Jean Monnet Center at NYU School of Law


Evolution & Effort:

The Development of a Strategy of Docket Control for the European Court of Justice & the Question of Preliminary References

Sarah E. Strasser [*]


©Copyright: Sarah E. Strasser, 1995.


Table of Contents

INTRODUCTION
I. FACTUAL PARAMETERS

A. VOLUME OF CASES BEFORE THE COURT OF JUSTICE

1. CASES BROUGHT

2. CASES DECIDED

B. DURATION OF CASES BEFORE THE COURT OF JUSTICE

1. INTERLUDE: A DAY IN THE LIFE OF A PRELIMINARY REFERENCE

2. STATISTICS ON DURATION

II. POLICY CONSIDERATIONS
III. POSSIBILITIES FOR DOCKET CONTROL

A. VOLUME REDUCTION

1. INCREASED DISCRETION FOR NATIONAL COURTS?

2. LIMITING COURTS WHICH MAY REFER

3. SELECTION OF CASES BY THE COURT OF JUSTICE: THE END OF MANDATORY JURISDICTION FOR PRELIMINARY REFERENCES?

B. CASE MANAGEMENT

1. IN THE NATIONAL COURTS

2. IN THE COURT OF JUSTICE: PROCEDURE

3. IN THE COURT OF JUSTICE: STRUCTURE

CONCLUSION
APPENDIX I: BLUEPRINTS FOR REFORM
APPENDIX II:SUMMARY OF OPTIONS OUTLINED IN PART III
APPENDIX III: JUDGMENTS DELIVERED BY THE COURT OF FIRST INSTANCE
APPENDIX IV: RATIO OF CASE VOLUME TO NUMBER OF MEMBER STATES
APPENDIX V: NATIONAL REFERRALS, COMPARED TO POPULATION & GNP
SOURCES


Introduction

The eve of the 1996 intergovernmental conference is a particularly appropriate point at which to consider reforms of Community Institutions, not least the Court of Justice. Indeed, litigation patterns in the Community suggest that such an examination is becoming increasingly urgent, as the volume of cases brought to the Court steadily expands, and delays in adjudication accordingly increase. Indeed, the Court cannot continue to hear all cases brought to its attention. If delays are allowed to rise, the legitimacy of the Court itself will be endangered. It has been said of the European Commission that "the mills of Brussels grind very slowly indeed." [1] The same might be said of Luxembourg, especially from the view of an outsider. The purpose of this study is to examine the trends which contribute to the duration of adjudication, and to suggest ways in which the Court and the Community might improve the situation.

This is not to say that the increased demands on the Court of Justice do not also indicate positive developments. The rising "popularity" of the Court as a forum is Janus-faced in its causes and effects. In 1971, fourteen years after the adoption of the Treaty of Rome, the Court observed that the "distinct increase in the volume of litigation" in that year, "apparent in all spheres of Community judicial activity...undoubtedly suggests that Community law is coming to have a practical impact". [2] But at the same time, the Court has become the "victim of its own success," as the very number of cases which have proved the acceptance of its jurisdiction have put the Court under a great deal of pressure. [3] It is this dichotomy which must be addressed. The growing case load has contributed to mounting and significant delays, despite the transfer of entire categories of cases to the Court of First Instance. [4] Whether the Court system is equal to the demands of the Community, both procedurally and structurally, is a question naturally arising from this state of affairs.

These concerns are particularly relevant to the Court's treatment of cases brought under Article 177 of the Treaty, which provides for interlocutory actions in the European Court of Justice to suits brought before Member State courts and tribunals. In these cases, the Court considers questions of Community law which are thought essential to determining a national issue. [5] Just as the rising number of cases brought to the Court illustrate the growing influence of Community law, so does the rise of preliminary references. Moreover, given the extent of delays and case backlog endemic to the Article 177 system, these actions provide a useful perspective from which to reconsider the Community judicial system. [6] For Article 177 is the Court's primary means of exercising judicial control, as it forms an important link between Member State and Community legal systems. The Court of Justice, through the preliminary reference process, is able to give authoritative rulings on the meaning and effect of the Treaty and other instruments of Community law, as long as a national court recognizes such a ruling is essential to the resolution of a domestic dispute. In turn, Article 177 gives national courts a means of "removing the obstacles which inhibit proper application of Community law within the national framework," by enabling them to ask for official interpretations of unfamiliar but relevant laws. [7]

If the smooth functioning of the reference procedure is essential to the efficacy of the Community legal system, then significant delays in the delivery of rulings pose a potentially serious problem. Preliminary references are merely interlocutory to the main action; once the Court of Justice delivers an answer, the domestic court must resume its own adjudicatory process. This fact makes delays in Luxembourg all the more problematic. Since the preliminary reference procedure is meant to function through a co-operative process between national courts and the Court of Justice, the obligation to minimize delays thus falls on both national courts and the Court of Justice. [8] Mutual responsibilities apply to each: national courts must refer a clear and justiciable question, and the Court must render a helpful ruling. [9]

In tension with the effort to encourage centralized Community law and a constructive national-Community court relationship, are concerns of efficiency and prompt adjudication, however. In 1993, the Court took on average over twenty months to deliver preliminary rulings. [10] This delay has had some serious consequences, causing some Member State courts to go so far as to refrain from making references, in situations which truly demand Luxembourg level deliberations. [11] For example, in the Portman Agrochemicals case, Mr. Justice Brooke of the English High Court, while admitting he was unsure how to interpret a question of Community law, chose not to make a reference. As both parties to the domestic litigation pointed out, and Brooke J agreed, the usual time spent by the Court to make preliminary rulings would ensure that the domestic case was moot by the time the Court delivered its decision. [12]

As Portman Agrochemicals demonstrates, these delays have a direct impact on the uniformity of interpretation and application of Community law, which Article 177 is meant to guarantee. Efficiency -- defined here as the ability to deliver answers within a reasonable period of time -- is thus closely bound up with the effectiveness of those rulings and of the Court itself. If national courts are discouraged from making references because the judicial process is too heavy, the Court will be denied the opportunity to consider what may be important questions. For the reference procedure is a voluntary one, without any "hard" compliance features. The obligation of national courts to make references is wholly within the bounds of their own discretion, and thus should be fostered to retain the usefulness and integrity of the system. The problem of delay is not simply the fault of the Community judiciary, however. As will be discussed below, national courts are also responsible for the growing case load, and thus the delays plaguing the Community legal system. [13]

This picture is complicated by the fact that the problem of delay is not static. The expansion of the Union and the addition of substantive areas to the Court's jurisdiction cannot but exacerbate the present situation. [14] And the dangers of delay are not limited to the element of time. As the ability of the Court to deliver decisions promptly drops, so shall the normative impact of those decisions; the German Bundesverfassungsgericht, the English House of Lords, and the United States Supreme Court demand authority in part due to their limited output, which prevents such long waiting periods. [15] The experiences of other supreme courts has shown that once the number of cases exceeds a certain, indeterminate level, the ability of lower courts to absorb the jurisprudence declines, the perceived significance of each decision drops, and the respect for and reputation of the supreme court as the ultimate arbiter of the law disintegrates. [16] This process may have already begun to take hold of the Court of Justice. [17]
The purpose of this paper is not to suggest that the Court has failed to take any actions to remedy the problems of case backlog and delay. Indeed, the contrary is mor accurate. As time has gone by, the Court has changed its method of approaching references, interpreting the facts and national law presented in references more strictly when making rulings. This trend has accordingly resulted in "a tightening of the conditions imposed on national courts when they refer questions," so to assist the Court to render a helpful and informative ruling. [18] The Court has also tightened its internal procedure, and has given special priority to the expedition of preliminary rulings. Nevertheless, there are still areas in which the Court's approach could be adjusted, to further reduce the amount of time spent in adjudication. The reforms discussed below take three basic forms: targeting case volume, or focusing on the management of cases at either the national or the Community level. Under each of these headings, a number of options is available. No single approach provides a magic answer to the Court's problem; in all likelihood, the most successful solution would incorporate a combination of the reforms suggested below. As the adoption of any single option or combination thereof depends on political reality, this study highlights policy concerns associated with these reforms, as well as evaluating the practical efficacy of each alternative.

The following analysis is divided into three sections. Part I lays out the background data, focusing on the actual numbers of cases brought to and decided in the Court and the Tribunal. Part II briefly discusses the policy considerations invoked by docket control in the Court. Part III, which forms the bulk of this study, systematically presents and analyzes the numerous areas in which control of the Court's case load might be asserted. The mechanisms required, as well as the possible success of these alternatives, form part of this discussion. Part III is intended as a guide, not a prescription. Specific blueprints for reform are thus suggested in an appendix to the main study, rather than in the main text.

Part I
FACTUAL PARAMETERS: CASE LOAD & DURATION OF PRELIMINARY REFERENCES IN THE EUROPEAN COURT OF JUSTICE

Given the enlargement of the Community, and the deepening and broadening of its legal system, it is unlikely that the increase in the Court's case load is a temporary phenomenon. [19] In all likelihood, the Court's burden will become greater in the future, through an expansion in the volume of questions from national courts on a growing number of issues, and by the growing complexity of the legal areas themselves. Should developments continue as they have begun, the duration of cases will continue to rise as well. To maintain the effectiveness of the Court of Justice and of the Community legal system, the eventual adoption of a policy of docket control is not only advisable, but essential.

To consider the direction such a policy might take, the past experiences of the Court of Justice should be considered. Predictions of the success or failure of proposed schemes depend on how different circumstances influence the Court's ability to control its docket. The accession of new Member States, the creation of the Court of First Instance and its progressively expanded competence, the increased use of the preliminary reference and other actions by Member State courts and Member States, and the demands posed by the procedures themselves, are among those factors that have influenced the ability of the Court of Justice to fulfill its functions. These experiences may be extrapolated from, to suggest how the system might respond to other changes. The interaction of these factors with specific schemes for reform will be discussed below, in Part III. The topic of this section is the history of case volume and duration in the Court. (Table A) This information will prepare the ground for a discussion of methods of docket control, by forming the factual boundaries of that consideration.

TABLE A
ACTIVITY OF THE COURT OF JUSTICE, 1960-1993 [20]

Year

Cases brought

judgmnts given

average duration (months)

177

total

177

direct actions

total

177

direct actions

1960

0

28

0

14

20

1961

1

27

0

8

12

1962

5

36

1

18

20

1963

6

111

2

13

17

1964

6

58

7

6

31

1965

7

62

7

17

52

1966

1

31

4

12

25

1967

23

37

12

6

24

1968

9

33

15

6

26

1969

17

77

13

6

30

1970

32

80

28

19

64

1971

37

97

28

14

60

5.5

8.5

1972

40

82

30

15

61

5.5

11

1973

61

192

53

11

82

6

10

1974

39

102

41

8

63

6

9.5

1975

69

131

45

13

79

6

9

1976

75

127

53

11

88

6

9

1977

84

158

75

12

103

6

9

1978

123

270

62

21

98

6

18

1979

106

159 [21]

83

37

139

9

18

1980

99

280

75

34

132

9

18

1981

109

323

65

21

28

12

12

1982

129

348

94

59

185

12

13

1983

98

299

58

53

151

12

14

1984

129

312

77

57

165

14

17

1985

139

434

109

64

211

14

20

1986

91

329

78

61

174

15

20

1987

144

396

71

100

208

18

22

1988

179

375

108

98

238

17.5

23

1989

139

385

90

64

188

17

23

1990

141

381

113

73

193

17.4

25.5

1991

186

342

108

90

205

18.5

24.2

1992

162

440

112

89

211

18.8

25.8

1993

203

486

128

69

204

20.4

22.9

1994

203

354

119

53

188

18.0

20.8

A. Volume of cases before the Court of Justice

The total case load of the Court is increasing, as reflected by the steady growth of cases pending before the Court. [22] Factual considerations of case volume may be broken down into two main groups: cases brought and cases decided. Within those categories, the figures relating to the total number of cases and to Article 177 actions may be distinguished.

1. Cases brought

Overall, the number of cases brought before the European Court of Justice has increased steadily since the 1950s. (Figure 1) In 1960, the total number of cases brought before the Court was 28; in 1970, the total number increased to 80; in 1980, the number had more than tripled over the decade, to 280; and in 1990, reached a total of 381. In 1993, a total of 486 cases was brought before the Court, reaching a volume of over seventeen times that in 1960. [23]

Despite the continuing increase in the total number of cases brought before the Court of Justice, the proportion of cases brought per Member State, particularly over recent years, has remained fairly constant. (Figure 2) The expansion of the Community to nine Member States in 1973, ten in 1981, and twelve in 1986 does not seem to have been a significant factor in the percentage increase of cases brought. After initial jumps after each wave of accession, the number of references brought by each Member State has remained fairly
constant, even if time is allowed for national courts to become accustomed to the concept and use of Article 177. If one assumes that seven years is sufficient for national courts to reach their "normal" usage of Article 177, a sense of case volume increases caused by more than the simple enlargement of the Union may be gained. Accordingly, the ratio of case volume to the number of Member States in 1974, 1980, 1988, and 1993 was as follows: 11.3, 31.1, 31.2, and 40.5. [24]

The volume of Article 177 references has not been immune to these trends. (Figure 3) Since the first preliminary reference was brought in 1961, national courts have increased their use of this procedure. [25] In 1970, 32 references were made, and by 1980 that number had tripled, to 99 references. In 1990, Member State courts made 141 references, and in
1993, 203 references, reaching double the 1980 level. Both the expansion of the Community and an increasing familiarity with the Article 177 procedure have contributed to this trend, although to what extent is uncertain. [26] 2.

2. Cases decided

The ability of the Court to render decisions has expanded over time, although not at the same rate as the increase in the number of cases brought. (Figure 4) This trend may be seen in the steady growth of cases pending before the Court, which reached 432 in 1993. [27] In 1960, the Court ruled on 20 cases. In 1970, the Court gave 64 judgments, in a year when 80 cases were brought. By 1980, this rate had dropped considerably, the Court ruling on 132 cases, while 280 were brought that year. In 1990, 193 judgments were given, and 287 [28] were brought. And in 1993, 204 cases were ruled on, while 486 were brought.

The Court has lagged behind in delivering preliminary rulings as well. (Figure 5) As the years have gone by, the gap between the number of references made and the number of rulings given has increased. In 1970, a year in which 32 preliminary references were brought by Member State judicial bodies, the Court delivered 28 preliminary rulings. In 1980, 99 references were brought, and 75 rulings were given. In 1990, 141 references were made, and 113 rulings were given. And in 1993, 203 references were made, and 128 rulings were delivered.

Some explanation of the dips in the Court's productivity may be found in the effect of the continual reappointment of potentially half of the bench every three years. [29] While Justices and Advocates General may be re-appointed, there is no guarantee that they will. (Figure 4) shows fairly regular declines or levelings-off in the volume of judgements, corresponding to the years of judicial turnover: 1991, 1988, 1985, 1982, 1979, 1976, and so on. The rises and falls in preliminary rulings delivered does not mirror exactly the pattern of decisions made overall, but instead suggests spurts of activity around the years in which judicial changes are made. (Figure 5) The general frequency of increases and decreases in productivity occur roughly every three years, however. [30]

Even though rulings have lagged behind references, the Court's ability to process cases has expanded over time. This enhanced capability may be linked to a number of factors, such as the expansion of the bench, concurrent with the expansion of the Community. Any impact that an increase in the number of judges has had is offset by the absolute increase in the volume of references brought, however; the overall ratio of case volume to Member States has risen over the lifetime of the Community. [31] (Figure 2) An increase in the use of chambers of five or three judges to consider cases -- rather than a full Court -- may have contributed to the ability of the Court to render a greater number of decisions as well. And it may be that the Court was simply functioning below full capacity in the past. If this last proposition is true, then the Court responded to the expanded case load by processing a greater number of cases. In all likelihood, some combination of these factors is responsible. Whatever the explanation, the number of cases pending before the Court has increased in absolute terms, and the average duration of cases has continued to rise.

B. Duration of cases before the Court of Justice

1. Interlude: a day in the life of a preliminary reference

At the Court, the conventional wisdom seems to be that, in the unlikely situation that no backlogs existed in the preliminary reference process, a minimum of nine months would be necessary to produce a ruling. [32] This assumption does not specify the optimal volume of cases to be decided within this time frame, however, and makes using the nine-month period as a realistic floor problematic. Would the Court take only nine months to deliver a ruling if it had only one ruling to deliver? Or is the "capacity" of the Court to process references within nine months higher? Although the technicalities of the Article 177 procedure allow for the possibility of delivering a ruling in under six or eight months, the reality is otherwise, as the twenty-month average for rulings in 1993 demonstrates. A brief look at the journey which a preliminary reference makes -- from when it first arrives in Luxembourg, to when it returns, as a ruling to the national court -- provides a useful context for understanding this discrepancy. This consideration is also necessary for speculating on which stages could stand to be abridged or otherwise altered.

After an order for a reference is submitted, that order is entered in the Court's Registry, and assigned to a Judge Rapporteur and an Advocate General. [33] At this point, opportunity is made for all the Member States, the Commission, and the Council to submit written observations on the case. A two-month, non-extendable time period is allowed for the delivery of these comments to the Court. [34]

Before this period begins to run, however, the order for reference must be translated into the languages of Member States, as each must receive notification of the reference in its own official language. [35] While the Court normally allows eight to ten days for this process to be completed, particularly complicated or unclear references will take more time. [36] A significant amount of delay may occur at this stage, despite the fact that priority is given to the translation of documents relating to preliminary references. [37] After the two-month period is over, so is the written stage of the procedure. If Member States choose to submit observations, which may be in their official languages, these documents must be translated into the language of the case. [38] Again, these translations may take time.

Once the written stage is completed, the Judge Rapporteur and the Advocate General are given time to formulate their opinions of the case, taking comments from the Member States and the Commission into account. [39] This stage usually takes at least six months. This period is problematic, particularly if the Court ultimately rules that the reference is inadmissible. [40]

The Court then makes an administrative decision, and decides whether to hear the case in full court or in a chamber. [41] If the case is heard in chambers, the size of that chamber must also be determined. [42] After the Judge Rapporteur's Report for the Hearing is distributed, the oral hearing commences. [43] At the end of the hearing, the Advocate General presents his opinion in open court. The case then goes into deliberation, and ends once the Court delivers its judgment. Discussion among the members of the bench can take months, however, and delay the process considerably. The amount of time allowed for deliberation is not bound by any limitations, and has been known to take a year.

Even before the deliberations stage is reached, then, the Court may take nine months to process a reference for a preliminary ruling. Depending on the difficulty and controversality of the case, deliberations may be completed in a few hours or a few days, or over a number of months. This situation is complicated by the fact that the judges have a number of cases in deliberations and in other stages at once, which renders scheduling time for discussing any single case difficult. Given all these factors, although nine months might be too short a period of time to expect a ruling to be delivered in normal circumstances, a year may not be an unreasonable target.

2. Statistics on duration

Having set up the difficulties inherent in the Article 177 process, the patterns may be examined. The amount of time which the Court of Justice takes to give preliminary rulings has increased at a fairly steady rate. (Figure 6) In 1971, the average duration of preliminary references was between 5 and 6 months. In 1980, that figure had risen to 9, and in the following year to 12. By 1990, the Court took an average of 17.4 months to give preliminary reference rulings. Between 1992 and 1993, this figure jumped over a month and a half, from 18.8 to 20.4 months.

Direct action proceedings have had a slightly different history. The rate of increase in duration has been more erratic, and levels reached on average have been higher than in the case of preliminary references. (Figure 7) In 1971, the average duration was 8.5 months. By 1980, that figure had more than doubled, to an average of 18 months. In 1981, the average duration dropped noticeably, to 12, but by 1985 was above the 1980 level, at an average of 20 months. By 1990, a direct action took on average 25.5 months to be processed by the Court. And in 1993, the average duration dropped slightly, to 22.9 months. This contrast may be explained by the fact that direct actions require a full written procedure, which is limited to the two-month period for the submission of observations in the case of preliminary rulings. [44] While additional time for filing may be received for direct actions, the two-month period applicable to Article 177 proceedings may not be extended. Moreover, there is greater impetus in the Court to adjudicate preliminary references quickly, as reflected by the priority given in the translation services to documents relating to rulings. [45]

This relentless increase in the duration of cases brought before the Court occurred despite the transfer of staff cases in 1989 to the then newly-formed Court of First Instance. Of the cases pending that year, 153 were removed from the Court of Justice's docket, and referred to the Court of First Instance. The duration of preliminary reference cases dropped slightly in 1989. But by 1991, delays exceeded 1987 levels, and continued their upward climb. What effect the 1993 expansion of the Court of First Instance's jurisdiction to all actions brought by natural or legal persons has yet to be seen. [46]

It should be noted that these figures for average duration do not take into account the amount of time taken by national judiciaries to prepare a reference. The gap between the decision to refer and the actual making of a reference varies between Member States and between courts, just as the national procedures for formulating and preparing references vary. [47] Accordingly, reforms applicable at the national level will be discussed in Part III.

Part II
Policy considerations

To a large extent, any decision to actively establish a docket control policy is dependent upon satisfaction with the current state of affairs. Opinions as to the desired or acceptable levels of case load, delays, and uniformity of Community law are integral to evaluating this situation, and considering what the next step should be. Many questions are raised in this respect. What is an "acceptable" number of cases for the Court to process, yet maintain the quality of its rulings? What delays are "reasonable," and at what point do they become "unreasonable"? And just how much "uniformity" of Community law across the Member States is desirable, or indeed realistic? The answers to these questions are partly subjective, and may turn on political issues in the Court and the Community.

Nevertheless, some attempt should be made to consider these questions objectively. This task is plagued with difficulties, however, for while it is easy enough to set boundaries at the extremes of "acceptable" and "unacceptable" situations, a large grey area exists between these "clear" situations. Taking the question of delays, one might conclude that the shorter amount of time spent, the better, as long as the quality of judgments is preserved. But what exactly is "too long" a wait? And what are the standards for "quality"? As was suggested above, expecting the Court to deliver most rulings in under a year is not unreasonable. For delays in the Court of Justice have implications reaching beyond Luxembourg, for they mean delays in the national proceedings from which the reference originated. And undue delays may reduce the willingness of national courts to refer, if the waiting period is too long. For these reasons, preferential treatment is given to preliminary references.[48] The conclusion here might be that if delays are such to cause national courts to stop referring as often as they might, then those delays are "unacceptable." [49] So are delays which effectively render the national case moot, due to a domestic statute of limitations or other circumstances. Beyond these benchmarks, however, there lies an uncertain realm -- short of crisis, perhaps, but not fully in it. A comparison with delays normally experienced in Member State courts might prove useful, but not necessarily dispositive. The Court of Justice is a different kind of court from even national supreme courts, and experiences a different type of pressure.

Moreover, the impact of delays on Member States varies, depending on their relative use of the preliminary reference procedure. Certain Member States make greater use of Article 177 than others. [50] In absolute terms, the countries making the most references are Germany, France, the Netherlands, Belgium, and Italy. In relative terms, however, this ranking is misleading. To have any real significance, the differences in size amongst the Member States must be normalized, although against what set of data is not clear. [51]

As with the question of delay, the Court's case load may be seen as "acceptable" if it is "workable." As long as the Court is able to render accurate and useful decisions, it is reasonable to say that the number of cases before the Court is at an acceptable level. Again, however, the problem of determining what quality of decisions falls into this category is beset by the difficulty of analyzing what is a fundamentally grey area. What exactly is a "useful" decision? The desired level of uniformity of Community law in the national courts are also at issue. But to determine what level of uniformity is "acceptable," knowledge of what uniformity actually exists is essential. And receiving adequate and accurate information on just how Community law is being used by judges in the Member States may prove difficult.

Although there seems to be a general consensus that controlling the volume and duration of the Court's work is necessary, the form and extent of that reform is not so clear. The data presented in Part I strongly suggests that the situation is a problematic one, and must be contained. In addition, recent instances in which the Court declared references by national courts inadmissible -- Meilicke, Lourenco Dias, and Telemarsicabruzzo, among others -- might be interpreted as evidence of a need for a docket control policy, if not an outright creation of one. [52] If the latter is true, the Court may already, in the absence of any explicit, legislative policy shifts, be adapting itself to the situation through judicial fiat. How far this adaptation can go without public debate is another question.

In the following section, two basic types of reforms are considered: those aimed at reducing the volume of cases before the Court of Justice, and those aimed at streamlining the management of cases. While the aim of all these reforms is to minimize delay, some options may have undesirable secondary effects, or might be difficult or virtually impossible to put into action. Reducing the total volume of cases of references brought before the Court might indeed reduce the duration of those cases, but national courts might be alienated, and important questions might fail to be considered by the Court. A filtering procedure might also be established. Such a system would need to be designed so as to avoid discriminating against cases and against national judiciaries. These last two methods would necessitate Treaty amendments, and thus require a intergovernmental conference and Member State consensus for adoption. Enhancing the discretion of national courts might also be effective at minimizing the flow of cases from the national courts, but has the probable cost of endangering the uniform development of Community law.

Focusing on the process of bringing references and ruling upon them also has implications for the relationship between national courts and the Court of Justice. Raising the quality of references made would certainly simplify the job of the Court of Justice, but might make national judiciaries somewhat defensive. These measures do have the advantage of being possible through means less disruptive than a Treaty amendment, however. Adjusting procedural and structural elements of the Court itself requires sensitivity to other concerns, which touch on the interaction between the national courts and the Court of Justice.

Part III
Possibilities for docket control: description & evaluation

At the moment, the mood at the Court is not one of urgency or panic. True, a substantial backlog of cases exists, to which the expansion of the Tribunal's jurisdiction has only been a partial remedy. And recent trends suggest that the situation will just become worse, particularly with the accession of new Member States. [53] But the system works, and, given the pressures exerted upon it, arguably works very well. Faced with the task of considering cases on a wide-ranging set of issues, from any of fifteen Member States, from any of the other Institutions, and in any of eleven languages, the Community legal system withstands a high level of strain. The fact that the system has not simply fallen apart might be reason to leave well enough alone. [54]

Just because the case backlog is not at crisis proportions does not mean the current situation should be ignored, however. Delays pose a problem. And the development of the Union and its legal system suggests that this dilemma will become more pronounced before it becomes less so. Once the system actuallecialize, are all ways of speeding up the disposal of preliminary references. Again, the question of what cost enhanced efficiency will have on justice and the purpose of Article 177 arises. Setting a balance between these concerns is ideal, but where that equilibrium lies is not so easy to determine.

A choice need not be made between these alternatives. Indeed, a successful solution will likely incorporatea number of different reforms. The actual choice of a program will require considering the varying amounts of time and inherent difficulties of adopting these reforms, once the will to implement them is established. For while some may be taken within the context of the existing powers of the Court of Justice, others will require more fundamental, legislative changes. A combination of small, short-term changes, along with more fundamental, long-term reforms might be the most desirable, for example. Then again, political reality may require another approach. While some changes to the Court's workings may be made through jurisprudential developments or by altering internal Court procedures, other measures will require adjustments to the Treaty itself, and thus an intergovernmental conference and the approval of the Member States.

A. Volume reduction

One means of reducing delays is simply to limit the number of cases which reach the Court in the first place. The possible benefits of this approach are self-evident: judges will be allowed more time for deliberation and adjudication of complicated and important cases, and will thus be better able to render prompt and thoughtful decisions, yet maintain the quality of those rulings. The probable disadvantages are equally clear: cases central to the uniform interpretation and application of Community law may fail to reach the proper forum. Of major concern is how any method will determine which cases or categories of cases deserve Court deliberation, and which do not. Leaving more discretion to Member State judiciaries could also reduce the burden on the Court, but might create more difficulties in the long term, by allowing for the development of inconsistent interpretations and applications of Community law.

The following discussion considers actions the Court of Justice could take to reduce the number of preliminary references sent to Luxembourg: increasing the discretion allowed to reference-making national courts, limiting which courts have the power to refer, and establishing a filtering procedure to determine which references are worthy of the Court's attention. Many of these options focus on the interaction between the Member State judiciaries and the Court of Justice, and necessarily invoke consideration of that relationship.

1. Increased discretion for national courts?

Before determining whether national court discretion to refer should be increased, the related question of whether national courts need be encouraged to refer should be considered. If the confidence of national courts to interpret Community law increased, the chance of references reaching the Court would probably decrease, and its case load would contract. Although the burden on the Court would be lightened, the interpretative role of the preliminary reference procedure itself might be put at risk. As arguably "borderline" subjects for rulings were retained by the national courts, the uniformity of Community law would be put threatened. Enhancing this discretion would reduce the necessity to make references in certain cases, but there is a danger that judges lacking a sufficient knowledge of Community law would fail to use this power appropriately. [55]

The behavior of domestic court judges is central to the preliminary reference mechanism, being at the heart of the relationship between Member State and Community law. [56] Article 177 fosters a system of judicial co-operation between national courts and the European Court of Justice, in which the Court is more at the mercy of the national judiciaries than vice versa. [57] For the ability of the Court to make rulings on Community law is limited by the judges presiding in national courts, in whom the responsibility to refer lies. Accordingly, the Court of Justice depends on national courts for the opportunity to consider such questions. This reliance renders the preliminary reference procedure particularly vulnerable, even as it makes possible legal developments that might not have survived political and formal legislative processes. [58]

The obligation imposed on the higher Member State courts in the third paragraph of Article 177 is not accompanied by a corresponding power to force compliance; the underlying assumption seems to be the hope that references will be made. [59] The Community legal system, characterized in part by the doctrines of direct effect and of supremacy, possesses a "system of judicial remedies and enforcement," distinguishing it from the more purely horizontal enforcement of classical public international law systems. [60] The strength of this system is not at its greatest with respect to preliminary rulings, however. Indeed, the viability of the Article 177 is dependent upon the willingness of national courts to comply. [61] This compliance has two parts: making preliminary references when appropriate, and applying the Court's rulings when they are made. Yet Article 177 is framed in extremely voluntary terms -- perhaps inspired by the general finding that non-compliance and defiance often result when judiciaries impose values not generally held. [62] There is no practical way to force a national court to adhere to its Article 177 obligation, barring an Article 169 action brought by the Commission against the offending court's Member State. Nor is there any means of ensuring that a preliminary ruling, once given, will be applied by the national court. These deficits are significant, given the centrality of the preliminary reference procedure to legal uniformity in the Community. [63]

Thus, the issue of national court "discretion" is really whether national courts need encouragement to refer, and not whether national courts should be urged to abstain from referring in more instances. As the case law suggests the concern of the Court is with the former, a major problem faced by the Court is how to ensure national judiciaries voluntarily comply with Article 177.

This bias is illustrated in the conservatism of those decisions which propose increased control by national courts. In 1962, the Court held in Da Costa that national courts of last resort may choose to not refer a question, if the Court has already ruled on an issue that is "materially identical" in substance to the one at hand, thus establishing the doctrine of acte clair. [64] The subsequent case law of the Court demonstrates an approach which fluctuates between encouraging and tightening access to the referral procedure, however. Almost twenty years later, in CILFIT, the Court underlined the ability of highest national courts to determine when a reference is necessary. A reference is not necessary if the ruling would have no impact on the outcome of the case, Da Costa applies, or the interpretation of Community law is "so obvious as to leave no scope for any reasonable doubt," even in the absence of a previous ruling on the subject. [65]

While these holdings might suggest that the Court capitulated in response to demands for greater control over the power to refer, it is possible to interpret these cases as disguised moves towards actually increasing control by the Court over the national judiciaries. Hjalte Rasmussen has argued that CILFIT did not herald a loosening of the obligation of national courts of last resort to refer, but was in fact a "tactical move" designed to enhance Luxembourg's influence over Article 177(3) courts. [66] The real purpose behind CILFIT, he concludes, cannot be to grant national courts more discretion over Community law, for that aim would conflict with the Court's consistent emphasis on the importance of legal consistency. [67] Thus, while the CILFIT doctrine does not guarantee compliance by national courts, it may have reduced the risk of non-compliance. [68] At the very least, through its decisions in Da Costa, CILFIT, and, later, Foto-Frost, [69] the Court consolidated its ability to control the development of Community law, particularly within the Member States. [70]

In more recent years, the Court has both tightened access to the reference process, and has encouraged courts to make references. This policy reflects the tension between two goals of the Court: to fulfill the aims of Article 177, and to do so efficiently. In Meilicke, the Court asserted its power to determine whether a preliminary ruling was necessary for the adjudication of a domestic dispute. [71] Two years later, in 1994, the Court suggested in Eurico Italia that, even if the Da Costa exception applies, it does not want to block national courts which wish to refer. [72] Certainly, national courts, particularly those falling under Article 177(2), already use discretion over whether to refer, aside from what the Da Costa and CILFIT line of cases permits. Recently, in Portman Agrochemicals, the English High Court decided not to refer a question, although the dispute was wholly dependent upon the interpretation of a Community directive. Mr. Justice Brooke noted that both parties had pointed out that the delays plaguing the preliminary reference procedure were such that the case would be moot by the time a ruling was given. Although lacking confidence in his ability to adequately resolve the issue, the High Court judge decided, in the interests of the parties and in consideration of the time and cost involved in obtaining a preliminary ruling, that he would interpret the directive himself. [73]

The frequency of requests for preliminary rulings strongly suggests that national courts are not hesitant to make references. This broad generalization aside, patterns of referral vary between and within Member States. [74] In absolute terms, Germany is far ahead in making references, while Ireland and Portugal are far behind. (Table B) As was discussed in Part I above, these statistics cannot be properly understood without taking into account other factors, namely the relative size of Member States, the length of their membership in

the Community, and the nature of their juridical culture. Only when appropriate adjustments are made can some sense be made of these trends. [75]

TABLE C

REFERENCES FOR A PRELIMINARY RULING MADE UP TO 31 DECEMBER 1990 [76]

Member State

References from "higher" national courts (%)

Total number of references made

Belgium

12.6

246

Denmark

27

36

Germany

30.7

684

Greece

13

23

Spain

0

11

France

12.0

399

Ireland

9.5

21

Italy

15.6

237

Luxembourg

32

28

Netherlands

16.8

328

Portugal

0

3

United Kingdom

6.7

118

Ultimately, whether to enhance national courts' discretion over the preliminary reference process should depend on the ability of national courts to decide questions of Community law accurately. National courts will be called on to make these decisions more often, particularly if Luxembourg loses the capacity to do so itself. Given that familiarity with Community law is still extremely uneven among Member State judiciaries and among their judges, and that there are national courts which will decide cases themselves if circumstances require, making an explicit move to extend this discretion seems unnecessary and even unadvisable. Although the number of references brought to the Court would likely decline, the damage to the uniform interpretation of Community law would be great. A worst-case scenario would be the atomization of Community law, in a way which would prevent the equal treatment of all Member States under its principles.

2. Limiting those courts which may refer

Overall, the majority of preliminary references are made by lower or appellate courts within the Member States. Looking at the origins of all references for a preliminary ruling made up to the end of 1990, the percentage of references have been made overwhelmingly by lower courts in each of the twelve Member States. (Table B) Relatively few of these cases were brought by national supreme courts. This phenomenon is interesting, considering that Article 177 only encourages lower courts to refer questions of Community law, while it positively requires the highest courts to do so. In no case have more than 32% of preliminary references been brought by the highest court in a Member State (as in Luxembourg), and in some cases this percentage has been zero (as in Spain and Portugal). There does not seem to be any predictable relationship between the percentage of references originating from higher courts and length of membership in the Community; Spain and Portugal, among the newer Member States, have not sent any references from their highest courts, but the United Kingdom, which joined in 1973, has only sent 6.7% of its references from the House of Lords.

Accordingly, the number of preliminary references brought to the European Court of Justice would surely be reduced if the group of national courts with the ability to use Article 177 was restricted or prohibited. [77] This end could be achieved in two ways: by deleting the second paragraph of Article 177, and thus restricting the use of the preliminary reference procedure to courts "against whose decisions there is no remedy in national law," or by limiting its use to "higher" courts, as identified in Table B. This approach would lighten the case load of the Court, but would have the danger of discouraging or preventing potentially important questions from ever being considered, simply because lower courts make extensive use of their power under Article 177(2). [78] (Table B) For a reference to reach the Court, the case must progress up through the hierarchy of national courts, until it reaches one from whose decision there is no appeal. Any method of docket control which functions by reducing the number of references presented to the Court must ensure that appropriate subjects of rulings are not denied access.

Deleting the second paragraph of Article 177 would restrict the power of referral to national courts and tribunals from which there is no appeal in domestic law. [79] This reform would not necessarily mean the limitation of referral powers to national "supreme" courts, however, as some specialized lower courts are arguably also courts "against whose decisions there is no remedy in national law." Labor and tax courts are two examples of this categorization. This restriction is already applied to preliminary references under the Brussels Convention, and was proposed in 1992 at the Edinburgh Summit by Germany. [80] In the context of the Brussels Convention, among the official reasons for excluding courts of first instance and other "lower" courts from the provenance of Article 177 were cited to include the following: preventing too many references being made, "particularly in trivial matters," and ensuring that the Court of Justice need only grant rulings where it is "fully informed," thus deferring the preliminary task of interpretation to the national courts. [81] If Article 177(2) were to be removed, the normal process for a Treaty amendment, including the approval of the Member States, would be necessary.

Limiting the use of Article 177 to paragraph three courts or to national supreme courts has some weaknesses. Simply put, the success of this reform requires that cases necessitating a preliminary ruling will actually reach the highest court of the land, so that a reference may be made. Sending a reference to Luxembourg is uncertain enough in the present system, without having this additional hurdle to overcome. Without Article 177(2), a case must have sufficient momentum to reach a court of last instance -- as determined by the time and money available to the parties to any given case, and the inclination to continue, particularly if a settlement can be made. Moreover, a lower court judge might choose to disregard issues of Community law, particularly if their relevance is ambiguous, rather than continue to drain the resources of the parties and the judiciary.

There is no guarantee that this restriction of access would ensure that the most worthy or important references reach the Court, for limiting referring power to highest courts is by no means a merit-based restriction. On the contrary, such a system would only guarantee that cases with substantial resources on hand could even reach the highest national courts. In addition, judges in the lower courts of the Member States might be encouraged to interpret provisions of Community law themselves, out of a concern for these elements of time and money. These considerations are made even now, without such limited access. [82]

Perhaps the Edinburgh proposal should be considered more seriously, however. Objections to deleting Article 177(2) as a means of docket control have been formed on both the practical and political level. As has been discussed, reducing Luxembourg's docket in this way would increase pressure on national courts, by keeping potential references within domestic judiciaries until they reach an Article 177(3) Court. Unless Member States streamlined their own procedures, the burden on these courts would increase. The political impact is framed in terms of a threat to the rule of law, if the Edinburgh proposal is believed to urge a reduction in Court of Justice jurisdiction. by preventing any court or tribunal from sending a reference to Luxembourg, the power of higher courts within the national systems is preserved. Under current Community law, the preliminary reference procedure has been a way for less powerful courts in Member States to circumvent their own supreme or other higher national courts; if a reference is considered by the Court of Justice, the referring court and all courts within its national system are obliged to honor the ruling.

Article 177 thus affords national courts against whose decisions an appeal may be some influence over their superiors. This situation creates an incentive for lower national courts to refer questions, and thus circumvent higher courts. On the other hand, higher courts dislike being on the same footing as the courts below them, even as they use Article 177 at much lower rates than Article 177(2) courts. If lower courts are deprived of the power to refer, then perhaps the incentives for higher courts to participate more in the reference process will increase. Indeed, the pressure on them to do so will rise, if the second paragraph is deleted.

While the majority of references are brought by Article 177(2) courts, that does not mean that these questions are the most important of those sent to Luxembourg. Indeed, references which have been filtered through a national court system tend to be more important, just because they have survived this extra consideration. Thus the deletion of the second paragraph of Article might increase the quality of references brought, as well as reducing their number. Ultimately, these concerns, and that of access to Luxembourg, must be balanced against one another.

Although preserving the right of lower courts to make references may be important to the integrity of the Court and the Community legal system, it may also be that the practical gains outweigh the political drawbacks. Preventing references from lower courts would restrict recourse to the Community legal system, and might endanger the uniformity of Community law. But if the higher courts of national judiciaries are pushed to make more use of Article 177, and receive more appeals from lower courts to do so, streamlining Member State methods for handling requests for references might result. Worries as to whether national courts will comply with their obligations under Article 177 if access to the Court of Justice is limited may be overstated. Ultimately, the loss of some uniformity may be worth the practical effect of lightening Luxembourg's load, and perhaps encouraging national judiciaries to adapt their procedures to the demands of Community law.

3. Selection of cases by the Court of Justice: the end of mandatory jurisdiction for preliminary references?

Limiting access to Article 177, based solely on the nature of the referring court, does not directly consider the relative merits of any given reference. A selective approach towards preliminary reference jurisdiction need not possess this shortcoming, however, by adopting a formal filtering procedure which evaluates the substance of requests. Eliminating requests based on their content, or redirecting certain references to another forum in the first instance, would reduce the burden on the Court of Justice. Alternatively, a less formalistic filter could be adopted through jurisprudential developments that indicate the appropriate form and content of a reference by example. A body of case law has already begun to develop along these lines, through the Court's inadmissibility rulings.

a. Formal filtering methods

The formation of an explicit, structured system to filter requests for preliminary rulings is one way to allow the Court to select and thus reduce its docket. What form this system should take is open to some choice. The methods used by some countries' supreme courts provide one type of model, in which courts directly exercise discretion over which cases to hear. These examples may not be appropriate to the European Community legal system, however. Alternatively, a supplementary level of Community courts could be created to handle this load, or the Tribunal could rule on references received from lower courts. [83]

i. Selection by the Court of Justice

The volume of preliminary references might be reduced by adopting a formalized filtering system, such as that which exists in the United States Supreme Court, the English House of Lords, the Australian High Court, and the German Bundesgerichtshof. [84] The Court of Justice could ensure that it would only have to rule on those cases having a significant impact on Community law, and not entertain trivial, unimportant, or inadmissible claims that are better heard by another forum or by national judiciaries.

One of the most examined filtering models is the United States Supreme Court's certiorari procedure, whereby the Court chooses which cases to consider on the basis of their importance or timeliness. From the late nineteenth century, when the case load on the Supreme Court was unmanageable, Congress litigation, culminating in the Judiciary Act of 1925, established almost purely discretionary jurisdiction for the Supreme Court. Now the Supreme Court hears only applications for review, and the cases it selected from that process. In 1976, petitions for certiorari made up over ninety percent of all cases heard by the Supreme Court. [85]

The United States Supreme Court's filtering techniques are potentially applicable to the situation in the European Court of Justice. [86] The "constitutional" questions presented to the Supreme Court play a similar role to preliminary references in the Court of Justice, particularly in their potential impact on the efficacy and development of the law. The analogy between these systems has its limits, however. First of all, while the formative period of American law is well past, the same is not strictly true of the Community. Indeed, it might be argued that a new, "formative" period of Community law occurs with the accession of each new Member State. Transitional time will be required for new Member States, even though they must accept the rule of law as established before their accession. As a result, it may not be for some years before the Community as a whole establishes a "normal" pattern of references. Secondly, as the Community legal system is now structured, the European Court of Justice occupies a different position than the Supreme Court: it is not a court of appeals, sitting at the top of a federal system of courts. This arrangement has implications for the relationship between national courts (particularly Article 177(3) courts) and the Court of Justice, which is different from the Circuit-Supreme Court dialogue.

Moreover, the volume of preliminary references is not so great in absolute terms that American-style certiorari would be particularly useful. The time spent considering petitions would not be considerably less than that spent in actual judgments, unless an abbreviated procedure was adopted. And such procedural changes might very well cause the reduction not just of time, but of sufficient judicial consideration.

ii. Shifting the burden: selection by the Court of First Instance

An alternative approach would be to allow the Tribunal jurisdiction to select cases for the Court's consideration. As the Treaty now stands, however, the Tribunal is not permitted to consider preliminary references. Nor is it allowed to hear all varieties of direct actions. In this context, two options exist for relieving the Court of Justice of some of its burden: to amend the Treaty to extend the Tribunal's jurisdiction to include Article 177 actions, or to permit the Tribunal to hear all direct actions, including those brought by Member States and Institutions.

Expanding the jurisdiction of the Court of First Instance is often suggested as a means of minimizing the Court of Justice's case load. [87] This direction has already been taken, particularly with the June 1993 Council decision to allow the Tribunal to hear "all actions brought by natural or legal persons," as provided by Article 168a TEU. [88] Changes to the Treaty mean that the Tribunal's jurisdiction may extend even further, however. Although the Council extended the jurisdiction of the Court of First Instance as far as the Treaty of Rome permitted, the Treaty on European Union has amended Article 168a, removing the limitation on the Court of First Instance's jurisdiction to actions brought by "natural or legal persons". Only the prohibition on hearing preliminary references remains the same. Potentially, then, the Court of First Instance might be able to hear cases brought by Member States or Institutions.

The main categories of cases now open to this reform are those brought under Article 169 and Article 177. With respect to Article 169 actions, some non-implementation cases brought by the Commission against Member States are purely formal, in that Member States do not contest them. Thus the effect of transferring these would be limited. Moreover, many of these cases should be heard by the Court itself, as they have constitutional significance. And even if Article 177 cases were diverted to the Tribunal, this transfer will not necessarily reduce the amount of time taken by references, but may merely redistribute the work. The Tribunal, which was envisaged as a fact-finding court, rather than a court of law, may not be the appropriate forum for preliminary references.

Since the extension of Tribunal jurisdiction to preliminary references seems unlikely, the effect of transferring all direct actions to the Tribunal, with appeal by leave to the Court of Justice, should be considered. [89] This move would leave the Court with two sets of cases to consider: references from national courts, and appeals from the Tribunal. There are not that many more cases that can be transferred after the extension of the Tribunal's jurisdiction in 1993, without causing the overload of the Tribunal, as well as of the Court. [90] The Tribunal is probably now at capacity, or will be in the near future. Simply put, treating the Court of First Instance as a general first instance court, from which appeals may be brought to the Court of Justice, will not alone correct the caseload situation. Shifting docket control problems is not the same as controlling them. [91] Moreover, the overall duration of cases within the European judicial system might even rise, given the increased opportunity for appeal.

The effect of the Tribunal to date has been significant, but its impact is not without limitations. While the Tribunal has reduced the proportion of Community litigation that reaches the Court, it cannot, and has not had an effect on the total volume of that litigation. The volume of Community litigation has only continued to increase since the start of the Tribunal's activity in 1989. While the Tribunal has reduced the case load of the Court in the amount of cases brought before it, its effect has been in redirecting existing cases, rather than altering their amount. Only if the Tribunal performed a filtering function, could it reduce the number of cases in the system.

As to the 1993 expansion of the Court of First Instance's jurisdiction, only time will tell whether any real change in the docket of the Court of Justice is achieved. This change now allows the Tribunal to hear all actions brought by natural and legal persons, excluding only actions brought by Community Institutions and Article 177 actions. [92] The impact of this transfer of jurisdiction depends on the number of cases that this shift will add to the Tribunal's docket, which will only be known through experience. [93] If the impact of the 1989 shift of staff cases to the Court of First Instance is any indication, the impact on cases remaining in the Court of Justice's provenance may be slight. For the beginning of the Court of First Instance's activities in November 1989 had little, if any impact on the average duration of preliminary rulings. The increase in the average duration of Article 177 cases may have slowed (rising to only 17.4 months in 1990, below 1988 levels), but there does not appear to have been any effect at all on the duration of direct actions (the average duration continued to rise after 1989, from 23 to over 25 months, and dropped perceptibly only during 1993, to an average duration of just under 23 months). (Table A)

iii. Selection by Regional Courts

The creation of another set of courts to make an initial judgment of a reference's admissibility, rather than merely shifting this burden to the Tribunal, may be more efficient in the long run. The addition of a set of Regional Courts, as outlined by Jacqué and Weiler, could perform this function. Preliminary references from the highest national courts would have automatic access to the Court of Justice, and references brought by lower courts are would be filtered through a set of four Regional Courts, each of which would have jurisdiction over the courts of a few Member States. [94] The Regional Courts rule on preliminary references from lower national courts within its region. Any Article 177 rulings made by the Regional Courts would be open to appeal by the parties, the Commission, the Council, the European Parliament, or the Member States -- but the Court of Justice would still possess discretion whether to hear the appeal. An appeal would be granted if the case raised an important issue of Community law, the Regional Court decision created an inconsistency between the jurisprudence of Regional Courts, or if the Court believed that a clear error of law was made. [95]

Although it might be countered that, given the right of appeal from Regional Court decisions, this structure would only further burden the system, the history of appeals from the Tribunal to the Court to date suggests otherwise. Since 1991, when the Tribunal first delivered decisions, the Court has ruled on only 36 appeals, out of a total of 266 Tribunal rulings. [96] The rulings of the Tribunal command sufficient authority; there is no reason, given sufficient time, that the Regional Court system could not achieve a similar reputation. On the other hand, giving a body other than the Court itself the power to screen cases might deprive the Court of its integrity. [97]

A more serious shortcoming of this system is that it assumes that references originating from national courts of last instance are more significant than those made by lower courts. This premise is often, but not necessarily true. The quality of references from higher and lower courts varies considerably, with the result that there are Members of the Court itself who find it impossible to say which level of courts consistently deliver the higher quality references. [98] The errors resulting from this valuation may not be substantial, however. A more serious shortcoming of this plan the hurdles of necessary Treaty amendments.

b. Informal filtering & inadmissibility rulings: a European certiorari? [99]

Should it wish to, the European Court of Justice might adopt a filtering process by applying the powers it already possesses. Through its case law, the Court may exercise some control over what cases are "justiciable" or "admissible" for consideration. This policy would not involve the legislative process and approval from the Member States that a structural change would necessitate. And, to some extent, this approach has already been set in motion. [100] Inadmissibility rulings provide a way for the Court to say what types of references are acceptable subjects of deliberation. [101]

Over the years, the Court has exercised preliminary reference filtering, through its jurisprudence, although perhaps without the same sophistication as a formal selection process would achieve. Because it lacks formality, a docket control method enacted through Court decisions will not be as efficient as the preliminary reference procedure demands. [102] Nevertheless, the Court has made use of its ability to refuse references. This approach has created more rigorous rules for the submission of questions, by demonstrating what requests are not successful. In Schwarze, while underlining the principles of judicial co-operation and uniform application of Community law, the Court established that national courts must adequately explain domestic law and facts, in order that the Court of Justice may make a real contribution to resolving the main proceedings. [103] While the Court did not refrain from ruling in this case, it set out the basis on which a refusal could be made.

While the Court has generally been extremely receptive to requests for preliminary references from national courts, it has not been so universally. In 1978, in the case of Mattheus v. Doego [104], the Court first declared a reference inadmissible, and refused to give a ruling. The line of cases that began with Mattheus may be evidence of a move towards more active management of the Court's 177 docket. [105] In Benedetti, the Court held that the uncertainty of facts and national law made it impossible for the Court to rule on the application of Community law. [106] More recently, in Telemarsicabruzzo, the Court used the same reasoning as in Benedetti, observing that the national court's imprecision made the Court's job of rendering a helpful ruling impossible. [107] The inadmissibility orders given in Banchero and Monin followed closely upon Telemarsicabruzzo, declaring requests for preliminary rulings "manifestly inadmissible" within Article 92(1) of the Court's Rules of Procedure. [108]

In some instances, inadmissibility rulings have set some of the boundaries for acceptable references. In 1992, the Court decided in both Meilicke and Dias to refuse either all or some of the questions referred, using exactly the same language in both cases to lay down certain principles of jurisdiction and review. [109] Drawing upon the reasoning used twelve years earlier in Foglia v Novello (No. 2), the Court maintained that it could review the appropriateness of a reference, thus reaffirming its own jurisdiction. [110] If, on the basis of the facts provided by the national court, it seemed that the reference was irrelevant to settling the domestic case, the argument ran, the Court should decide that a preliminary ruling was unnecessary. [111] In Dias, the Court said that if a question "concerns the interpretation of a provision of Community law, the Court is, in principle, bound to give a ruling." [112] The Court was convinced of the relevance of only six of the eight questions referred in this case. Although the facts and national law were sufficient for the Court to render a ruling, in two of the questions the relationship of the ruling to the main proceedings was not clear. By contrast, the Court's dissatisfaction with the reference in Meilicke was that the referring court failed to establish the national law that applied, prior to making the ruling. [113]

Inadmissibility rulings cannot be used as a substitute for a formal docket control policy, however. As will be discussed below, refusals to give rulings have done more to maintain the quality of references, than to provide an efficient means of docket control. And given the length of time it often takes to deliver even rulings of inadmissibility, their strategy may not gain much time. [114] In either case, however, the question of by what point in the procedure inadmissibility should be declared remains. If the decision not to admit the reference is made too early, some worthy references may be excluded. On the other hand, hand, waiting too long will delay the domestic procedure unnecessarily. It may be difficult to understand the significance of a particular ruling until the Advocate General's opinion is delivered; even then, the real issues may not be clear. The Court is as involved in defining the issues as in resolving them. [115] It may be six months before the Rapporteur and the Advocate General can "get to grips with a case," much less the Court reach the deliberations stage. [116]

Moreover, should the Court of Justice refuse to answer a question which is genuinely needed by a national court to settle a case, the national court will be pushed to make interpretations of European law itself, without any opportunity for review by Luxembourg. This result would in turn lead to the danger of divergent interpretations of Community law. The Court of Justice has made specific efforts to avoid just this problem; in Dzodzi and Gmurzynska-Bscher, the Court ruled on questions which were "European" in scope only so much as provisions of Community law had been extended by domestic law. [117] It would seem odd if the Court explicitly advocated the national courts' extensive application of their own interpretative powers.

B. Case Management

Another way of approaching the problem of the duration of cases in Luxembourg is to consider how preliminary references should be treated once they are added to the docket. This question looks to the efficiency of the Community legal system generally, and of the preliminary ruling process in particular. These reforms are aimed directly at reducing the duration, rather than the actual number of cases.

Should the Court wish to streamline the efficiency of the reference process, there are a number of ways in which it might do so. First, it could encourage the referring courts to suggest answers to the questions posed, and otherwise achieve a uniformly higher level of precision in the references submitted. This development would demand an effort by national judges to reach a certain level of familiarity with Community legal concepts. Second, reforms could be made within the Court of Justice and at the level of the Community legal system itself. These changes would involve tightening of Court procedure, possibly expanding the role of the Tribunal, altering the appointment and organization of the bench, or enhancing the relationship between the European Court of Justice and the referring courts.

1. Reforms in the national courts

To what extent do the practices of national courts contribute to backlogs and delays in the preliminary reference procedure? [118] The answer to this question will indicate the changes which must take place within the Member States. Certainly, the extent to which Member State courts are familiar with Community law cannot but directly contribute to the time taken by the Court to dispense with Article 177 cases. Moreover, knowledge of Community law and its procedural rules influences when and on what basis references are made. The form in which national courts present their references directly affects the ability of the Court of Justice to render accurate and prompt rulings. Accordingly, one response to the rising duration of preliminary rulings is to go to the inefficiencies that originate in the national courts. Raising awareness of Community law among judges in the national courts, and encouraging a standard for the form and content of references made would alleviate some of the burden on the Court. Referring courts might also be encouraged to draft answers to the questions presented, so to give the Court a better indication of the issue involved and of the national courts' concerns.

a. Raising familiarity with Community law in the Member States

The preliminary reference procedure is complicated by the facts that not all Member State judges possess the same knowledge of Community law, and that the form of references generally remains unstandardized. Just as Member State courts need to consider carefully whether a case should be referred to the Court in the first place, they must ensure that an accurate and complete statement of the facts and relevant domestic law is part of the request. Otherwise the Court cannot be expected to do its own job as promptly and as accurately as it might otherwise.

Improving judicial awareness of Community law in the Member States is thus fundamental to streamlining the Article 177 process. At this time, the level of knowledge of Community law required to qualify as a practicing lawyer or a judge varies significantly between Member States, and is minimal or nonexistent in some. [119] Efforts have already been made to train national judges and magistrates in Community law, so as to ensure its uniform application. In 1993, the Seminar on Training of Judges in Community Law, sponsored by the European Centre for Judges and Lawyers and the European Institute of Public Administration of Maastricht (EIPA), made some preliminary observations on the use of Community law in the national courts. Community law, it was observed, is becoming increasingly integrated into the everyday work of national judges. The lack of systematic training in or knowledge of Community law required for national judicial posts means that not only do most national judges and magistrates have an insufficient knowledge of Community law, but there seems to be a "widespread prejudice in judicial quarters in the Member States...to consider Community law an extraordinary law," rather than law relevant on a regular basis. In order to change this attitude, national judges should be made aware of the fact "that they belong to a Community of judges and a profession which is increasingly called on to apply Community law and that, consequently, they have a vocation to act as Community judges." [120]

The response to this shortcoming has been to advocate a system of "basic training" for judges and magistrates in the Member States. The general scheme envisioned at the EIPA conference seems to be based on voluntary compliance, whereby regional, national, and European bodies would be responsible for the training. The general consensus was that the majority of training should occur at the regional and national levels, perhaps in cooperation with European institutes. [121] If the Court of Justice and the Community became involved in this educational effort, its impact would be much greater, however.

It is also important that lawyers qualified to practice in the Member States possess some knowledge of Community law. [122] This condition is not as easily controlled or regulated by the Court and the Community, given their location in the private law realm. For the advice of lawyers is often central to the judge's consideration of whether to make a reference, particularly if the judge is unfamiliar with the Community law and procedure.

b. A standard form for references?

Another cause of delay is the vagueness which often characterizes preliminary references. Failures to articulate the factual and domestic legal basis behind a request create problems both for the Court, which must determine the answer to the question posed, and for concerned Member States and the Commission, which may wish to submit opinions during the written stage. [123] During the formative period of its jurisprudence, this sort of imprecision was not as problematic, for the volume of cases was smaller, and the Court used the opportunity to make wide-ranging pronouncements of law. Since the 1960s, however, the approach of the Court has been to limit the scope of rulings to those answers that the national courts truly need. [124] As a result of this shift, which has resulted in a much more narrow construction of references, the actual composition of requests has assumed greater importance. A full explanation of the facts of the case and relevant national law is essential to the Court's ability to render a ruling. Although its focus is on questions of law, the Court needs to understand the factual background to give a useful answer. Moreover, the Court is not intimately familiar with the legal systems of all Member States, and thus depends on an adequate explanation of the national law on which the case is based. [125]

One means of remedying this lack of clarity is, as was discussed above, to enhance knowledge of Community law in Member State courts. A complementary reform, also effective at the national level, would be the adoption of a standard form for submitted references. The quality and form of references is highly variable among the courts of the Member States. [126] When the elements of the reference are set out in a clear and standardized form, the Court is better able to render rulings promptly and accurately with respect to the issues invoked at the national level. For example, if this clarity had been achieved by the referring court in the Telemarsicabruzzo case, perhaps it would not have taken the Court twenty-eight months to declare the reference inadmissible: the Court might have been able to render a ruling or declared the case inadmissible at an earlier stage.

While standardization may be adopted at a Community-wide level, it need not be; standardization at the Member State level would be sufficient, as long as certain criteria were satisfied. A directive -- setting out guidelines for the presentation of the facts, relevant national law, and, as far as possible, the understood need for a reference -- is one way to achieve this reform. The practice of the German judiciary might be used as a model for such a set of guidelines. German references are presented in a fairly standard format, in which the facts of the case, and the relevant provisions of domestic law are clearly laid out. [127] Alternatively, a "procedural committee of judges," drawn from both the national and Community courts, could settle upon an appropriate set of guidelines, either formal or informal. [128]

Even if a Directive requiring of a standard form at the national level is not possible, the Court might be able to encourage this kind of approach through its jurisprudence. This development has arguably already begun, most notably in the Telemarsicabruzzo ruling, where the Court indicated that references must contain an explanation of background facts and law sufficient for the Court to render a decision. [129] There, the Court indicated what it submitted references should include. Not only does this approach avoid the difficulties of legislating, but it would have the added benefit of not over-defining the process. [130]

For at the level of principle, a real objection to standardization -- in a legislative or a jurisprudential format -- is that it might over-formalizing the reference procedure. If the voluntary compliance mechanisms associated with Article 177 outweigh the benefits of a more efficient process, this option may not be advisable. The demands for greater efficiency would seem to trump these concerns, however.

It may be that inadmissibility rulings themselves are pleas for some level of formalism. Some would argue not. In Vaneetveld, Advocate General Francis Jacobs argued that the Court's inadmissibility rulings in Meilicke, Telemarsicabruzzo, Banchero, and Monin were by no means departures from the Court's basic assumption that the preliminary reference procedure "is intended to provide for a form of judicial co-operation in which formalism should be avoided." Jacobs observed that while in Meilicke the Court refused to give a ruling because the questions were purely hypothetical, and thus beyond the scope of the Court's jurisdiction, the latter three cases were examples of where the national court's definition of the factual and legal context of the questions were crucial to the Court's consideration. In Vaneetveld, however, this requirement of clarity was not so crucial, and did not render the Court unable to deliver a ruling helpful to the national court. [131]

c. Encouraging the submission of draft answers: "green light"

If referring courts were encouraged to "draft" answers references, the Court of Justice could adopt a procedure to reduce the amount of time spent on preliminary rulings as a whole. This recommendation, described as the "green light" procedure, outlines a method of docket control that would avoid the negative side-effect of discouraging national courts from bringing references. [132] The referring court would simply submit a draft answer to the question it poses, along with the questions themselves, as German courts often do already.

If this scheme was put into action, the initial stages of the Article 177 procedure would be the same as now used, in that after preliminary references are made to the Court of Justice, the case would be assigned a Judge Rapporteur and an Advocate General. In cases where the national court judge states a preferred interpretation, however, the following analysis would take place: if both the Judge Rapporteur and the Advocate General agreed that the case did not raise a point of Community law requiring Court consideration, and that the suggested interpretation was correct in its outcome, that case could be returned to the national court for judgment, after a preliminary hearing at which the Court would state that it did not object to the submitted interpretation. [133] Functionally, this procedure could be adopted by revising the Court's Rules of Procedure, or, less formally, through case law. If the jurisprudential approach was adopted, the Advocate General's opinion or the Judge Rapporteur's report could advocate the answer proposed by the referring court, when appropriate. To adopt the "green light" approach, either a change to the Court's Statute, or to the Rules of Procedure would be necessary. The former, which is part of the Treaty, would require consideration at an intergovernmental conference; the latter would require Council unanimity.

Not only would the workload of the Court be reduced, but this approach might have other positive effects, namely encouraging national courts to be more careful in framing their references. For by submitting their own interpretations, the national courts will also gain experience in applying Community law -- in essence creating a forum for the review of their decisions by the Court of Justice.

The main problem with the "green light" procedure is that national judges must have a basic level of familiarity with Community law for the system to work. A criticism of the "green light" is thus the following: if national judiciaries knew enough to be able to draft a answer to the question posed to the Court, a reference would likely be deemed unnecessary in the first place. According to this position, requiring draft answers would have the effect of discouraging references from courts unsure of how to formulate their proposals. [134] But courts with even a small level of knowledge of Community law may have an idea as to how they would prefer to have the ruling made, even if that sense is intuitive or based on domestic legal principles. The danger of discouraging courts from making references could be avoided simply by strongly suggesting the submission of draft answers, rather than requiring them. This encouragement would be underlined by the possibility of shorter delays for national actions.

d. Other possibilities for co-operation with national courts

No provision exists to allow the Court of Justice to ask for clarification of references when the need arises; either the Court must interpret unclear references or declare the questions as inadmissible. The suggestion has been made that the Advocates General might be enabled to act as intermediaries between the Court of Justice and national courts in these situations. [135] This use of the Advocates General would avoid any awkwardness that

might result if Justices from the Court, ultimately responsible for giving a preliminary ruling, were to question the national judge themselves. The peculiarities of national law may not allow this possibility, however, as some Member States do not allow the referring court to take any action until the Court of Justice returns a ruling. [136]

Allowing a national court to "certify" the urgency of particular references might also increase the efficiency with which the Court answers particularly significant questions. The best way of operationalizing such a mechanism would be to build it into the Court's Rules of Procedure -- giving the device more bite than jurisprudence would create, yet avoiding the difficulty of that a change in the Court Statute would involve.

Focusing on the role of national courts in the preliminary reference procedure has produced several possibilities for docket control. These reforms concentrate on the "life" of a reference before it reaches the Court of Justice, and emphasize the co-operative nature of the process: increasing the familiarity of national judges and attorneys with Community law, establishing a standard form for the submission of references, and encouraging the inclusion of draft answers. An examination of the later life of a reference, and its transformation into a ruling, will reveal other opportunities for streamlining the process as it occurs within the Court itself.

2. In the Court of Justice: procedure

There are ways in which the Court might increase the efficiency of the preliminary reference process. Indeed, a tendency towards doing so has already been demonstrated. Time limits have been placed on oral arguments [137]; the same might be done for the deliberation period, the length of time perhaps depending on the number of judges involved in the case. An example might be taken from the practice of the Tribunal, in which all oral pleadings have been dispensed with. Since the Tribunal deals primarily with issues of fact, and not of law, this reform may not be appropriate for actions before the Court of Justice, however. As the procedure in the Court now stands, oral hearings are no longer automatic, but only happen a party requests one, and the Advocate General's opinion is no longer delivered in open court. [138] In certain cases, the Court can simply deliver its own opinion by reference to the Advocate General's, although it has done so in only a few instances. [139]

a. Oral hearings

The issue with respect to oral hearings is not so much whether they may be dispensed with, for that occurs fairly often. [140] Whether they may be dispensed of without the consent of the parties is the real question. This move would be a very difficult one to make, and probably not advisable. If national judiciaries felt that the Court of Justice did not guarantee due process, the legitimacy of the Court as an Institution, and Community law as a system, would be threatened. In any event, the Court tries to minimize the procedure, when it is clear that a ruling is not a viable option. [141]

A more practical means of streamlining is to adopt more widely what Tom Kennedy has termed the "fast track" method, by which the Court of Justice would dispense with oral proceedings, or even return the case to the national court, if existing Community jurisprudence adequately addresses the issue. [142] The use and advisability of this treatment depends on the value put on oral hearings, and the value the Court gives to its judgements as precedents. The effect of a ruling may be very different, depending on particular legal and other circumstances within a Member State. A ruling that is satisfactory for a situation in one Member State might not do so in another, even if the question is essentially the same.

There is no real consensus on the value of oral hearings in the preliminary reference procedure. While the oral part arguably provides an invaluable and otherwise unavailable opportunity to respond to an opponent's written pleadings, the oral hearing does not necessarily add much to the progress of the case, depending on the practice of attorneys. [143] Failures by advocates to contribute significantly to the outcome of a case may be explained by tendencies of different legal cultures, and the extent to which an advocate is accustomed to exploiting the opportunity. For while common law systems value the advocacy potential of oral arguments, the same is not as true in civil law systems. [144]

Determining the importance of oral hearings according to the definition within a particular legal system is not appropriate, as it fails to account for the variety of legal systems in the Community. European Community, and not national law should be the basis of any attitude, in the spirit of the Community legal system as a "new" and unique legal order, which is different than the sum of law of the Member States. [145] One approach is that the "non-contentious" nature of preliminary reference proceedings, in contrast to the other types of cases heard by the Court, allow a loose interpretation of the oral pleadings requirement, permitting their disposal in certain circumstances. [146] Since the Court differentiates between contentious and non-contentious proceedings on the basis of cost to parties, the argument runs, "it could on the very same grounds liberalize the requirement of written and oral proceedings and dispense with the latter" under certain conditions. [147]

b. Using orders to declare inadmissibility

Using orders to dispense with inadmissible references possesses similar drawbacks. In Banchero, the Court dismissed a reference as inadmissible, using an order rather than a full judgment to do so. Since the national court failed to precisely lay out the facts and relevant provisions of domestic law, and to state the exact reasons for the reference, the Court argued, rendered it impossible for the Court "to give a useful interpretation of Community law." [148] The time gained by using an order was considerable. Banchero was expedited in under eleven months, almost half the average duration of that year's rulings. The Court relied on both statute and case law in this decision. Article 92(1) of the Court's Rules of Procedure provided a basis for its conclusion in Banchero. According to this provision, "[w]here it is clear that the Court has no jurisdiction to take cognizance of an action, or where the action is manifestly inadmissible, the Court may, by reasoned order, after hearing the Advocate-General and without taking further steps in the proceedings, give a decision on the actabruzzo and Banchero emphasizes the temporal advantages of adopting the use of orders to declare inadmissibility. In Telemarsicabruzzo, the Court held that the reference was fragmentary and lacking in detail, and refused to answer the questions presented. The Court took almost 28 months to decide that it would not decide, however. This delay is undoubtedly exceptional, but the contrast with Banchero underlines the efficiency of using orders. Preliminary rulings are interlocutory to the main, domestic proceedings -- thus using orders to reject a reference will directly reduce delays in national courts waiting for rulings. Orders may be used inappropriately, however. If a valid reason for a reference exists, but a full procedure is necessary to make that reason clear, then using an order would be unwise. Given the Court's tendency in the past to interpret questions presented to them, a widespread use of orders may be unlikely. Banchero may just have been a response of the moment to the problems of docket control, without any intention to develop a general principle. [151]

Alternatively, the Court might use the "reference back" when national issues remain in dispute indicating in its ruling what the national court needs to determine so that the ruling may be applied correctly. In effect, the ruling would contain options, chosen according to the domestic situation. This approach does not ensure uniform application of Community law, and may also produce frustration in those national courts which receive a non-conclusive ruling. [152]

3. In the Court of Justice: structure

While altering the procedure applied to Article 177 actions provides one way of reducing delays in the preliminary ruling process, reconsidering the organization of the Court itself leads to another set of possibilities. These options focus on the internal structure of the Court, although they may be intertwined with procedural issues as well.

a. Full court or chambers?

How many judges are necessary to render a preliminary ruling? The Court may make a decision either as a full Court, a petit plenum of nine judges, or as a chamber of five or three judges. [153] The Treaty on European Union modified the second paragraph of Article 165, so that convening a plenary session is no longer required to consider a preliminary reference, unless a Member State, the Commission, or the Council requests one.

In order to make better use of the Court's resources, the defaultsize of the deciding body might be shifted down to the chamber of five or the petit plenum, so that more cases could be considered simultaneously. [154] Although no "parties" exist in the traditional sense to request a plenary session for Article 177 cases, the Member States and the other European Institutions serve this function. Moreover, the uniformity of Community law might be threatened. [155] Opinions range on this issue: while some believe that the common law practice of a single-judge bench would be appropriate, others do not agree. [156] In any event, the pressures on the Court to reduce the delays points to the increased use of chambers over the plenary or petit plenum. Nevertheless, it is unlikely that any Member State would prefer to have an important case decided in anything less than a full court, thus limiting the practical possibility of using chambers more intensively.

b. Number of judges

Increasing the number of justices sitting on the Court of Justice has occurred along with the accession of Member States. But this practice may not be feasible when the Union expands much beyond fifteen Member States, and alternatives should be considered. While enlarging the bench increases the resources adjudication, it also means more points of view, and thus more time spent in deliberations. [157]

c. Judicial specialization

On either a formal or an informal basis, the Court of Justice might push the Judges and Chambers to specialize, although to date the Court has resisted this trend. [158] No formal procedure exists by which the past experience of Judges and Advocates General is taken into account. Thus, although social security legislation is particularly complicated, Judges who have familiarized themselves with the issues and statutes will not necessarily be able to apply that knowledge to other decisions. While this practice might lead to excessive narrowing of expertise, and possibly rigidity of view, it would also make the process more efficient.

d. Judicial qualifications & length of judicial tenure

There are no standardized qualifications for Judges appointed to the Court of Justice or the Tribunal. Judges and Advocates General must be selected from among those "who possess the qualifications required for appointment to the highest judicial offices in their respective countries or who are juriconsults of recognized competence," but there is no requirement that they be well-versed in Community law, or able to function easily in French, the working language of the Court. [159] Neither is there any guarantee that judges assigned to a case be familiar with its working language -- which in Article 177 actions, is the language of the referring court.

This lack of qualifications specific to the appointment of judges to a Community court, coupled with a relatively brief tenure, presents problems for maintaining efficiency in the judicial process. The Treaty currently provides for a term of six years, with a partial replacement of seven and six judges to occur every three years. [160] Thus, every three years, half of the bench may be new. And the ability of new judges to function at highest efficiency is not guaranteed. Potentially, a judge might need two or three years to gain adequate expertise in substantive and procedural Community law, adjust to the system, and achieve fluency in French. During the last year of tenure, a judge will wind down his involvement. As a result, some judges may have only two or three years of truly active work. As was noted in Part I, this situation has caused a cycle of declines and rises in the Court's productivity, occurring roughly every three years. (Figures 4 & 5)

Establishing more specific requirements for eligibility as a Community judge has the main disadvantage of offending Member States with respect to their selection of Luxembourg judges, however. The possibility of making the necessary Treaty amendment is thus extremely unlikely. An alternative lies in the amount of time judges and Advocates General stay in Luxembourg, however. If the tenure of Judges and Advocates General was extended, perhaps to eight or nine years, considerable efficiency gains might be achieved, and the dips in Court productivity would be less frequent, and perhaps minimized. By making the term non-renewable, any dangers of political entrenchment could be avoided. This change would involve a Treaty amendment, however, and all of the political difficulties that such a process entails. Indeed, this proposal has been made already. In preparation for the Maastricht Summit, the European Parliament suggested amending Article 167 of the Treaty to provide for a maximum twelve-year tenure for Judges and Advocates General, without renewal. [161]

e. Limiting the number of working languages for references

All orders for references received from national courts must be translated into all Member State languages before any other steps can be taken. [162] Until recently, this meant a total of ten languages; with the recent accession of Austria, Finland, and Sweden, an additional two languages must now be accommodated. [163] The translated order is sent to the Member States, which then have two months to submit written observations, as do the Commission and the Council. These translations normally take eight to ten days, but unclear or lengthy orders for references potentially add weeks to the process. [164]

As has been mentioned, these required translations are a major source of delay in the preliminary reference procedure. For while the language of Court procedure is French, preliminary reference case language is that of the referring court's Member State. The speed at which a reference reaches the oral phase depends in great part upon language of case, length of documents, and translation backlog. [165] Given the recent enlargement of the Community to fifteen Member States and eleven languages, this issue cannot be ignored. [166]

Indeed, it may be time to put the "regime of linguistic equality...to sleep," especially as the Court's jurisdiction expands into the Eastern European countries. [167] The House of Lords EC Committee has had the idea of limiting submissions to the Court to two or three languages, and to charge the parties extra for those cases submitted in additional languages, so to force the creation of a group of legal experts with certain linguistic abilities. [168] The obvious problem with limiting the languages accepted for submissions is simple: how to choose those languages? But the solution is complicated. Each Member State has a natural prejudice for its own language. Working languages could be chosen by lot, or by extent of usage, both of which have their problems. Without a selection of languages by lot, there is a danger that the politically most powerful States will achieve primacy in the Court as well. And while not all countries would be happy with a French-English-German choice, but the difficulties of a Court which accepted only Finnish, Portuguese, and Greek documents are also evident. [169]

f. translation services

A less politically volatile alternative would be simply to enlarge the translation staff at the Court. [170] Given the number of pages currently required for translation, there is little doubt that additional translators would be gainfully employed. Furthermore, the Court might lighten the load of the translation service by selecting which judgments to publish, as occurs in the United Kingdom and the United States. The practice of dropping the publication of the Report of the hearing has had a good result on the translators' work load. [171]

Conclusion

There are advantages and disadvantages associated with choosing action -- or inaction -- as the engine in framing a docket control policy. It may be that the political difficulties of undertaking any fundamental, structural changes may outweigh the benefits that such reforms might have. On the other hand, the comparative importance of sustaining the uniform interpretation and application of European Community law is arguably too important to leave to chance developments.

What is becomingly increasingly clear is that the Court cannot wait much longer to take up a more active attitude. The ultimate victim will be the integrity of European law, which the continued use of the preliminary reference procedure is designed to promote. Proactive reforms should be undertaken, so as to prepare the Court of Justice and the Community legal system for the increased pressure that the expansion of the Union and the growth and deepening of its policies will bring. Ultimately, productive "evolution" will require active effort by the Court and the Community. The reality of political climates, however, may mean that any reforms to the European Court of Justice, particularly structural changes, will not neceay be simply those on which consensus is easy to reach.

The reforms, or combination of reforms, that are ultimately undertaken depend as much on political as on legal issues, if not more so. Appendix II, which summarizes the discussion in Part III, is meant to provide a guide to the options available, which can be applied as policy concerns dictate.

Appendix I
Blueprints for reform

There are two general approaches that docket control policy reforms might take: minimalist and maximalist. The characterization of these approaches is largely determined by the methods necessary for the adoption of particular alternatives. At one end, a combination of options in which the Court makes changes by its own decision might be adopted. These reforms are relatively uncontroversial, primarily involving issues internal to the Court. At the other extreme, a more radical, aggressive approach might be taken, in which amendments to the Treaty itself are central, and thus require an intergovernmental conference for their adoption. In between these poles lies those options which require the intervention of the Council, such as changes to the Rules of Procedure, or which may be put into place by Directive or Regulation.

The following are possible minimalist and maximalist views of Court reforms leading to docket control.

1. a minimalist approach:

--educate national judges in Community law, through administrative action;

--standardize form and content of references, through jurisprudence (or a directive);

--encourage submission of draft answers, through jurisprudence and altering the Court's Rules of Procedure;

--expand translation staff, or increase its productivity, by administrative act.

2. a maximalist approach:

--expand the capacity of the Community legal system, adopting the Regional

Courts system, and provide a judicial filte

European Community

ECJ

European Court of Justice

IGC

Intergovernmental Conference

juris

Court of Justice jurisprudence

MS

Member State(s)

nat'l cts

Member State judiciaries

Ref(s)

Preliminary reference(s)

Reg

Regulation

RP-ECJ

Rules of Procedure of the ECJ

TEU

Treaty on European Union

A. Volume

Method

Reduce volume?

Reduce duration?

Means of adoption

Advantages

Disadvantages

Increase discretion for national courts

possibly; depends on the juridical culture of MS

not directly

juris

involve national courts more closely; lighten case load

threat to uniform interpretation of EC law

Limit those courts which may refer

yes

possibly

IGC

fewer refs

important refs may never be made

Selection or filtering of cases

by ECJ

yes

maybe

IGC

most important refs heard

still has to review; procedural justice curtailed?

by CFI

yes

maybe

IGC

burden off of ECJ

CFI shifting to busy CFI will not solve problem

Regional Cts

yes

yes

IGC

lower cts can appeal 177 to ECJ

radical restructuring

Informal

yes

maybe

juris (inadmiss. rulings)

inadmissible selected out

impractical; important refs not heard

B. Case management

Method

Reduce duration?

Means of adoption

Advantages

Disadvantages

Nat'l cts

Increasing familiarity with EC law

Yes

educational programs, sponsored by ECJ (admin); Dir

further integration of EC law throughout MS

Some nat'l courts might not make refs where should

Standardization of form & content of references

Yes

Dir, juris, or Reg

standardization of reference format & content

reduce trust of nat'l courts in ECJ's respect for them

Encourage draft answers ("green light")

yes

juris

less work for ECJ

might discourage refs, if thought draft is obligatory

Enhance ECJ-nat'l ct co-op (AGs as intermediaries)

Perhaps

Reg or Dir

ECJ can go back to nat'l ct for explanation

nat'l cts might have to modify own procedure

Enhance ECJ-nat'l ct co-op ("certif-ication" of refs)

Yes, in selected cases

RP-ECJ

guide ECJ; foster ECJ-nat'l ct co-operation

nat'l ct might not identify correct cases

ECJ: procedure

Oral hearings

yes

juris; Council

speed process

some refs not clear without oral part

Use of orders

yes

juris

speed process

some refs not clear without oral part

ECJ: structure

Chambers (increase use)

possibly

admin

more decisions, same number judges

MS might demand plenary

Number of judges

uncertain; might actually increase deliberation time

IGC (165 TEU)

more to divide into chambers

inefficiency; difficulty of reaching decisions

Specialization

perhaps

admin

enhance expertise of judges in areas

limit expertise in broad range of areas

Qualifications

probably

IGC (167 TEU)

longer period of efficient tenure

offend MS; reduce their trust in ECJ

Judicial tenure

perhaps

IGC (167 TEU)

longer useful tenure

possible entrenchment

Languages (reduce those for Art. 177 use)

perhaps (if less translating to do)

Council unanimity (104(1) RP-ECJ)

less work for translators

discourage references; choosing working languages; possibly no effect on translator productivity

increase translation staff

probably

admin

reduce time lags

cost; offend translators

Appendix III

Judgments delivered by the Court of First Instance

direct actions

staff cases

total

1990

6

52

58

1991

26

15

41

1992

41

19

60

1993

43

4

47

1994

19

41

60

Note: of all the cases heard by the Court of First Instance through 1994, only 36 have been appealed to the Court of Justice; 16 of these appeals were dealt with by the Court of Justice during 1994.

Appendix IV

Ratio of Case volume to number of Member States

Member States (number)

Year

case volume/number of Member States

6

1960

4.6

1961

4.5

1962

6.0

1963

18.5

1964

9.3

1965

10.3

1966

5.2

1967

6.1

1968

5.5

1969

12.8

1970

13.3

1971

16.1

1972

13.6

9

1973

21.3

1974

11.3

1975

14.5

1976

14.1

1977

17.5

1978

30.0

1979

17.6

1980

31.1

10

1981

32.3

1982

34.8

1983

29.9

1984

31.2

1985

43.4

12

1986

27.4

1987

33.0

1988

31.2

1989

32.8

1990

31.8

1991

28.5

1992

36.6

1993

40.5

1994

29.5

Appendix V

Normalizing the volume of references from Member States [173]

TABLE V.A

PRELIMINARY REFERENCES COMPARED TO

POPULATION LEVELS IN MEMBER STATES [174]

Member State

References brought

Population (1000s)

References/1000 persons

Belgium

246

9987.0

24.63

Denmark

36

5146.5

6.99

Germany

684

79,753.2

8.58

Greece

23

10,120.0

2.27

Spain

11

38,993.8

0.28

France

399

56,893.2

7.01

Ireland

21

3,518.7

5.97

Italy

237

57,746.2

4.33

Luxembourg

28

384.4

72.84

Netherlands

328

15,010.4

21.85

Portugal

3

9858.5

0.30

United Kingdom

118

57,530.0

2.05

When population levels are taken into account, the concentration of Article 177 use in each Member State is weighted differently than if absolute numbers only are taken into account. When set against population figures, the frequency of referral among Member States is ordered as follows: Luxembourg, Belgium, the Netherlands, Germany, France, Denmark, Ireland, Italy, Greece, the United Kingdom, Portugal, and Spain. Luxembourg, a country populated by some 384,000 persons, sends preliminary references to the European Court of Justice at a rate of over 72 for every thousand persons. This figure may be somewhat distorted by the disproportionate number of European enterprises registered in that country. In any event, the concentration of references from Luxembourg well exceeds Italy's, which lies at 4.33 per thousand persons, although Italy has actually made 237 references to date, compared to Luxembourg's 28. While the most dramatic effects are observed with respect to Luxembourg, but also reveals that the concentration of references made by the United Kingdom is much lower than first appeared (ranking tenth out of the twelve Member States, and not fifth), and Italy somewhat less so (falling from fifth to eight place). Ireland is shown to have a higher concentration of references than the absolute numbers suggested; given the low population, the total of 21 preliminary references made by the Irish courts results in the figure of almost 6 references per thousand persons (putting that Member State in seventh, rather than tenth place). All of this is just to say that the concerns of Member States with respect to docket control may not be identical, and that any decisions made may need to examine certain particularities. [175]

TABLE V.B

PRELIMINARY REFERENCES COMPARED TO

GROSS NATIONAL PRODUCT (GNP) PER CAPITA IN MEMBER STATES [176]

Member State

GNP

References/GNPx1000

Belgium

17,580

13.99

Denmark

22,440

1.60

Germany

22,360

30.59

Greece

5,980

3.84

Spain

11,010

0.99

France

19,420

20.54

Ireland

10,390

2.02

Italy

16,940

13.99

Luxembourg

29,460

0.95

Netherlands

17,850

18.37

Portugal

5,190

0.57

United Kingdom

16,020

7.36

The normalization of reference figures is somewhat readjusted when GNP figures are used. The ratio between references brought and GNP puts the Member States in the following order of intensity of use of Article 177: Germany, France, the Netherlands, Belgium, Italy, the United Kingdom, Greece, Ireland, Denmark, Spain, Luxembourg, Portugal.

TABLE V.C

PROPORTION OF COMMUNITY LAW QUESTIONS REFERRED TO THE COURT OF JUSTICE,

1972-1983 AND 1984-1993

1972-1983

1984-1993

31% (163/525)

28% (286/1496)

Germany

26% (371/1429)

29% (416/2205)

Ireland

50% (11/22)

38% (19/88)

Italy

29% (132/452)

16% (176/1074)

Luxembourg

20% (7/35)

30% (21/69)

Netherlands

31% (173/563)

14% (200/1471)

United Kingdom

20% (42/215)

21% (120/569)

total

27% (1029/3830)

19% (1445/7657)

Comparing the number of references made to the total number of domestic cases involving Community law questions provides another basis for normalization. Preliminary references represent only a fraction of all cases concerning Community law in any given Member State. Jonathan Golub has provided this information, in the table which is reproduced above. [178] This proportion varies between Member States, and sometimes between the two periods considered within a single Member State. These percentages may be used to put reference rates in some context. But this information should be used with some caution, for while this information may illuminate referral trends among Member States, the explanation for these proportions will differ between States. These numbers cannot explain for the influence that differences in juridical culture may have. Moreover, the percentage of cases referred should be considered in light of the absolute figures involved. For example, while the proportion of references to Community law cases only rose one percentage point in the United Kingdom over the two periods considered, the total number of references and EC law question cases increased considerably. Moreover, in the Netherlands, the percentage dropped by over half, but the total volume of cases implicated came closer to tripling.

Sources

Books & articles

Andenas, ME COURT. Chicago: American Bar Foundation, 1976.

Common Market Law Review. Quis Custodiet the European Court of Justice? 30 COMMON MKT. L. REV. 899 (1993).

Congressional Quarterly, Inc. THE SUPREME COURT, JUSTICE, AND THE LAW. 2nd edition. Washington, D.C.: Congressional Quarterly, 1977.

Court of First Instance of the European Communities. Reflections on the Future Development of the Community Legal System. 16 EUR. L. REV. 175 (1991).

Cruz Vilaça, José Luis da and Luis Miguel Pais Antunes. The Court of First Instance of the European Communities: A Significant Step towards the Consolidation of the European Community as a Community Governed by the Rule of Law. 10 Y.B. OF EUR. L. 1 (1990).

Dine, Janet and Sionaidh Douglas-Scott, Ingrid Persaud. PROCEDURE AND THE EUROPEAN COURT. London: Centre of European Law/Chancery Law Publishing, 1991.

Dué, Olé. The Court of First Instance. 8 Y.B. EUR. L. 1 (1988).

Dungworth, Terence and Nicholas M. Pace. STATISTICAL OVERVIEW OF CIVIL LITIGATION IN THE FEDERAL COURTS. Santa Monica, Calif.: RAND, 1990.

Edward, David. The Nature of the Community Judicial Process: How the Court of Justice Works as a Judicial Body. In Kirsten Thorup and Jens Rosenlov, eds., FESTSKRIFT TIL OLE DUE. Copenhagen: GEC Gads forlag, 1994.

European Law Review. Statistical information from the Court of Justice and the Court of First Instance. 18 EUR. L. REV. 177 (1993).

Frankfurter, Felix, and James M. Landis. THE BUSINESS OF THE SUPREME COURT: A STUDY IN THE FEDERAL JUDICIAL SYSTEM. 1927; New York: The Macmillan Company, 1993.

Golub, Jonathan. Rethinking the Role of National Courts in European Integration: A Political Study of British Judicial Discretion. EUI Working Paper LAW No. 94/12. Badia Fiesolana, San Domenico: European University Institute, 1994.

Jacobs, Francis G. Amendments to the Rules of Procedure. 5 EUR. L. REV. 52 (1980).

----------. Procedural Reform in the European Court. In Jürgen Schwarze, ed., PERSPECTIVES FOR THE DEVELOPMENT OF JUDICIAL PROTECTION IN THE EUROPEAN COMMUNITY, 191-197. Baden-Baden: Nomos Verlagsgesellschaft, 1987.

----------. Proposals for Reform in the Organisation and Procedure of the Court of Justice of the European Communitites: with Special Reference to the Proposed Court of First Instance. In F. Capotori et. al., DU DROIT INTERNATIONAL AU DROIT DE L'INTÉGRATION -- LIBER AMICORUM PIERRE PESCATORE, 287-298. Baden-Baden: Nomos Verlagsgesellschaft, 1987.

Jacqué, Jean Paul and J.H.H. Weiler. On the Road to European Union--A New Judicial Architecture: An Agenda for the Intergovernmental Conference. 27 COMMON MKT. L. REV. 185 (1990).

Kapteyn, P.J.G. and P. Verloren van Themaat. INTRODUCTION TO THE LAW OF THE EUROPEAN COMMUNITIES: AFTER THE ACCESSION OF NEW MEMBER STATES. London: Sweet & Maxwell, 1973.

----------. INTRODUCTION TO THE LAW OF THE EUROPEAN COMMUNITIES: AFTER THE COMING INTO FORCE OF THE SINGLE EUROPEAN ACT. Deventer, Netherlands: Kluwer Law & Taxation Publishers, 1989.

Kennedy, Tom. The Essential Minimum: The Establishment of the Court of First Instance. 14 EUR. L. REV. 7 (1989).

----------. First Steps Towards a European Certiorari? 18 EUR. L. REV. 121 (1993).

Koopmans, Thijmen. The Future of the Court of Justice of the European Communities. 11 Y.B. EUR. L. 15 (1991).

----------. La Procédure Préjudicielle -- Victime de son Succèss? In F. Capotori et al, DU DROIT INTERNATIONAL AU DROIT DE L'INTÉGRATION -- LIBER AMICORUM PIERRE PESCATORE, 347-357. Baden-Baden: Nomos Verlagsgesellschaft, 1987.

Lasok, K.P.E. THE EUROPEAN COURT OF JUSTICE: PRACTICE AND PROCEDURE. 2nd ed. London: Butterworths, 1994.

Lenaerts, Koen. Form and Substance of the Preliminary Rulings Procedure. In Deirdre Curtin and Ton Heukels, eds., INSTITUTIONAL DYNAMICS OF EUROPEAN INTEGRATION: ESSAYS IN HONOR OF HENRY G. SCHERMERS. Vol. II, 355-380. Dordrecht: Martinus Nijhoff, 1994.

Mancini, G. Federico and David T. Keeling. From CILFIT to ERT: the Constitutional Challenge Facing the European Court. 11 Y.B. OF EUR. L. 1 (1991).

Pescatore, Pierre. Address on the application of Community law in each of the Member States, Judicial and Academic Conference of the Court of Justice of the European Communities, 27-28 Sept. 1976. Luxembourg: Office for Official Publications of the European Communities, 1976.

Rasmussen, Hjalte. The European Court's Acte Clair Strategy in C.I.L.F.I.T.; Or, Acte Clair, of Course! But What Does it Mean? 10 EUR. L. REV. 242 (1984).

----------. ON LAW AND POLICY IN THE EUROPEAN COURT OF JUSTICE: A COMPARATIVE STUDY IN JUDICIAL POLICYMAKING. Dordrecht: Martinus Nijhoff Publishers, 1986.

Schermers, Henry G., C.W.A. Timmermans, & A.E. Kellerman, eds. ARTICLE 177 EEC: EXPERIENCES AND PROBLEMS. Amsterdam: North-Holland, 1987.

Schermers, Henry G. and Denis Waelbroeck. JUDICIAL PROTECTION IN THE EUROPEAN COMMUNITIES. Deventer, Netherlands: Kluwer Law and Taxation Publishers, 1987.

Voss, Reimer. The National Perception of the Court of First Instance and the European Court of Justice. 30 COMMON MKT. L. REV. 1119 (1993).

Weiler, J.H.H. The European Court, National Courts and References for Preliminary Rulings -- The Paradox of Success: A Revisionist View of Article 177 EEC. In Henry G. Schermers, C.W.A. Timmermans, & A.E. Kellerman, eds., ARTICLE 177 EEC: EXPERIENCES AND PROBLEMS, 366. Amsterdam: North-Holland, 1987.

----------. Journey to an Unknown Destination: A Retrospective and Prospective of the European Court of Justice in the Arena of Political Integration. 31 J. COMMON MKT. STUD. 417 (1993).

----------. The Transformation of Europe. 100 YALE L.J. 2403 (1991).

Witt, Elder. CONGRESSIONAL QUARTERLY'S GUIDE TO THE UNITED STATES SUPREME COURT. 2nd ed. Washington, D.C.: Congressional Quarterly, Inc., 1990.

Official & other statistical sources

BASIC STATISTICS OF THE COMMUNITY. 28th & 29th eds. Luxembourg: Office for Official Publications of the European Communities, 1991 & 1992.

Office for Official Publications of the European Communities. SYNOPSIS OF THE WORK OF THE EUROPEAN COURT OF JUSTICE. Luxembourg: Office for Official Publications.

Registry, European Court of Justice. Statistiques Judiciaires. Luxembourg: Logiciel integre pour le traitement des informations du greffe, Feb. 1994.

Smythe-Wood, Isobel. THE EUROPEAN COMPANION 1994. London: Dod's Publishing & Research Ltd., 1994.

United Nations Statistical Division. STATISTICAL YEARBOOK. 39th issue. New York: United Nations Statistical Division, 1994.

United States Bureau of the Census. STATISTICAL ABSTRACT OF THE UNITED STATES: 1993. 113th ed. Washington, D.C., 1993.

World Bank. WORLD TABLES 1993. Baltimore: Johns Hopkins University Press, 1993.

Interviews
(to be put on file at Harvard Law School)

David A.O. Edward, Judge, Court of Justice, in Luxembourg (10 Nov. 1994).

Hendrick von Holstein, Deputy Registrar of Judicial Affairs, Court of Justice, in Luxembourg (10 Nov. 1994)

H. Jung, Registrar of the Court of First Instance, in Luxembourg (10 Nov. 1994).

David Keeling, Chambers of Advocate General Francis Jacobs, Court of Justice, in Luxembourg (10 Nov. 1994).

Tom Kennedy, Head of the Information Office, Court of Justice, in Luxembourg (9 & 10 Nov. 1994).

Koen Lenaerts, Judge, Court of First Instance, in Luxembourg (9 Nov. 1994).

Blanca Pastor, Registry of the Court of First Instance, in Luxembourg (10 Nov. 1994).

Christiaans W.A. Timmermans, Deputy Director-General, Legal Service of the European Commission, in Brussels (8 Nov. 1994).

Walter Van Gerven, Former Advocate General, Court of Justice, in Cambridge, Massachusetts (12 April 1994).

Ulrich Völker, Legal Service of the European Commission, in Brussels (11 Nov. 1994).

 

 


Questions or comments about this site?
Email Enfellows@exchange.law.nyu.edu

Top of the page