office to the trial court, where the court considers that its admission would not adversely affect the fairness of the procedure or the rights of defence as enshrined in Articles 47 and 48 of the Charter of Fundamental Rights of the European Union, shall be admitted in the trial without any validation or similar legal process even if the national law of the Member State where the court is located provides for different rules on the collection or presentation of such evidence. ' The revision by the Greek Presidency, relying completely on national law for the definition and scope of the enforcement powers, is at first sight a simple design based on the lex locus. The EPPO can set up Permanent Chambers with full freedom and steer towards the most appropriate jurisdictions to investigate and prosecute, without having to take into account any statutory criteria for those decisions. However, under the (Greek) proposal as it is now published, it also has to take into account that the rules on the admissibility of evidence are not harmonised, which can be an obstacle to or a facilitator of effective enforcement. It is also questionable what the added value is to Eurojust and the existing joint investigation teams. 26

Introductory remarks
As pointed out in the previous section, both the proposed EPPO and Eurojust more or less abandon, and certainly mitigate, the traditional viewpoint that law enforcement is an exclusive matter for national authorities. These European agencies have the capacity to set their own agenda on operational matters and to commence investigations and deploy coercive measures all over the territories of the participating Member States. It is also clear that the proposed frameworks for Eurojust and EPPO will give rise to many questions, which are ultimately all somehow related to the fact that the rules on the determination of the applicable legal regimes are unclear, and that the gaps and duplications between those regimes are not removed by the proposed regulatory frameworks. The focus on the integration of both EU bodies into the national legal orders of the Member States is so strong that the EU legislator seems to have lost sight of the inherent transnational setting of both bodies. In this particular context, not only will the relevant authorities face problems in determining the applicable legal regime, but also -and we think that this is even worse -the legal position of the citizen will become seriously complicated.
In the remainder of this article we will therefore go into the question of how such concepts as European territoriality and the mutual admissibility of evidence influence the proper operation of the fundamental rights and liberties that are related to criminal justice, including the right to privacy (Articles 6 and 7 CFR), to a fair trial and an effective remedy (Articles 47 and 48 CFR), as well as the principle of ne bis in idem (Article 50 CFR). We will focus on a) how these concepts, that together shape the notion of shared enforcement, influence the traditional requirements that the criminal law must be accessible and foreseeable to all citizens, whether they are defendants or otherwise involved in criminal proceedings (Section 4.2), b) how they interact with judicial control (Section 4.3), as well as c) how the rights of citizens -with a focus on the rights of the defendant -are effectuated in a transnational legal area (Section 4.4).

Determining the applicable legal regime: accessibility and foreseeability revisited
It is a common feature of every criminal justice system of the Member States of the European Union that government action that interferes with the civil rights and liberties of its citizens needs a sound legal basis in the law. According to standard Strasbourg case law, and in order to pass the test of the rule of law, interferences with such rights and liberties must have a sound legal basis and must meet minimum requirements of accessibility and foreseeability. 27 There is no reason to doubt that this is any different for an EU system of criminal justice, as is also apparent from Article 2 TEU.
We already noticed in the previous section that both the EPPO and, to a far lesser degree, Eurojust have been given powers to deploy intrusive coercive and covert investigative measures. 28 Both proposals indicate, in fairly broad terms, what powers ought to be available to Eurojust (members) or the EPPO, and then refer back, as a general rule, to the legal order where those measure are to be deployed for their precise scope and content, including the available safeguards. With respect to certain series of more intrusive measures (Article 26(1) (a-j)), the original EPPO proposal however requires the Member States to have in place a series of safeguards, amongst which are prior judicial authorization and the existence of reasonable grounds for the use of those measures (see Article 26(2)(4)).
The question that interests us here is whether and how this decentralized system meets the requirements of accessibility and foreseeability in a transnational common enforcement area in which the free movement of citizens is strongly promoted. From the perspective of citizens (suspects, but also others), such a decentralized system has a series of disadvantages. As said, the combination of a regime that largely 29 refrains from rules on where to locate investigative and prosecutorial activities ('forum choices') and a mixture of EU and divergent national rules on investigative powers will lead to a situation in which citizens have great difficulty in assessing the scope of investigatory powers and their legality. Or, from the reverse angle, such a combination leads to situations in which executive powers enjoy great discretion in determining the applicable legal regime. That could lead to situations in which forum choices are made deliberately with the aim of circumventing the procedural safeguards of a particular legal order. In other words, there is a risk of forum shopping and of a race to the bottom, particularly now that sources of evidence -witnesses, documents, data, et cetera -and methods of collecting evidence -online searches, for instance -have also become mobile.
The question is therefore whether and how a patchwork of EU and diverging national legal rules relate to the well-known fundamental rights standards that we mentioned. There are roughly two opposite approaches to this. The first position takes the stance that interstate differences are an intrinsic part of the EU's legal order, and that the original EPPO proposal does introduce a series of safeguards that ought to be in place in any event (Article 26(2)(4)) and which will guarantee a certain level playing field concerning procedural safeguards. In such a context, it is precisely because all legal orders adhere to certain standards of fundamental rights that mutual trust is justified in each other's legal orders and, therefore, evidence may be mutually declared admissible. 30 A much heard related argument in that regard is also that the forum shopping argument has a reverse side. Free movement rights may be abused by citizens: 'Criminals must find no ways of exploiting differences in the judicial systems of Member States' . 31 In other words, the EU's shared legal order is not under any legal obligation to allow its citizens to assess under which legal order they are most likely to evade justice.
We have difficulty in accepting this position and propose an alternative, second approach. 32 To support that approach, we must mention, first of all, that we consider the number of safeguards that are introduced in the proposal to be very limited and at any rate insufficient to support the claim of a level playing field of procedural safeguards in general. Arrest and pre-trial detention are for instance excluded from the harmonizing measures. The same holds true for important safeguards, such as a specific, qualified degree of suspicion, 33 and/or purpose limitation. 34 Regarding the latter, it must also be noted that national requirements of purpose limitation may no longer be used in EPPO investigations, regardless of the existence of such restrictions in comparable national cases. Finally, we would like to mention that the series of measures that have not been subjected to minimal procedural safeguards at the EU level still contain, inter alia, the summoning of witnesses, access to premises (including houses?), and controlled deliveries. Regarding the 'reversed forum shopping' argument, it must moreover be added that the proposed measures will not only affect (alleged!) criminals, but also third parties. Their houses may be searched, too. Their electronic/digital communications may also be of interest to the law enforcement bodies.
Thirdly, and most importantly, the aforementioned first position provides citizens with no answer as to the question of on which of the potential applicable legal regimes they should rely for effective judicial control. Obviously, as the ECtHR has also held on many occasions, the goal of the safeguards of Article 8 (or 5) ECHR is not to allow citizens to predict with great precision under which circumstances certain measures will be deployed; rather, the goal is to give citizens an adequate indication of the conditions and circumstances in which the authorities are empowered to resort to any such measures. 35 That goal not only serves the purposes of legal certainty, but also, and perhaps even more, the purpose of legal protection, as it allows for ex-ante and ex-post control of the executive.
The problem in our case is not the absence or vagueness of a (single) law that offers such a point of reference, but the presence of a multitude of diverging applicable laws, all of which in themselves may be perfectly in line with Articles 8 ECHR / 6 and 7 CFR. Nonetheless, can we really say that in such a decentralized system, which provides the executive with an almost unlimited choice between multiple applicable legal regimes, an adequate indication to citizens is offered as to the scope of intrusive investigatory and prosecutorial powers? The aforementioned first position offers no conclusive answer to this problem, mainly emphasizing that all legal orders in themselves indeed meet the requirements of the ECHR and the Charter.
At present, there is no authoritative legal source that confirms the viewpoint that executive discretion in choosing between a multitude of legal regimes indeed meets (or fails to meet) standards of 'adequacy' . We are of the opinion, however, that it is unrealistic and, arguably, in contradiction with the concept of the free movement of citizens, 36 that EU citizens are required to know all the legal regimes that they may possibly be subjected to. What is more, even if they would be able to do this, then it would still be impossible to exert effective control over investigatory actions, as evidence has also become mobile and it is likely that there will always be a legal regime that produces the outcomes desired by the executive powers. In our view, this situation boils down to allowing the executive to define or change the rules of the game during the course of the game just as much as with the absence of an appropriate legal basis, or a basis that is insufficiently determinative in defining the scope and impact of criminal law powers.
There is one more step to be made. The specific, transnational setting of the EPPO and Eurojust may call for a mitigation of ECtHR standards. 37 Our analysis above after all leads us to the conclusion that, as far as the EPPO and Eurojust are concerned, it is not enough to have mutual trust in the legal systems of all EU Member States. We are in need of additional mechanisms that offer the authorities 'adequate' guidance on which of these systems to choose and that allow citizens to exert a reasonable degree of control. Instead of accepting that this would be 'disproportionate' (as the European Commission), 38 we 33 The reader should know that the criterion of a (reasonable/strong/almost irrefutable) suspicion can relate to both the assessment of the chance of success of the investigative measure, or to the possibility that the defendant has actually committed the crime of which (s)he is suspected/accused. 34 The requirement of purpose limitation serves to achieve that (particularly intrusive) interferences with, for instance, the right to privacy are allowed only for certain well-defined (serious) offences, thereby excluding those interferences for the investigation of other (less serious) offences. argue that such additional mechanisms are essential for the EU's ambition to establish an AFSJ. It is somewhat contradictory to accept that, on the one hand, the EU actively promotes the free movement of citizens, while it simultaneously defends the position that any further-reaching measures to fight crime than are currently on the table are 'disproportionate' . EU citizens -alleged offenders, victims, third parties -will only truly enjoy their rights to free movement and settlement, services, capital, et cetera (and will only be able to live up to their corresponding civic duties), if they are simultaneously protected by fundamental rights, also in a transnational setting. 39

Assuring effective judicial control in pre-trial proceedings
It has been a deliberate choice, both with respect to Eurojust and -more surprisingly 40 -the EPPO, that judicial protection against (operational) decisions of the European agencies in the pre-trial stage is located at the national level, with the competences of national courts to ask the Court of Justice for preliminary rulings. 41 As regards procedural measures, the EPPO will be considered as a national authority for the purpose of judicial protection (Article 36 original proposal). 42 As seen, the original proposal moreover holds in Article 26(4) that 'Member States shall ensure that the [most intrusive] investigative measures referred to in points (a)-(j) of paragraph 1 are subject to authorisation by the competent judicial authority of the Member State where they are to be carried out. ' Other investigative measures will be subject to judicial authorization if this is required by the national law of the relevant Member State. Finally, the EPPO may request from the competent judicial authority that it arrests or detains, on a pre-trial basis, a suspected person in accordance with national law (Article 26(7) original proposal).
There is no doubt that the choices that have been made are a significant departure from the ground rule that EU courts control the legality of actions by EU bodies, whereas national courts exert control over the actions of national authorities. 43 The Court of Justice held in Foto Frost that this system guarantees the coherence and uniformity of EU law by preventing, inter alia, national courts from delivering contradictory rulings on Union acts. 44 The Explanatory Memorandum to the EPPO original proposal justifies this departure with reference to the specific nature and different position of the EPPO (compared to other EU bodies and agencies), as all of its (investigative and prosecutorial) acts are closely related to the trial stage before national courts and its effects will therefore be felt mainly within the legal orders of the Member States. 45 What presumably plays an important role here is that, although it is not explicitly stated, national criminal courts are not (primarily) seized to rule on the validity of actions by EU bodies, but on the guilt or innocence of alleged offenders. 46 The legal basis for this departure from the EU's judicial organization is then found in Article 86(2) and (3) TFEU, stipulating, inter alia, that EPPO regulations determine the rules which are applicable to the judicial review of procedural measures taken by it in the performance of its functions.
Be this as it may, we are not so convinced that these reasons are satisfactory. No matter how much it is stressed that the legal consequences of EPPO activity are ultimately felt within the legal orders of the Member States, the fact remains that the EPPO is a European body, which is entrusted with a series of tasks that -by their very definition -cannot be clearly attributed to a single Member State. The development of prosecutorial policies, the decision to start investigations, the decisions to deploy certain investigative measures in a particular state and/or to bring criminal charges in another; these are all decisions which involve multiple Member States. The question therefore is to what extent individual national courts are in the position to effectively exert control, now that the regulation does not offer them much guidance. For instance, (how) do we prevent that EPPO officials, after having unsuccessfully tried to obtain judicial authorization for a certain measure in one Member State, consequently try their luck in another State? How do we prevent that contradictory rulings are given by different national courts on the legality of a certain investigative measure? Are courts at liberty to refuse to hear a case if they think that a trial in their jurisdiction is not in the best interest of justice? What criteria do they apply for that? What remedies are available for alleged offenders who may have a legitimate interest in prosecution in a Member State other than the one that was chosen by the EPPO? The proposal is silent on all of these issues, which will ultimately affect the legitimate interests of defendants and other EU citizens concerned. In addition, EPPO investigations may not even reach the trial stage. 47 Which legal order is competent in those instances?
One would have expected that in order to justify such a clear deviation from Foto Frost, these issues would have been explicitly addressed in the proposal. The problems occur precisely because investigative and prosecutorial decisions are taken by a body that is not national, but European, and which has been given powers to operate on a European scale. 48 It is for that reason that the system of EU judicial organization seeks to exclude gaps in judicial protection by introducing limited, but exclusive jurisdiction to European courts where the Treaties so provide. It is therefore equally difficult to accept Article 86(3) TFEU as the legal basis for such a departure. 49 Having said that, the challenge remains how to reconcile the need for coherent and uniform (ex ante and ex post) judicial control on the validity of EPPO actions (opening of investigations; investigatory measures; exercise of prosecutorial discretion) by national and/or European courts with the tasks of national courts to determine the guilt or innocence of alleged offenders in fair proceedings, as attributed to them by Article 86(2) TFEU. This not only requires a certain degree of harmonization of the tasks and powers of national courts, but also the presence of effective remedies at the European level. 50 With respect to the latter, it must be noted that the current European system may not be quite satisfactory either. Particularly with regard to annulment actions under Article 263 TFEU, the Luxembourg Court will only review measures that are capable of bringing about a distinct change in the legal position in the case of an act or decision against which an action for annulment may be brought. 51 With respect to the exercise of procedural rights, it is well known that it has been very reluctant to hear cases where it is not even clear that procedures will indeed follow. That situation has occurred many times in OLAF investigations, as OLAF has no powers to force national authorities to start investigations or commence prosecutions. A similar problem currently also exists with Eurojust. 52 However, the foregoing is not to exclude the possibility that actions for annulment may never be successful in other cases, 53 such as unwarranted infringements of civil liberties (for instance the rights to privacy, liberty or property), or situations of forum shopping. 54 Regarding the latter, Inghelram has pointed to Rendo and Others v. Commission, a competition law case also involving certain import and export restrictions, 55 in which the Commission decided to suspend proceedings under Article 85 EEC with respect to certain import restrictions and to proceed under Article 169 EEC. That also meant that the procedural rights of the former (Article 85) proceedings were (temporarily) no longer available to the applicants in the latter (Article 169). The General Court held that: 'Since the Commission's deferral has the effect of interrupting the procedure initiated under Regulation No 17 for a considerable period, it must be stated that consideration of some of the issues raised by the applicants in their complaint (…) has been taken out of that procedure, in which the applicants have specific procedural rights, and left to proceedings under Article 169 of the Treaty in which the applicants have no such rights. Whilst the procedure under Regulation No 17 is held over, the complainants will be deprived of the effective exercise of their procedural rights. ' 56 It is difficult to determine whether this particular case provides us with an argument for the thesis that forum choices (in a decentralized system of EU and -divergent -national rules) do bring about direct changes in the legal position of the persons concerned. What may have influenced the outcome in Rendo could have been that the Commission unilaterally 'switched' procedures 'during the game' . Reynolds Tobacco, however, reveals that 'forum choices' as such will certainly not always fall within the scope of Article 263 TFEU, not even where cases are transferred outside the legal order of the EU. 57 Ultimately, what remains to be seen is therefore how the EU Courts will appraise the opening of criminal investigations as such, and forum choices within that particular context, in light of their capacity to bring distinct changes in the legal position of the person affected. We are sympathetic to the idea that prosecutorial bodies should not be able to influence the scope of the rights and duties of their adversaries, or other parties concerned, such as victims.
Be all this as it may, waiting for what the European Court may possibly decide in the future with respect to Eurojust's and the EPPO's activities may not be the wisest course of action. With respect to the introduction of new EU mechanisms for pre-trial judicial control, various suggestions have been made as to how such mechanisms could best be incorporated into the existing Treaty framework. 58 Within the limitations of that framework, a system that integrates the various conflicting interests at hand will have to leave the (ex-post) review of the validity of EPPO actions with the EU courts. Simultaneously, such a system would also have to introduce mechanisms to guarantee ex-ante judicial control for certain intrusive investigative measures, as required by the ECtHR. 59 The question then is to what extent such ex-ante authorizations are to be considered as a review of the legality of acts of EU bodies, as defined in Article 263 TFEU. In our opinion, they are a (often mandatory) 60 part of a system of procedural safeguards which EU law must respect and which intends to exclude the possibility of arbitrary interferences with the fundamental rights of citizens. 61 The focus of the test is therefore on the question of whether such interferences are in accordance with the rule of law, and not on the legality of EPPO activities as such. That means that, in principle, such measures must be left in the hands of national courts (cf. also Article 274 TFEU), although preliminary references may sometimes be necessary. National trial courts, finally, would then decide on whether the materials that were gathered may be used as evidence, also in cases where those materials were produced by unlawful decisions of the EPPO. 62 That brings us to the following point of interest: the position of European citizens, particularly defendants, during the trial stage.

Testing the evidence in a fair trial
It is well known that one of the main problems in international criminal cooperation is the use of evidentiary materials that were collected abroad. 63 Despite all of the efforts that were put into increasing transnational cooperation with respect to the gathering of evidence, international or supranational rules with respect to their use barely exist. This is not only due to the complications that are related to the harmonization of the laws on evidence, 64 there are also doubts as to the necessity of such an exercise, at least as long as the scope of such a rule of 'free movement of evidence' is not clearly demarcated and defined. 65 Spencer has noted, for instance, that '[b]roadly speaking, it is surely up to each Member State to frame its own rules as to the admissibility of evidence in criminal cases; and it is not the business of the EU to tell Member States what types of evidence should or should not be admissible in their criminal courts, unless there is some compelling practical reason to require this. ' 66 Indeed, if taken to the limit, the free movement of evidence could mean that 'a written statement taken by a juge d'instruction from a witness in France (and as such admissible in French criminal proceedings) would be automatically admissible at a trial in England, irrespective of the hearsay rule, which would normally require the witness to give evidence orally; and similarly, it would be automatically admissible in Germany, in contradiction of the German counterpart of the hearsay rule, the "Grundsatz der Unmittelbarkeit". ' 67 Under such a broad interpretation, the admissibility rule would not only cover the gathering of materials, but also their transfer, as well as their use as evidence in court.
It does not really come as a surprise that the Eurojust proposal is silent on the use of evidence gathered abroad. Unsurprisingly also, the same is not true for the (original proposal for the) EPPO. Article 86(1) TFEU stipulates after all that the EPPO shall exercise the functions of a prosecutor in the competent courts of the Member States in relation to such offences. Therefore, the regulations must also contain the rules of procedure applicable to its activities and those governing the admissibility of evidence (cf. Article 86(3) TFEU; Article 4(3) original proposal). In line with that, Article 27 of the original proposal entrusts the EPP with the task of choosing, in light of the proper administration of justice and the specific criteria of Article 27(4), the competent jurisdiction and trial court. 68 That court is subsequently obliged, on the basis of Article 30, to admit the evidence presented by the EPPO without any validation or similar legal process, provided that the court considers that its admission does not adversely affect the fairness of the procedure or the rights of defence as enshrined in Articles 47 and 48 of the Charter of Fundamental Rights of the European Union. Such evidence must also be admitted in cases where the national law of the Member State where the court is located provides for different rules on the collection or presentation of such evidence.
The proposed Article 30 is the final, logical consequence of the principle of European territoriality. It is moreover supported by the partial harmonization of procedural safeguards (Article 26) and defence rights (Articles 32-35), guaranteeing a minimum degree of mutual understanding. To that extent, this proposal is also a clear deviation from the principle of forum regit actum, that is advocated in many instruments for mutual legal assistance and their successors under mutual recognition schemes. 69 Trial courts will have to accept evidence that was collected elsewhere in the European Union, also under different rules.
When compared to other studies available, what comes to mind is that the EPPO original proposal introduces quite a wide rule on the mutual admissibility of evidence. At first reading, it does not even make a distinction between evidence that was lawfully obtained and evidence obtained unlawfully, unlike, for instance, similar provisions in the relevant Council of Europe Treaties, 70 or the Corpus Juris, that contains a specific provision on this issue (Article 33). We can take it, however, that the rule only applies to materials that were gathered lawfully in another state. In the second place, Article 30 of the original proposal does not seem to differentiate between national rules on the gathering of evidentiary materials and the rules on their use as evidence. That is so because Article 30 not only refers to differences in national rules on the collection of evidence, but also on their presentation. That means that the example given by Spencer above indeed falls within the scope of Article 30.
The wide scope of the proposed solution, as well as the absence of harmonizing rules on admissibility (such as in the Corpus Juris), makes the EPPO proposal susceptible to the same criticism that we have seen in the previous sections. There is a risk of forum shopping and of deliberate circumventions of the evidentiary standards of a particular legal order. 71 Gleβ has for instance pointed out that defence rights may fall between two stools in cases of interaction between a legal order (where the materials are used during trial) that puts strong emphasis on defence rights at the pre-trial stage, and another other legal order (where the evidence was originally gathered) which relies on the trial stage for this. 72 This brings us to another remark. The question is to which extent the current proposal indeed goes far enough to prevent defence rights -and hence the legal position of citizens -from falling between those two stools. While it is true that the proposal does pay a certain amount of attention to the harmonisation of defence rights, these rights are nevertheless harmonized in the proposal only partially, and, even more importantly, a harmonisation of the content of the rights as such does not necessarily mean that they can also be effectuated in a transnational setting. This is best illustrated with reference to the right to hear witnesses (cf. Articles 48 CFR / 6(3) ECHR). 73 With respect to that right, the Strasbourg Court has consistently refused to deal with that right on a 'stand alone' basis. It always assesses it in light of the proceedings as a whole. 74 The right to hear witnesses is vital to test both the lawfulness and the reliability of evidence. In line with this, Article 35 of the original proposal rightly stipulates that the suspect and accused person will have, in accordance with national law, the right to present evidence for the consideration of the EPPO. Moreover, the suspect and the accused will have, in accordance with national law, the right to request the EPPO to gather any evidence relevant to the investigation, including appointing experts and hearing witnesses.
On the basis of Article 35, we take it that the principle of European territoriality also works, on occasion, in favour of the defendant to the extent that (s)he can avoid time-consuming mutual legal assistance procedures (or Evidence Warrants/Investigation Orders), by directly addressing -through the EPPO structure -the legal order where the evidence, witness or expert is present. 75 Nevertheless, the precise scope of this article is unclear. The reader will have noticed that the wording of Article 35 is even broader -and therefore leaves more room for ambiguity -than the defence rights contained in Articles 48 CFR and 6 ECHR. It (only) grants a defendant the right to ask the EPPO to hear witnesses. 76 Secondly, the proposal only mentions a 'right to request' , implying implicitly a certain degree of discretion for the EPPO not to grant the request, or not to specify the reasons for that decision.
Finally, and perhaps most importantly, how does the proposal guarantee that those rights are effectuated during the trial stage? In order words, how does the EPPO proposal make sure that foreign witnesses are indeed present during the trial, or that they are otherwise heard? The EPPO proposals are completely silent on the powers of the trial courts in this regard. May they use the EPPO structures for that, by instructing the competent officials? Otherwise, should the trial courts use their own MLA arrangements, evidence warrants, or, in due time, investigation orders, in order to hear the experts or witnesses they think are necessary? If so, what should be done in cases where refusal grounds apply? May the requested/executing national authorities, for instance, refuse to execute MLA requests because national investigations on related offences are ongoing? May trial courts in turn accept a certain mitigation of fair trial standards and refrain from hearing a witness or only under restrictive conditions, because transnational cooperation is still time-consuming and slow? 77 Or, on the contrary, do the concepts of European territoriality and the free movement of evidence have as their corollary that evidence that cannot be tested by the defence (and is important to the case) may not be used in trial proceedings? The EPPO proposal is silent on all of these issues.
In our opinion it is rather inconsistent to advocate the concepts of European territoriality, the free movement of evidence and fair proceedings while simultaneously accepting that the courts of the trial state lack the power to assure, for instance, the presence of witnesses during trial on grounds that are essentially related to the interests of the nation state and national sovereignty.

Concluding remarks
The rise of European agencies with operational powers in the criminal judicial area is a new phenomenon. Their added value lies in the fact that they are able to deal with issues that national authorities are unwilling or unable to deal with. Their unique institutional design allows them to operate on a European scale and to develop their own investigative and prosecutorial policies.
From our preceding analysis, we can deduce that the establishment of European judicial law enforcement agencies in the AFSJ and their institutional design and governance do not result automatically in European agencies that apply European law in an European common area, the AFSJ. When it comes to the legislative design of the empowerment of Eurojust and of the EPPO, in particular the definition of their investigative and prosecutorial powers, they are highly decentralized and integrated into the institutions and regulatory regimes (the applicable law) of every single Member State. Eurojust acts through its national members and the EPPO would act through its delegates in the Member States that would apply mostly national law.
The emphasis that is put on integration into the national legal orders of the Member States begs the question of whether the inherent European dimension of the EPPO and Eurojust is lost out of sight. The national empowerment of the European agencies is, first of all, problematic from the point of view of the agencies themselves, as they have to address the increase in transnational crimes, but also the increase in national crimes that need transnational cooperation to be investigated, prosecuted, adjudicated or executed. The European agencies have to work with a patchwork of national regimes and cultures, which results in great legal uncertainty as to the enforcement powers that can be used and their modalities. Transnational problems in the common AFSJ are still addressed -just like under the traditional instruments for mutual legal assistance in criminal matters -by an accumulation of approaches under national jurisdictions.
What is more, the national empowerment of European agencies is also problematic from the point of view of the addressees of the enforcement in criminal matters: witnesses, victims, third parties, and certainly also defendants. Their civil rights and liberties depend on discretionary choices in a regulatory patchwork, which can end up in forum shopping and a run to the bottom of the lowest protective denominator of safeguards. If so, citizens will have great difficulties in defending their rights in court, because national court organizations are not apt when it comes to their European dimension and because defence rights and procedural safeguards are still designed to be applied within the national context.