THE EVOLUTION OF THE EUROPEAN LEGAL ORDER

Legal order Common Market Law Review 41: 303–316, 2004. © 2004 Kluwer Law International. Printed in the Netherlands. 303 THE EVOLUTION OF THE EUROPEAN LEGAL ORDER FRANCIS G. JACOBS* 1. Introduction It is a privilege to give the keynote address at this conference, marking the fortieth anniversary of the Common Market Law Review; but it is at the same time a daunting challenge, especially before so expert an audience. First, I would like to pay tribute, as a regular reader, to the Common Market Law Review, to its successive editors and publishers, and not least to its Associate Editor, Alison, and her predecessors. The Review has, over the period of forty years which we are celebrating today, consistently published articles and case notes of the highest standard. The articles have themselves had, as I will suggest, an influence on the evolution of the European legal order, and should continue to do so. The expert case notes systematically survey the most important cases; they regularly compare the Opinion of the Advocate General and the judgment of the Court; they are almost invariably, as a result, illuminating in ways in which other case notes are not; and they sometimes also point the way to the future development of the law. In this opening address I am asked to give an entirely personal overview of the evolution of the European legal order. It is necessarily subjective, and indeed impressionistic; but some of the themes will be treated more fully, in this conference, by others. I am all too conscious of what I have omitted: much of the all-important historical and political context; and many highly significant subject areas, not least the rich case law on external relations. I have taken three themes: the market; individual rights; and constitutional foundations. These themes will, I believe, prove to be interrelated, as well as revealing certain common threads. * Advisory Board of this Review. Advocate General, Court of Justice of the European Communities. 304 Jacobs CML Rev. 2004 2. The market In examining the evolution of the EU legal order, it seems appropriate to take the market first. It was, and remains even today, fundamental to that evolution. It was the key concept. It came, and still comes, first in the Treaty. It was the original idea: the underlying notion of the European Communities: first the common market in coal and steel (the European Coal and Steel Community); later the common market in all other industrial (and agricultural) goods (the European Economic Community) and a separate common market in atomic energy (the European Atomic Energy Community). It was also original in a more profound sense: the remarkable originality of using the technique of market integration as a means towards achieving broader economic, and ultimately political, aims. The market well illustrates the purposive approach of the Court when interpreting the broadly worded provisions of the Treaty. It also demonstrates the role of the Court and the interaction of its far-reaching case law with EC legislation, and with Treaty amendment. Since others will cover the subject in more detail, I will content myself with an outline. In the briefest outline, the early years of the evolution of the EC legal order, especially the 1970s and 1980s, are largely the story of the case law on the market. The removal by the Court of barriers to trade sometimes induced, or even compelled, the legislature to act – but in areas such as professional qualifications, rather than on the free movement of goods. Later, and despite, or perhaps partly because of, the achievements of the case law, it was recognized that the requirement of unanimity for market legislation had to go: the Single European Act of 1986/7 accordingly amended the Treaty and introduced qualified majority voting, and paved the way, around 1990, for the ambitious programme of legislation designed to “complete” the “internal market”. Much of the case law is now devoted to the interpretation of the internal market legislation; but the legislation is often influenced by the case law. An excellent example of the interaction of legislation and case law is in the field of trade marks, where the Trade Marks Directive1 took over principles developed by the case law as part of the free movement of goods, for example the concept of exhaustion of rights, and where the interpretation of the legislation is still guided by the fundamental principles on the free movement of goods. Thus the Court’s tasks, although now often concerned with interpreting 1. First Council Directive 89/104/EEC of 21 Dec. 1988 to approximate the laws of the Member States relating to trade marks, O.J. 1989, L 40/1. Legal order 305 legislation, have not greatly changed, and its approach has remained generally consistent. It may be sufficient here to emphasize, without elaboration, the following points: First, the fundamental criterion of market access, not always consistent, and only occasionally articulated, in the case law; but generally reflected, in practice, in the Court’s approach. Second, the use of the concept of the “internal market”, in effect almost as an economic model. Third, the virtues of mutual recognition in preference to harmonization as a guiding concept advanced by the Court in Cassis de Dijon,2 and subsequently taken up by the Community legislature. In some quarters the approved term is now “equivalence recognition”. Fourth, for the user of the system, the availability, and the effectiveness, of the reference procedure as a means of removing obstacles to market access. And fifth, the contribution of legal scholarship – and not least of the Common Market Law Review.3 Mention should also be made of the scope for further development of the case law: (a) developing a coherent theory of limitations on the fundamental economic freedoms; (b) assimilating the express Treaty exceptions and the case law exceptions in a coherent framework; (c) confronting and resolving conflicts e.g. between the free movement of goods and environmental protection. It need hardly be mentioned that there is scope here for further contributions of legal scholarship! As elsewhere, the question constantly arises of the limits of Article 28. At one level, there is the need to balance the free movement of goods with the express exceptions in Article 30. At another level, there is the need to reconcile the Treaty provisions with the Court’s own category of public-interest objectives which, under the Cassis de Dijon case law, may take precedence over the free movement of goods. More generally, the case law, the legislation and progressive Treaty amendments seek to delimit the contours of the market: to strike a balance between deregulation and other values – with an evolving recognition, for 2. Case 120/78, Rewe-Zentral (Cassis de Dijon), [1979] ECR 649. 3. To take just one example (however invidious that may be), it is often suggested that the article by Eric White, “In search of the limits to Article 30 of the EEC Treaty” 26 CML Rev. (1989), 235, has influenced the Court’s case law. 306 Jacobs CML Rev. 2004 example, of the importance of social policy and of public services. The search for limiting principles, defining the boundaries of the openended categories of the Treaty, is a recurring and developing topic. If I have concentrated on the free movement of goods, that is not to underestimate other aspects of the market. In some respects, goods led the way and other freedoms have followed: notably the current programme for liberalization of financial services and the development of a capital market. The free movement of persons can appropriately be considered under the second theme: individual rights. 3. Individual rights The second theme is that EU law creates rights for individuals. Three aspects may be considered, however briefly: direct effect; the economic freedoms of the Treaty and their perception as fundamental freedoms; and fundamental rights in the classical sense of civil and political rights. However, this second theme of individual rights cannot be considered separately from the first theme, the market, and the third theme, constitutional foundations. This will be apparent as soon as we look a little more closely at these three aspects. 3.1. Direct effect The first aspect, and perhaps the key aspect, is direct effect: a provision of EC law can be applied by national courts, even in the absence of implementing national provisions. (This is of course closely related, as a concept, to the concept of self-executing treaty provisions, but it has been taken further by the Court.) It is appropriate, since we are looking at 40 years of evolution, that it was 40 years ago, in Van Gend en Loos,4 that the Court took the view that the Treaty was not merely the affair of States and Community institutions but conferred rights on individuals. Thus the Treaty enables individuals to rely on EC law and (in conjunction with primacy) to challenge national measures which prevent or obstruct the exercise of “Community rights”. As was apparent from the terms of the Court’s ruling in that case, and has been spelt out in subsequent case law, the conferment of rights is one consequence of direct effect; it is not identical with it. Direct effect has other manifestations also. 4. Case 26/62, [1963] ECR 1. Legal order 307 At least in the early years, the context of the principle of direct effect was the market freedoms: indeed it was perhaps appropriate, given the structure of the Treaty (starting with the elimination of customs duties, then the elimination of quotas and measures having equivalent effect) that the first case concerned the stand-still clause on customs duties. But there is also a close connection with the constitutional foundations of the Community. Direct effect and primacy are often (and perhaps rightly) considered together as constituting the essential characteristics of EC law.5 In looking briefly at the evolution of direct effect, we may focus on three aspects. First, it was applied initially to negative obligations imposed by the Treaty on Member States – the stand-still clause in relation to tariffs being a perfect example. To illustrate the dramatic relaxation in the conditions for direct effect, as regards the Treaty, it is sufficient to cite Defrenne II,6 a mere 13 years after Van Gend en Loos. There, direct effect was accorded to a “principle”, the principle that men and women should receive equal pay for equal work; and a principle of which Member States were required by Article 119 only to ensure and maintain the application. The obligation was positive rather than negative; its content, although in part defined by the Article, left many questions unanswered; and the provision was held to have direct effect, not only against the State, but against employers generally. So evolutionary, or even revolutionary, was the decision that it was found necessary to qualify its effect by the unprecedented step of limiting its temporal application by excluding all new claims in respect of past periods of employment. The Court thereby recognized the importance of legal certainty and sought to reconcile it with the principle of equality between the sexes. Secondly, direct effect was extended very broadly to other instruments than the Treaty: most notably, to directives, and to treaties other than the Community treaties. Third, in recognizing the direct effect of directives – a radical innovation – the Court has sought to draw a line by admitting only vertical, and not horizontal, direct effect. Yet to compensate for the lack of horizontal direct effect, the Court has employed various techniques which have progressively minimized any consequential loss of effectiveness. Any evaluation, however summary, of the implications of direct effect would have to recognize two basic points. On the one hand, it has clearly proved fundamental in creating the EU legal order and, indeed, in transforming the treaty from a classical instrument of international law into (or to5. Opinion 1/91, [1991] ECR 1-6079, para 21. 6. Case 43/75, [1976] ECR 455. 308 Jacobs CML Rev. 2004 wards) the constitution of a quasi-federal organism. On the other hand, the Court’s approach, historically at any rate, has not been to promote the rights of individuals for their own sake or as a matter of ideology; its approach has been essentially pragmatic and the recognition of individual rights has been almost instrumental, being seen as necessary to ensure the effectiveness of the legal order. That is, at any rate, the way in which direct effect is explicitly justified in Van Gend en Loos itself. The underlying notion is effectiveness. 3.2. Economic freedoms The second aspect is the economic freedoms of the Treaty and their perception as fundamental freedoms. With the assistance of direct effect, the freedoms set out in the Treaty – the free movement of goods, the free movement of workers, the right of establishment, the freedom to provide services, and the free movement of capital – are transformed into individual rights. Traditionally these freedoms were, and are, also found in commercial treaties, bilateral and multilateral (including the GATT). In the EC legal order, they become rights which can be invoked by individuals and enterprises in the national courts – very often in practice not their own courts, but the courts of the State against which they seek to enforce those rights. And the content of those rights is defined not by those (“foreign”) courts, but, where necessary or appropriate, by the ECJ itself. Thus the rights are, typically, broadly construed. The best illustration is in the area of personal rights of free movement. In the early years – the first ten years of the period we are considering – the Court’s main contribution in this field was to protect imaginatively the social security rights of migrant workers under virtually the first EEC legislation – Regulations 3 and 4. Later the Court, again interpreting broadly the notion of “worker” and the substantive content of the rights of workers under the Treaty and the legislation, paved the way for the development of the concept of Community citizenship. In turn, legislation such as the 1990 Directives7 broadened the personal scope of these rights. When the concept of citizenship of the EU was formally introduced by the Treaty on European Union, it merely consolidated the status quo: the rights of EU citizens were those which they already enjoyed under the Treaties and the legislation. But the introduction of the concept of EU citizenship encouraged the Court to give it a 7. Council Directive 90/364/EEC of 28 June 1990 on the right of residence, O.J. 1990, L 180/26; Council Directive 90/365/EEC of 28 June 1990 on the right of residence for employees and self-employed persons who have ceased their occupational activity, O.J. 1990, L 180/28; Council Directive 90/366/EEC of 28 June 1990 on the right of residence for students, O.J. 1990, L 180/30. Legal order 309 more substantive content, even if, hitherto, only in a rather tentative way. Once again, the concept illustrates the mutual interaction of Treaty change and development of the case law. 3.3. Fundamental rights Independently of the economic freedoms contained in the Treaty, the Court developed the principle of respect for fundamental rights of the classic kind – often termed “civil and political rights” – on which the Treaty was silent. Again the historical development is striking. In the early years of our period, the Court is cautious, even negative, seeing an invocation of fundamental rights protected by national constitutions as a threat to the primacy of EC law. But around 1970, in response to pressure from national courts – pressure which illustrates the reality of the “judicial dialogue” engendered by references for preliminary rulings – the Court declares that a guarantee of fundamental rights can be derived from the Treaty, so that the Court itself can determine the contours of protection within the EC legal order – and the effect is to reduce the risk of challenge to the primacy of EC law, so that the Court’s approach might again be seen as inspired in part by concerns for effectiveness and by pragmatic considerations. In any event, it must be a welcome development that the Court can itself use fundamental rights as a guide to the interpretation of EC legislation and, if necessary, in controlling its validity. The subject is, in various ways, a good illustration of the evolution of the EU legal order. It shows the Court’s initial caution, followed by an imaginative development of case law principles to supply the omissions of the Treaty. Similarly, the Court was at first cautious in its references to the benchmark of human rights protection in Europe, the European Convention on Human Rights; in recent years, such references, and indeed references to the case law of the Strasbourg Court, have become more frequent. The subject also illustrates the interaction of the Court’s case law with Treaty amendment: what is now Article 6(2) of the Treaty on European Union takes over, almost verbatim, the case law of the Court in stating that the Union shall respect fundamental rights, as guaranteed by the Convention and as they result from the constitutional traditions common to the Member States, as general principles of Community law. Similarly in relation to the EU’s own projected catalogue, the EU Charter of Fundamental Rights, the Court has shown characteristic caution and has not yet cited the Charter in any of its judgments, although the Charter has been cited by Advocates General and by the Court of First Instance. It is appropriate to mention in this context that an essential reason for cre- 310 Jacobs CML Rev. 2004 ating the Court of First Instance, by a Decision taken pursuant to the Single European Act, was, according to the preamble to the Decision, in order to “improve the judicial protection of individual interests”. The establishment of the Court of First Instance in 1989 was the single most significant change in the Community’s judicial system in the past forty years. Its jurisdiction, initially limited to competition cases, coal and steel cases, and staff cases, has been progressively extended, and the amendments to the EC Treaty made by the Treaty of Nice now recognize it as not merely “attached to the Court of Justice” but as being responsible, together with the Court of Justice, for the observance of the law in the interpretation and application of the Treaty. 4. Constitutional foundations Three topics may be mentioned, all of them perhaps characteristic of a constitutional order: 1. Preserving a balance of power. 2. Developing, in the absence of a formal constitution, fundamental principles governing the legal order. 3. Ensuring the fullest scope for judicial review – so that the Community/ Union is, as far as possible, based on the rule of law. 4.1. Preserving a balance of power Balance of power, within the EU, involves both institutional balance – a balance among the political institutions (Parliament, Council and Commission) (and to a lesser extent other Union bodies) – and balance between institutions and Member States. As regards the former – which is covered in more detail elsewhere8 – it may be sufficient here to refer to the Court’s principle of “institutional balance”. The notion itself is not novel. Indeed it has been pointed out, in the context of the Community’s evolving polity, that the concept was an important part of republican discourse in the fifteenth and sixteenth centuries,9 and no doubt it can be traced still further back. But it was put to novel use by the Court. Twenty-five years ago, for example, when the Parliament had very 8. See the contribution by Jean Paul Jacqué in this Review. 9. See Craig, “The nature of the Community: Integration, democracy and legitimacy”, in Craig and de Búrca, The Evolution of EU Law (OUP, 1999), p. 1 at 38 and note 137. Legal order 311 limited powers, the Court showed itself concerned to protect their exercise to the full in the “Isoglucose” cases.10 Subsequently a concern for institutional balance led the Court to re-write the Treaty, first to enable the Parliament to be sued, and next to allow the Parliament to sue, despite the absence of any Treaty provisions to that effect. Again, the interaction of the case law with Treaty amendment is illustrated by the Treaty amendments which, after the event, incorporated the Court’s innovations on both points into the Treaty text. On preserving a balance between the Institutions and the Member States, perhaps the most significant aspect is the concept of the primacy of EU law and the contours of that primacy. Forty years ago, shortly after Van Gend en Loos, the Court laid down, in Costa v. ENEL,11 the primacy of EC law. The notion itself was not novel: under international law, it is a commonplace that treaties prevail over national law. What was novel was that the Court should lay it down as a principle to be applied by the courts of the Member States – as indeed was the case for direct effect in Van Gend en Loos. And as in Van Gend en Loos, there was a subsequent evolution: the principle, first applied to the Treaty (where it could justifiably be said to be inherent in the very system of the Community), was subsequently extended to all forms of EC legislation, with the consequence that every provision of EC law, however lowly, prevails over every provision of national law. Moreover the principle had spill-over effects on national constitutions, illustrated for example in the ruling in Simmenthal12 in 1978 that national courts were under a duty themselves to disapply national provisions which conflicted with EC law, and must not wait for the national constitutional court to set the national provisions aside. Nonetheless primacy, although far-reaching, is not absolute. EC law is not to be applied by national courts in all circumstances. In brief, there are limits on direct effect; there are limits imposed by respect for the procedural autonomy of the legal orders of the Member States; and a national court is not bound in all circumstances to invoke EC law of its own motion. Closely related to the notion of primacy is the notion of preemption. The notion of preemption has not been immoderately used by the Court. The Court has been cautious in investing the Community with exclusive competence – some would say, in the field of external trade, excessively cautious. In areas where the competence is not exclusive, but where the Community 10. Case 138/79, Roquette, [1980] ECR 3333 and Case 139/79, Maizena, [1980] ECR 3393. 11. Case 6/64, [1964] ECR 585. 12. Case 106/77, [1978] ECR 629. 312 Jacobs CML Rev. 2004 has acted, it is not excessive to hold that Member States no longer have competence to act. A final word on balance between the Institutions and the Member States: it may be worth mentioning just one point concerning the all-important question of the standard of review by the Court of Community measures and Member State measures respectively. It is sometimes suggested that the Court has employed a double standard, especially for example when applying the flexible principle of proportionality – being more ready to strike down Member State measures than Community measures. However such criticisms sometimes overlook that what is at stake is very different. Member State measures, diverging from State to State, are by their nature liable to cause obstacles to trade. Community measures will be designed to eliminate such obstacles. They must therefore be assessed differently. Nonetheless, where Community measures, however beneficial their aim, impose a disproportionate burden on traders or others, they must be struck down. 4.2. Fundamental principles governing the legal order The principle of proportionality is of course only one example of the rich corpus of constitutional and administrative law principles which the Court has developed: other examples, as well as the protection of fundamental rights, already mentioned, are equal treatment, legal certainty and legitimate expectations, the right to a fair hearing, and legal professional privilege. It could even be said that it is largely by the use of such principles that the Court has been able to fashion the Treaties and Community legislation into a coherent legal order. At the same time, the principles provide the best illustration of the interaction of Community law and national law: in developing the law in this way, the Court has been able to draw on the diverse legal cultures of the Member States. Here too, as with the scope of the fundamental economic freedoms and the scope of fundamental rights, the Court is still searching for the appropriate boundaries. What is the scope of the protection of fundamental rights in Community law? Do the ERT13 and Akrich14 judgments go too far? What is the scope of the prohibition of discrimination on grounds of nationality? And – perhaps a closely related question – what are the contours of citizenship of the European Union? 13. Case C-260/89, [1991] ECR I-2925. 14. Case C-109/01, judgment of 23 Sept. 2003, nyr. Legal order 313 4.3. Judicial review The scope of judicial review is perhaps the most fundamental issue of all, since upon it depends the legal order itself. It is therefore worth examining rather more fully the evolution of the law in this area. In 1986 in Les Verts,15 the Court was able to say that “the Community is a Community based on the rule of law, in which all measures taken by the institutions and the Member States are subject to judicial review”. This principle expressed a fundamental theme running through the case law. This part of the story is particularly familiar. Thus, the Court has taken a generous view of the types of acts which are susceptible to review. Under the first paragraph of Article 173 of the EEC Treaty (now Article 230 EC), the Court was originally competent to review “acts of the Council and the Commission other than recommendations and opinions”. Article 189 of the EEC Treaty (now Article 249 EC) defined binding Community acts as regulations, directives and decisions. It might have been thought, on the basis of those provisions, that the Court was only competent to review regulations, directives and decisions adopted by the Council or the Commission. However, in AETR,16 the Court was willing to review the legality of Council proceedings regarding the negotiation and conclusion by the Member States of an agreement on the working conditions of the crews of vehicles engaged in international road transport,17 on the ground – essentially – that the purpose of the procedure for judicial review laid down in Article 173 of the EEC Treaty (which is to ensure observance of the law in the interpretation and application of the Treaty) would not be fulfilled unless it was possible to challenge all measures, whatever their nature or form, which are intended to have legal effects.18 In Les Verts,19 the Court was asked to review two measures, adopted by the European Parliament, on the reimbursement of expenses incurred by parties taking part in the 1984 elections. In declaring that action admissible, it held that while “Article 173 refers only to acts of the Council and the Commission ... an interpretation of [that provision] which excluded measures adopted by the European Parliament from those which could be contested would lead to a result contrary to both the spirit of the Treaty as expressed in Article 164 [now Article 220 EC] and to its system”.20 15. Case 294/83, [1986] ECR 1339. 16. Case 22/70, Commission v. Council, [1971] ECR 263, paras. 39 to 42 of the judgment. 17. The European Road Transport Agreement. 18. For an application of that principle to a Commission Communication, see Case C-57/ 95, France v. Commission, [1997] ECR I-1627. 19. Cited supra note 15, paras. 24 and 25 of the judgment. 20. See also Case 2/88, Zwartveld [1990] ECR I-3365, paras. 23 and 24 of the judgment. 314 Jacobs CML Rev. 2004 When deciding which institutions are entitled to bring proceedings for annulment under the Treaty, the Court has not adopted a strict reading of the Treaty text either. Prior to the entry into force of the Treaty on European Union, the first paragraph of Article 173 of the EEC Treaty provided that the Court had jurisdiction “in actions brought by a Member State, the Council or the Commission”. The absence of any reference to the European Parliament in that provision did not, however, prevent the Court from holding in Chernobyl21 that “an action for annulment brought by the Parliament against an act of the Council or the Commission is admissible provided that the action seeks only to safeguard its prerogatives”,22 for while “[t]he absence in the Treaties of any provision giving the Parliament the right to bring an action for annulment may constitute a procedural gap, ... it cannot prevail over the fundamental interest in the maintenance and observance of the institutional balance laid down in the Treaties”.23 The only exception to the Court’s remarkably broad approach to judicial review concerns standing for individuals to challenge a measure directly before the Court (now the Court of First Instance), where the Court has essentially, if not totally, maintained the very narrow test of standing laid down in Plaumann24 forty years ago. Here there has been little evolution over the 40year period we are considering. But in the UPA25 judgment in 2002 the Court did draw attention to the possibility of Treaty amendment, and the 2003 Draft Constitution, if it enters into force, would amend the standing rule: although the amendments are not ideally drafted, they go a long way to remedy the defects in this important area of judicial protection. There must of course be limits against abuse of judicial process, but these can be fashioned by the Community courts, and must not be a basis for depriving the individual of legitimate protection of his rights and interests. After all, the individual should be at the centre of the grand European design, not an unwelcome intruder. The Court’s restrictive approach to standing provides a striking contrast to its expansive case law on judicial review and remedies in the national courts. Here the Court has been bold, and its case law in this area can certainly be viewed as part of the constitutional foundations of the Community, and forms an appropriate coda to this third theme. The starting-point was the principle that national courts must afford individuals remedies which are effective, and which are in any event equivalent 21. 22. 23. 24. 25. Case C-70/88, European Parliament v. Council, [1990] ECR I-2041. Para 27. Para 26. Case 25/62, [1963] ECR 199. Case C-50/00 P, Unión de Pequeños Agricultores v. Council, [2002] ECR I-6677. Legal order 315 to those provided for corresponding claims under national law. The principle as it has evolved has made significant incursions into national legal systems and even national constitutional rules. Among the clearest illustrations are: (1) the Simmenthal case, already mentioned, in 1978, in which the Court held that, where national legislation conflicts with Community law, a national court must give effect to the Community provisions and must not wait for the national law to be set aside either by a constitutional court or by the legislature. (2) Factortame II26 in 1990: where, in a case involving Community law, a national court considers that the sole obstacle to the granting of interim relief is a rule of national law, Community law requires it to set aside that rule. (3) Francovich27 in 1991: Member States are liable to compensate individuals for loss caused to them by a violation of Community law for which the Member State is responsible. 5. Conclusions As already suggested, there are some common threads in the above survey. First and most obviously, it demonstrates the importance of case law in the evolution of the European legal order. It illustrates also the systematic and purposive character of the Court’s interpretative techniques, and its use in particular of the principle of effectiveness. Although initially cautious, the Court has pursued a consistently purposive approach, and has developed European law, by incremental stages, in such a way as to secure the objectives which it saw as fundamental, thereby frequently supplying the omissions of both the Treaty and the Community legislature. The systematic character of the Court’s approach is apparent from the connections between each of the themes considered. The cooperative relationship between national and Community judiciaries has played a vital role: the national courts have been willing to identify key questions and to refer them to the Court of Justice, and although not all constitutional courts have fully accepted the claims of EC law, the dialogue has so far proved effective in avoiding instances of conflict between EC law and national constitutional law. Second, the survey illustrates the interaction of case law, legislation and Treaty amendment. On the one hand, the case law has supplemented the written sources and filled the gaps; on the other hand, it has sometimes also 26. Case C-213/89, [1990] ECR I-2433. 27. Cases C-6 & 9/90, [1991] ECR I-5357. 316 Jacobs CML Rev. 2004 provided the inspiration for legislative developments and for Treaty amendment. Moreover, through judicial dialogue and the system of preliminary rulings, EC law has also interacted with the national law of the Member States. Principles of national law have fertilized the European legal order, and the concerns of national systems and national courts can be taken into account. Third, the Court’s case law can often be seen as a search for a coherent set of boundaries or limiting principles for the concepts which it has boldly developed. There may here be an element of uncertainty, but perhaps no more than is necessary in exploring frontiers, where occasionally the Court takes a step too far and may then find it prudent to retreat. The overall pattern however has remained progressive, and is indeed well expressed in the concept of “evolution” (itself also embodying an occasional retreat in an overall progression) in the title allotted to me. A final conclusion gives cause for concern. Judicial review, as we have seen, is at the heart of the European legal order, and it is only through judicial review that the European Union can be subject to the rule of law. Yet past Treaty amendments have created large judicial no-go areas, in which the Court’s jurisdiction has been limited or entirely excluded. The jurisdiction of the Court is now largely confined to the first pillar, with a somewhat arbitrary patchwork in certain other areas. The current pattern raises a fundamental question. How far can it be said today, as it could be said of the Community in Les Verts, that the European Union is a Union based on the rule of law? The 2003 Draft Constitution, which is still under discussion at the time of writing, would fill many of the gaps. First, it would abolish the three-pillar structure, so that the Court’s jurisdiction would extend broadly, with limited exceptions, to the area of freedom, security and justice, and to a certain extent to the common foreign and security policy. Second, judicial review would not be limited to measures adopted by the institutions, but would extend to all EU bodies. Third, it would confer greater standing on individuals, at least to challenge regulatory measures, if not primary legislation. I would welcome these developments – which indeed correspond largely to my own agenda for the Union’s judicial system. But these proposals also serve to highlight the serious holes in the system as it stands if they are not adopted. A failure to fill the gaps would halt the evolution of the European legal order as we have seen it develop over the past forty years; it would mean that the Union could not be said to be based on the rule of law; and it might raise grave doubts about the very legitimacy of the Union.