National Parliaments and the ECJ: A View from the Bundestag


Tracy Slagter
University of Wisconsin Oshkosh
800 Algoma Blvd, Oshkosh
WI 54901, USA
Tel +11 920 424 7161


The German government largely disregards impending decisions of the ECJ, even if those decisions have far-reaching implications for Germany. Decisions of the Court may, however, be useful for the opposition in securing its policy preferences. Equal treatment decisions are used to evaluate expectations about legislative anticipation of Court decisions.


What is the relationship between the European Court of Justice (ECJ) and national legislatures in the European Union? While we have substantial evidence from many authors that legislators are attuned to the jurisprudence of their own national courts and even, at times, alter legislation in anticipation of their national courts' decisions, we do not know much about how national legislators relate to the jurisprudence of the ECJ. I examine legislators in one European Union national legislature – the German Bundestag – and their anticipation of and reaction to decisions of the ECJ. Under what circumstances do legislators account for decisions or possible decisions of the ECJ and incorporate that information into their work? This article is limited to a discussion of the ECJ's rulings on equal treatment cases brought before it via the preliminary rulings procedure outlined in Article 234 of the Treaty.

This article proceeds as follows. Section I introduces the ECJ and the procedures that connect it to national legislatures, particularly the preliminary reference procedure described in Article 234 of the Treaty. Section II outlines the literature about the ECJ and national courts as well as the very limited literature on the ECJ and national parliaments. Section III outlines some expectations based on previous literature followed by a summary of the case studies used and the impact of the individual case evidence. I then discuss the behaviour of the majority and minority in parliament and the overall unwillingness to engage the ECJ before concluding.

I. The European Court of Justice and Article 234

The European Court of Justice holds a prominent place in the European Union web of institutions, and has maintained its role as chief interpreter of Community law since its establishment in 1951 with the Treaty of Paris. When states enter the European Union, they must accept the jurisdiction of the ECJ, and via Article 292 its competence is exclusive – Member States cannot submit claims to any other international authority. These are important provisions, since Community law has ‘direct effect’ in the Member States, meaning that all Community legal norms are the law of the land in the European Union, supplanting any ‘lower’-level national laws. Tied to this is the doctrine of supremacy, developed by the ECJ itself in 1964: ‘in the sphere of application of Community law, any Community norm, be it an article of the Treaty (the constitutional Charter), or a minuscule administrative regulation enacted by the Commission, “trumps” conflicting national law whether enacted before or after the Community norm’ (Weiler, 1999, pp. 20–1). The ECJ is given the competence to decide where the ‘sphere of application of Community law’ begins and ends (Kompetenz-Kompetenz).

Article 234 of the Treaty sets forth the Court's ‘preliminary rulings’ procedure. According to this article, national courts have direct access to the supranational European court whenever they encounter a case that requires interpretation of Community law; the Article obliges courts to seek the ECJ's opinion before proceeding if the referring court is a court of last instance.1 When such a case comes before a national court of last instance, that court is obliged to forward it to the ECJ in the form of a ‘preliminary reference’. The ECJ will not make a ‘decision’ on the case per se, but will instead interpret the Treaty for the national court and allow the national court to make the actual case decision in light of the ECJ's interpretation. In this way, the ECJ has transferred the implementation of its decisions to national courts, an innovative mechanism for increasing compliance with ECJ decisions on the assumption that Member State parliaments are more likely to conform to the decisions of their national courts than they are to obey the supranational court. The preliminary ruling has been one of the chief mechanisms through which Community law has been developed. Of the 9,052 cases brought before the ECJ between 1961 (when the procedure was established) and 1995, fully one-third of those were preliminary references (Dehousse, 1998, p. 30), and of those, the majority were cases that demanded the conformance of national with European law. It is in these cases, via national courts, that the ECJ is able to extend its reach into national legal systems and with it the scope of its jurisdiction.

Member State legislatures come into contact with the ECJ via their national courts. As noted, national courts at any level request preliminary rulings on a case when the case requires an interpretation of Community law, and often these cases question the ‘supranational constitutionality’ of national laws; in this way, the ECJ constitutes a second level of judicial review of national legislation. When the ECJ responds to a preliminary reference, it gives the national court guidance as to how to rule on the case (e.g. whether or not a challenged law conforms to the Treaty), and often that guidance directs courts to require that legislation be changed. In this way, the legislature responds to the ECJ through its own national courts. That is, the national courts require that the legislature responds to the ECJ in the same way it responds to its own national courts.

II. The ECJ and National Courts and Legislatures

Judicialization, the process by which legal argumentation is brought into the legislative process, is the consequence of the power of constitutional judges to affect legislative-sphere decision-making because their decisions can declare laws unconstitutional, sometimes even before the laws are promulgated. Under certain circumstances, judicialization is so intense that legislators, hoping to avoid censure from the court, alter their behaviour to comply with the court's potential ruling on a case – this is legislative anticipation. Judicialization thus sets the stage for legislative anticipation.

Figure 1 shows the different paths of judicialization in Europe. The relationship between national courts and national legislatures has been well documented, and is represented by a solid line in the figure.2 In Europe the creation of special constitutional courts, originally designed to keep the boundaries between courts and legislatures distinct, has in some instances resulted in the further blurring of these distinctions via judicialization. Also well developed is scholarship on the relationship between national courts and the European Court of Justice, noted by another solid line. This relationship is in many respects ‘automatic’, in that the ECJ harnesses the power of national courts to enforce its rulings via the preliminary reference procedure, and requires that national courts refer relevant matters of law to it. Much less studied, however (and represented here by a dashed line), is the relationship between national legislatures and the ECJ: many studies subsume discussion of national legislatures into discussions of national court–ECJ interactions, and this neglects the unique processes that are at work in the legislature as it grapples with its role between two powerful court systems, national and supranational. The legislature's position in this arrangement is such that the legislature is compelled to take into account decisions of both its national court and the ECJ as it formulates legislation. This also leaves open the possibility that legislators in national parliaments anticipate ECJ rulings and modify their behaviour and their product accordingly.

Figure 1.

Where Judicialization Occurs

There is a large literature on the European Court of Justice, and a significant portion of that literature deals with the relationship between the ECJ and national courts, neglecting or subsuming any discussion of the ECJ and national legislatures.3 But what of national legislatures and the ECJ? It is this arm of the triangular relationship drawn earlier that has been addressed by only a few scholars. For example, Eeckhout (2000) compares the ECJ's attitude towards the Community legislature and national legislatures. Is the Court even-handed in its treatment of national legislation and Community legislation? Eeckhout finds that, overall, the Court gives no deference to the national legislature when it reviews legislation:

The Court is concerned with ensuring the uniform application of Community legislation – and ultimately ensuring the primacy of Community law [. . .] it is obvious that a more deferential approach would be liable to undermine the uniform application of Community law and in many areas of Community legislation that is difficult to tolerate without putting in jeopardy the very aims which that legislation seeks to achieve. (Eeckhout, 2000, p. 8)

Indeed, Eeckhout explains that the Court must be more strict with the national legislatures simply because they are held to the entire corpus of Community law, whereas the Community legislature is bound only by the Treaty. That is, by joining the Community, Member States agree that all of Community law will be fully transformed into national law, and doing so is the responsibility of national legislatures. The Community legislature, however, must follow only the letter of the Treaty as it creates law, so it operates with fewer restrictions. This is perhaps why, when the ECJ gives a preliminary ruling on a case submitted by national courts, the ECJ is quite specific in its instructions to national courts about what should happen to a national law in question: ‘[. . .] the result is often that the referring court has no discretion when finally deciding if national legislation is to be set aside or not’ (Eeckhout, 2000, p. 3, author's own italics). This suggests that, although the ECJ's influence is procedurally ‘filtered’ through national courts before it touches the legislature, the filter is very weak. Although national courts intervene in the relationship between national legislatures and the ECJ, it seems safe to assume that their intervention has minimal impact on the supranational message sent to the national legislature. It also hints at the possibility of a contentious relationship between the ECJ and the legislature, since the two bodies are working for two different masters: the ECJ for Community law, and the legislature for national law.

The legislature–ECJ relationship is also the focus of Danny Nicol's research. His book EC Membership and the Judicialization of British Politics (2001) details the British decision to join the European Community. Nicol illustrates how British politicians were largely unaware of the implications of EC membership, and did not realize the extent to which British parliamentary sovereignty would be undermined by the ECJ. He notes, for example, how major ECJ decisions, decisions that were the building blocks of its power (for example, Van Gend en Loos), were not even noted in parliamentary debates or in the British press: ‘Thus the Community's judicial transformation from international organization to sui generis, proto-federal entity went unreported’ (Nicol, 2001, p. 38). Interestingly, Nicol also finds evidence that parliamentarians firmly believed that, even if the ECJ was very powerful, they could evade its decisions like other Member States seemed to do (Nicol, 2001, p. 42). The book reveals that, at nearly every turn in the development of Community law, British parliamentarians (as well as legal scholars, Whitehall's many lawyers and the media) had their heads in the sand and ‘a revolution occurred without political debate’ (Nicol, 2001, p. 252).

Aside from these few pieces, very little research has been done on the impact of ECJ rulings on national legislation and national parliamentary debate. There is a vast, unexplored territory to be covered in the gap between the literature on judicialization, which is almost (with the exception of Nicol) solely focused on national courts and national legislatures, and the literature on the relationship between the ECJ and national parliaments. The aim of this research is to begin closing that gap and to explore, at the parliamentary level, the process of judicialization from the supranational level and the circumstances under which that process leads legislators to anticipate ECJ decisions.

III. Uncovering Legislative Anticipation

Anticipation by the legislature is often also referred to as autolimitation, or legislative self-restraint. Autolimitation occurs when the governing majority ‘sacrifice[s] initially held policy objectives in order to [. . .] reduce the possibility that a bill will either be referred to the court, or be judged unconstitutional’ (Stone Sweet, 2000, p. 75). Autolimitation can be thought of as a complex game, where the actors in the game are the parliamentary majority, the minority and the courts, and the courts have the power of abstract judicial review.4 Only the court knows how it will rule; the minority and the governing majority can only speculate and use the court's past decisions (if precedent exists) as a guide to its potential decisions. The governing majority has incentive to pass legislation that fulfils its electoral promises, and the opposition has a similar incentive to block legislation that goes against its own electoral promises. Both the majority and minority want to maximize their chances of re-election, and neither of them wants to compromise. In the end, however, the majority may give up some of its policy objectives so that it does not come into conflict with the court, assisted by the minority that works to convince the majority that compromise is the best (least painful) solution.

The term ‘autolimitation’ is most frequently employed to describe a behaviour of a governing majority as it is drafting new national legislation. The majority, concerned that the new legislation might be struck down as unconstitutional in the future, refrains from drafting it in accordance with its own preferences and instead takes into account the preferences of the court. Here, I suggest that autolimitation can be extended to include not only instances in which the governing majority is drafting new legislation, but also to instances in which the government is reconsidering existing legislation. As governments alternate between parties or coalitions, for example, it is common for past legislation to be updated or revised according to the objectives of the new majority. That new majority has an incentive, then, to think about the court as it revises existing legislation and attempts to pass it anew. Governments bring the courts in when they explicitly reference previous rulings on issues that might impact that legislation at hand, or when they show evidence of having studied previous rulings in legislative debates,5 documents or in the media.

There is a difference in expectations about autolimitation depending upon whether or not the legislation under examination is new or existing. In the case of the ECJ and national parliaments, it is more likely that the governing majority will be examining existing legislation, since the ECJ's rulings are a reaction to national legislation that has already been passed. The governing majority might attempt to anticipate the ECJ if the legislation it is reconsidering is of national significance or has some cultural relevance. On the other hand, it might refuse to engage in any type of anticipation of the ECJ's decisions if it merely wishes to maintain the status quo – it might then simply wait until its hand is forced to make a change to existing legislation, paying little or no attention to the court.

The way the governing majority, in particular, plays the anticipation game is quite complicated, as Stone Sweet (2000) notes. On the one hand, the government may refuse to compromise on bills that are very important to it (and to its electorate), and therefore refuse to play to the court and engage in autolimitation. If it does this and the court strikes legislation down, the minority has secured an important victory and the majority is unable to fulfil its election promises. If the majority does compromise on important legislation, however, it may alienate some of its constituency, resulting in defeat at the polls in the next election. Because no one knows for certain how the courts will rule, the governing majority is often between a rock and a hard place. On the other hand, however, the governing majority can choose to compromise with the minority in order to please the court, and then blame the need to compromise on the court. If and when the compromised legislation is struck down, the majority can throw up its hands and proclaim that it did its best to save the will of the electorate from the hands of the court (Stone Sweet, 2000, p. 77). In the end, then, autolimitation is more art than science – winning the game is all in how the governing majority plays it.

Because the process of autolimitation is so complex and full of differing incentives for different actors, systematic accounts of how and when it occurs are few. Most accounts are forced to rely on anecdotes; while we are certain that autolimitation happens, ‘proving’ it is difficult. Landfried, for example, gives detailed accounts of incidents where legislative anticipation in the Bundestag of the German Federal Constitutional Court's rulings had a profound impact on the legislation that was passed. As legislators draft policy, Landfried notes that ‘politicians ask jurists to give them detailed interpretations of the decisions of the Court and sometimes the experts go in for “Karlsruhe astrology” ’ (1985, p. 528).6

In this article I explore how the governing majority and the minority in the German Bundestag react under the pressure put on them by the supranational European Court of Justice. Is the existence of a ‘supranational veto point’ enough to force the majority to reconsider its position on important issues (i.e. engage in autolimitation)? How does the jurisprudence of the European Court of Justice impact the tactics of the minority in parliament?

IV. Legislative Anticipation of the ECJ

The ECJ issues hundreds of preliminary rulings each year and through those preliminary rulings its preferences are revealed. Do legislators (both in the majority and the minority) have an incentive to determine the ECJ's position in advance of a possible court case? If so, under what circumstances? As noted above, literature on the relationship between national parliaments and national courts suggests that legislators should indeed have one ear cocked towards this supranational court.

There are several reasons why majorities and minorities in parliament should attempt to anticipate the decisions of the ECJ. One reason the governing majority should engage in anticipatory behaviour is because it is the chief guardian of Member State interests. The ECJ makes significant rulings that impact every facet of life in the European Union. There are some issues, however, on which it is not supposed to rule, issues that are reserved as exclusively within Member State competencies and that are not covered by any of the many treaties that are the basis of Community law. If a government suspects that the ECJ may be treading on those Member State competencies (which it has done in the past), the government has an incentive to anticipate a decision and work around it. If it does not, critical national interests may be at stake, and losing an important battle with the ECJ might not be good for the majority either in parliament or at the ballot box. This is especially true if the ECJ makes a ruling that impacts existing national legislation and would require substantial revision. Thus, when the government and the ECJ have different views on an issue, we should be more likely to witness anticipatory behaviour from the government. This would be especially true if the positions of these actors are very divergent. This argument assumes that the government wishes to maintain existing legislation (or has announced no specific plans to change it). The more the preferences of the government and the ECJ diverge, the more the government should, in theory, attempt to anticipate the ECJ's rulings in order to either conform to them or evade them.

The above applies to the governing majority if it is ‘protecting’ existing legislation. If the majority is introducing new legislation, however, it might also have an incentive (and perhaps a greater incentive) to anticipate decisions of the ECJ if those decisions could adversely affect the new programmes it is attempting to enact, as previous studies of parliament/national courts suggest. The majority might reconsider its position on a bill in order to conform to how it believes the ECJ will rule (based on the ECJ's previous rulings), or it might use the information it has about the ECJ to draft a bill in such a way that it appears to satisfy ECJ requirements but actually evades them.

On the other hand, previous literature has examined national parliaments and national courts, institutions bound by the same rules and constitutions. Frequently, as in Germany, members of the courts are chosen by members of the legislature and so the link between them is obvious. The ECJ, however, is not formally connected to any Member State parliament, parliaments cannot directly submit potential legislation to it for review, and thus the connection between the ECJ and national parliaments is much more ambiguous and distant. So it is also possible that the governing majority pays little attention at all to the ECJ and its rulings unless it is directly impacted (e.g., the ECJ makes a ruling that requires legislative action). If this is true, then the government will not engage in anticipation of the Court's decisions at all; unlike the pattern we see with national courts, the government might simply ignore the court and wait until it must act.

Importantly, the behaviour of both the majority and minority in parliament should vary depending on the importance and visibility of the case under consideration at the ECJ. Additionally, the significance of legislation that might be impacted plays into the government's incentive to anticipate the ECJ. The governing majority would likely have more at stake and be more willing to expend the effort to anticipate an ECJ decision if the case under consideration had wide-ranging consequences for the polity and if the public cared about the outcome of the case. I expect that when the case under consideration by the ECJ is high profile and is socially and culturally significant, we should see the governing majority actively attempting to anticipate ECJ decisions.

The majority–minority relationship comes into play when their known preferences are divergent on certain issues before them. If the rulings of the ECJ support a minority party's position, and if previous court rulings show a clear pattern in how the Court views such issues, a minority party can use the Court to monitor the majority. The minority serves as a watchdog, perhaps abusing minority rights to remind the majority of the wrong-headedness of its proposal and decrying its policy position (typical minority party behaviour). This is a threat to the majority: not only is the minority wasting valuable time in parliament by monitoring the majority's every move on an issue, but the minority might also bring negative attention to the majority that could damage its reputation. Increased publicity might also result, and this could then lead to a public backlash against the majority if the issue is salient enough to the citizenry. A party in the minority, then, can use compliance with the ECJ's past or potential rulings to intimidate the majority.

It follows that a party in the minority thus has a strong incentive to anticipate decisions of the ECJ, an incentive perhaps stronger than the majority party's. If a minority party can anticipate a ruling and use that potential ruling to support its own preferences (whether for new or existing legislation), it could make it more likely that the minority party will get what it wants in parliament because the ECJ is on its side, no matter what the position of the majority party. Thus, in addition to noting that a party in the minority is likely to monitor the majority's compliance with ECJ past and potential rulings when their preferences diverge, the minority is more likely to attempt to anticipate ECJ decisions as its preferences diverge from the majority's preferences. Anticipation by the minority is also more likely when its preferences are closely aligned with those of the ECJ.

V. Anticipating the ECJ: Case Selection

The next task of this article is to evaluate the evidence to determine if it is consistent with expectations derived from the literature on national court/national parliament relations. Using the German Bundestag as a representative of a strong Member State parliament, I have limited my focus to German equal treatment cases. I chose these cases from the hundreds of possible cases for this initial investigation for several reasons. First, there are many equal treatment cases to choose from, and, more so than other types of cases, the cases fall in a sequence that is more conducive to studying the legislative response to ECJ decisions. Second, I wanted sequences of cases that had different interpretations from case to case and equal treatment cases fit that requirement. I did not want to deal exclusively with cases that tweaked very minor aspects of a former decision, or that merely reaffirmed a previous ruling, which is the difficulty in selecting economic/competition policy cases. Third, it was important for me to select cases where there was variation in the governing majority, the opposition and the ECJ's position on an issue. I also wanted to include samples of cases where the parties wished to propose something outside the status quo and equal treatment cases provided that as well. Such a focus admittedly limits this study, but nonetheless provides a useful starting point for future, broader examinations of parliamentary reactions to the ECJ.

Given the above considerations I conducted detailed examinations of groups of legal cases in two broad issue areas submitted to the ECJ via an Article 234 preliminary reference from German courts.7 The group case studies and the ECJ decisions they focus on are as follows (ECJ case numbers in parentheses):

Equal Treatment and Access to Employment

Von Colson and Kamann v Land Nordrhein-Westfalen (C-14/83)

Dorit Harz v Deutsche Tradax GmbH (C-79/83)

Nils Draehmpaehl v Urania Immobilienservice OHG (C-180/95)

Eckhard Kalanke v Freie Hansestadt Bremen (C-450/93)

Hellmut Marschall v Land Nordrhein-Westfalen (C-409/95)

Equal Treatment and Military Service

Angela Marie Sirdar v Secretary of State for Defence (C-273/97)8

Tanja Kreil v Bundesrepublik Deutschland (C-285/98)

Alexander Dory v Bundesrepublik Deutschland (C-186/01)

In each group of cases in these two broad issue areas, I examined the ECJ preliminary ruling on the case as well as the report of the Advocate General (AG), which typically echoes the ECJ's eventual ruling and appears a few months in advance of the Court's decision. I also looked at the German laws in question; most of the time the laws under investigation were federal, but in a few instances they were state laws. Most importantly, however, I followed Bundestag debates and related parliamentary documents from the time prior to, during and after the ECJ's ruling to look for evidence that legislators had taken the ECJ's past or potential rulings into account. Finally, I interviewed key individuals (cabinet advisers, legislative assistants affiliated with the parties) at the German Bundestag in Berlin to get their general impressions of the relationship between the ECJ and the Bundestag.9

VI. Summary of Case Studies

I have conducted lengthy analyses of the cases listed above, carefully tracing the debates in the German Bundestag before, during and after the ECJ's ruling on the individual cases. Here I present an abbreviated timeline of the major events that occurred as the cases were decided and as they impacted (or did not impact) on behaviour in parliament.

Equal Treatment and Access to Employment: Sequence of Key Events

1984ECJ rules on Von Colson and Harz, noting problems with §611a BGB.10 In these cases, which were nearly identical factually and eventually received identical rulings, women were denied positions on the basis of their sex. The ECJ notes that §611a of the German Civil Code (BGB), which provides a weak remedy for the ‘loss incurred by candidates who are victims of discrimination as a result of their belief that there would be no discrimination in the establishment of the employment relationship’ is an insufficient sanction against discrimination.
SPD (Opposition) proposes bill to remedy §611a BGB issues.
1985ECJ rules on Commission v. Germany, accusing Germany of not adequately transposing the European Equal Treatment Directive into German law. §611a BGB is cited by the ECJ as being part of the problem.
1986SPD proposed remedy to §611a BGB rejected.
1988Greens (Opposition) propose remedy to problems with §611a BGB; their proposal also suggests the use of employment quotas in hiring.
SPD proposes first version of Equal Treatment Law, which also remedies §611a BGB.
1990Bundestag debates on SPD and Green proposals; both are rejected by CDU/CSU–FDP coalition majority.
1991Bundestag debate on SPD and Green proposals, which were reintroduced and carried over into the new legislative period.
1993CDU/CSU–FDP coalition government introduces Second Law on Equal Treatment, which does not remedy §611a BGB and discourages the use of quotas in hiring. Proposal references Von Colson, Harz and Dekker.
 SPD reformulates Equal Treatment Law to counter government proposal, with reference to Von Colson; new Equal Treatment Law recommends the use of hiring quotas.
1994Hearings begin before the ECJ on the Kalanke case. In this case, a man (Kalanke) applied for a job in the Bremen Parks Service. His competition for the post was a woman, and she was hired despite the fact that Kalanke was more qualified. A Bremen law states that in fields where women are underrepresented, women should be promoted before men.
No mention made of Kalanke proceedings by any party in the Bundestag.
1995AG delivers opinion on Kalanke in April; ECJ rules on Kalanke in October. Decision is that quotas for hiring were discriminatory and must be eliminated. Case is high profile. No discussion of AG opinion in the Bundestag between April and October.
SPD proposes bill on ‘Advancement of Women’ in the EU.
Extensive debate on Kalanke and SPD's proposal in the Bundestag. SPD and Greens are outraged by the decision and call for the government to encourage the European Commission to change the European Equal Treatment Directive. Other members cite the decision as ‘confusing’.
1997ECJ rules on Draehmpaehl, which again criticizes §611a BGB. Case involves a man who applied for a position advertised as for ‘an experienced female assistant’. ECJ focuses on the insufficient compensation for victims of discrimination provided for in §611a BGB, something it has already ruled on earlier but was never remedied.
 SPD introduces proposal to remedy §611a BGB, citing ECJ decisions.
 First full debate in Bundestag on the ramifications of ECJ decisions on §611a BGB and quotas; all parties finally willing to engage ECJ decisions, as well as potential decisions.
 ECJ rules on Marschall. Case involves a teacher who applied for a promotion in Nordrhein-Westfalen but was denied since women at his grade level were underrepresented. Nordrhein-Westfalen law states that women should be promoted where they are underrepresented unless reasons specific to a man's situation tilt the balance in his favour. ECJ rules that Marschall was not discriminatory since it
 contained a provision for promoting men (‘savings clause’). This was a surprising ruling given the Court's position on Kalanke. This decision opens the door for quotas to be used again in certain circumstances.
1998CDU/CSU–FDP coalition government submits bill to remedy the issues with §611a BGB; after debate in the Bundestag, the government's proposal is adopted.
1999New SPD–Green coalition government proposes new equal treatment law to replace CDU/CSU–FDP version. The new law includes quotas, and the SPD continually references the fact that they are on solid ground with recent ECJ jurisprudence.
2001Debate on SPD–Green proposal; proposal adopted.

Equal Treatment and the Military: Sequence of Key Events

1994Federal Administrative Court makes ruling on low-profile Bundeswehr (German Army) case. FDP (in coalition government with CDU/CSU) questions government about possible inequality/discrimination in Bundeswehr.
1998Kreil case is referred to the ECJ for review. The case involves a German woman who applied for a position as an electrical engineer with the Bundeswehr. Her application was rejected due to a provision in the German Basic Law (Germany's constitution) that excludes women from any military service that requires the use of a weapon (Article 12a). Women are restricted to the military band or hospital units. Kreil took her case to a German court, which then referred it to the ECJ. German court asked whether or not the constitution violates European law.
1999FDP (now in opposition) submits questions to SPD–Green government about prohibition on women serving in the Bundeswehr; SPD–Green Government does not respond.
 FDP proposes bill to amend Article 12a of the Basic Law; they are ridiculed in parliament but persist. FDP notes that Article 12a might violate European law.
 ECJ rules on UK-referred Sirdar, a British case involving a woman who wished to work as a cook in the Royal Marines. Her application was rejected. The ECJ upheld the rejection, noting that discrimination on the basis of sex in this case was fundamental to the ‘interoperability’ of the Royal Marines. This was consistent with Member States' expectations that the ECJ would not intervene in the Member State defence organizations. After a short time, AG delivers opinion on Kreil.
 FDP's proposal debated in the Bundestag; FDP references ECJ several times.
2000ECJ rules on Kreil. Decides that the Basic Law contravenes European law and must be changed, as Basic Law Article 12a is discriminatory on the basis of sex. This surprises many, given the Sirdar ruling of just a year earlier. Kreil case is very high profile.
 Legal Affairs Committee holds public hearing on Article 12a.
 FDP requests debate on the future of the Bundeswehr; party positions shift towards consensus on Article 12a in the wake of the ECJ decision.
 SPD–Green government proposes bill to remedy Article 12a, reflecting a complete reversal of previous opinions on the issue.
 SDP–Green Government, FDP and CDU/CSU submit bill to change Article 12a; government bill is adopted with consensus from all parties.
2001Dory case brought before the ECJ. In this case, a young law student maintains that the compulsory service requirement is discriminatory against men, since after Kreil women were allowed to voluntarily serve. During the time this case comes forward, the first women are allowed to serve in the Bundeswehr combat units as a result of Kreil.
2002AG delivers opinion on Dory, noting that the compulsory service requirement is not discriminatory and that the ECJ should stay out of debates on how the Member States organize their defence operations. This is contradictory to what it did in Kreil.
2003ECJ rules on Dory, consistent with the AG's opinion. Dory case is never discussed in the Bundestag, despite the fact it brought up issues related to another long-standing issue in the German military, the compulsory service requirement (Wehrpflicht).


Access to Employment Cases

I initially expected to find evidence of the majority's anticipation of an ECJ decision when the majority's position is distant from the position (or perceived position) of the ECJ, and furthermore that the majority should engage in anticipation when the case has a high profile. This series of access to employments cases is not a tidy ‘test’ for those expectations, since the government's position was distant from the ECJ's on the issue of §611a BGB but the government's position was exactly that of the ECJ on the issue of quotas. In any case – whether the government and the ECJ were in agreement or not – at no time during the more than decade-long debate on discrimination in access to employment did the government (the CDU/CSU–FDP for most of the time period under investigation) attempt to anticipate or even demonstrate concern for an impending ECJ decision in a serious fashion, even when the much-publicized Kalanke case was under consideration by the Court. Indeed, the only time the government appeared to be looking ahead to a supranational decision was in the brief period after Kalanke but before Marschall, when the government used Advocate General Jacobs' opinion on the Marschall case (with which the ECJ ultimately did not agree) to bolster its argument that quotas in hiring were illegal. Had the government been more attuned to the jurisprudence of the ECJ, it could have attempted to squelch SPD/Green proposals for hiring quotas from the outset. The Kalanke decision, in particular, would have played right into the government's hands, and the publicity surrounding the case and its outcome would have been highly beneficial to the government. Given that the Kalanke decision completely conformed to the government's views on the use of quotas, the fact that they did not even acknowledge that it was going on shows the ‘wait and see’ approach the government takes towards the ECJ. Although it is true that Kalanke was in many respects a landmark decision on the issue of hiring quotas, it was nonetheless still one case in a line of important cases involving the EU's Equal Treatment Directive that the government could have taken into account (see, for example, Dekker, Von Colson, Harz). Additionally, the ECJ had developed a ‘trend’ of ruling against national provisions (and especially German provisions) on equal treatment issues 70 per cent of the time (see Cichowski, 2004), a fact the government is not likely to have missed.

The government's behaviour is not surprising given its behaviour early in the debate about remedies to discrimination and §611a BGB. For over a decade, the government not only refused to change the deficiencies to the BGB that were clearly and undeniably the target of the ECJ's decisions in Von Colson and Harz, but it neglected even to admit that there was a problem. It did not refer to the decisions in parliamentary debates and it did not cite the decisions in the proposals it drafted through the late 1980s and early 1990s. It was not until 1993, for example, that Von Colson and Harz even entered the government's rhetoric, a full ten years after those cases were decided. Indeed, it was not until a critical mass of cases (Von Colson, Harz, Kalanke, Draempaehl) were decided on the issue of discrimination in access to employment that the ECJ had the government's full attention, and then only because the SPD (opposition) had plenty of ammunition for an ECJ-based attack. In the post-Kalanke era, the government's attitude towards the ECJ in parliament was more one of lip-service to the ECJ than true respect for that Court's rulings. In effect, the government justified its inaction on ECJ rulings by noting that it was merely waiting for more ECJ rulings – the approach was never proactive, but rather ‘wait and see’.

This group of cases seems to conform to some expectations about the minority party's behaviour, namely, that the minority party will both monitor the government's compliance with ECJ decisions and attempt to anticipate the ECJ's decisions when the minority's preferences are divergent from those of the majority. The reconstruction of events in this case makes it very clear that the parties of the opposition were especially eager to ‘monitor’ the government's responses to ECJ decisions and this occurred from the very beginning of the discussion about discrimination in access to employment. The SPD was particularly engaged in the jurisprudence of the ECJ as it repeatedly called the government to account on fixing the provisions in §611a BGB, and was later joined in its monitoring of the government by the Greens. The SPD's reasons for doing so are fairly self-evident: by questioning the government's responses to key ECJ decisions, the opposition hoped to expose an important hole in the government's grander plan for an equal treatment law. The opposition then hoped that the hole could be filled by its own policies, which were self-consciously attuned to ECJ jurisprudence.

Importantly, however, there is surprisingly little evidence of anticipation on the part of the minority here, especially on the part of the SPD. There were plenty of opportunities for the SPD to attempt to predict how the ECJ was going to rule on Kalanke – indeed, the case received a lot of press11 and the SPD was already actively engaged in policy-making on the exact same issue (as were German courts). The SPD could have used the decision to its advantage, and could have attempted to achieve policy goals by writing legislation that would have used quotas in a way not prohibited by the ECJ's ruling in Kalanke. In April 1995, six months before the ECJ made its decision, Advocate General Tesauro argued that ‘the right of the individual not to be discriminated against on grounds of sex should not yield to the rights of the disadvantaged group in order to compensate for past discrimination suffered by that group’ (Fredman, 1998, p. 203). If the SPD knew that the Kalanke case, which resulted in a ruling against the very type of quota system the SPD was promoting, had been pending before the ECJ since 1994, Bundestag debates show no evidence of it. Indeed, in the months leading up to the Kalanke decision, the SPD pushed harder than ever for quotas.

Equal Treatment in the Military Cases

‘The ECJ is an unpopular court here’, the Ministerialrat (head of a ministry department) I interviewed said in an interview at the Bundestag in November 2005.12 He then explained that the Kreil decision, in particular, was one that legislators were not impressed with, given that many (despite ‘inter-party consensus’ rhetoric in parliament) believed that the ECJ had overstepped its boundaries by intervening in matters traditionally left to Member State discretion (especially in light of the Sirdar decision). He also noted that, despite the fact that the Bundestag receives tens of thousands of European Union-related documents from various institutions of the EU each year, legislators as well as members of the public were not well informed about the European Court of Justice or about the significance of its decisions for German politics.13

A member of the research staff for one of the Bundestag's Fraktionen was similarly sceptical about legislators' information on the ECJ: ‘We have such a diversity of information’, she said, ‘and sometimes issues are too complicated at the national level to pay attention to what is going on at the European level[. . .]. It's not that we don't care about the ECJ or what it does. It's a problem of capacity’.14 The information overload could lead one to believe, on the surface, that legislators do not or could not attempt to anticipate the decisions of the ECJ or use those decisions in any strategic manner. But when asked if she thought that legislators attempted to anticipate the decisions of the ECJ, the same research staffer replied, ‘Oh, most certainly we do. But it isn't easy’.

The evidence in this case suggests, at least in the highly publicized ECJ ruling on Kreil, that some legislators do engage in some form of legislative anticipation, and that they do indeed attempt to use the Court and its eventual decisions to influence legislative outcomes in parliament. The existence and decisions of the Court create new opportunities for parties to secure advantages in parliament, advantages they would not have had without its influence, no matter how indirect. The decisions of the ECJ have a very indirect relationship to the work that goes on in the Bundestag: the ECJ's decisions are funnelled through other bodies, notably the national court system, before they have any bearing on legislation. It is undoubtedly more work for the legislator to attempt to anticipate this Court's decisions or use them strategically, and the payoff for successfully accomplishing either task is smaller (due to, for example, the fact that there is no direct electoral connection, and it is unlikely that voters would decide a party's fate based only on that party's reaction to an ECJ decision). But even a small payoff might be worth the amount of legwork it takes to stay a step ahead of the ECJ for certain legislative players. In this case, keeping one ear in Luxembourg certainly helped a small party, the FDP, accomplish a goal it had striven towards for over a decade.

The FDP was engaged in the ECJ's decision on Kreil almost from the beginning, and this was due to the fact that the issue of women in the Bundeswehr was something the FDP had on its agenda since the late 1980s. In fact, they had brought the issue to the Bundestag's attention a few times before, and each time their proposal was struck down without much debate. A member of the FDP's research staff commented that ‘the FDP typically receives ECJ decisions very warmly’ because ‘they push Germany in the right direction’.15 She also noted that although staying one step ahead of the rest of the Bundestag by attempting to anticipate what the ECJ might decide on an important issue is difficult, the difficulty ‘doesn't limit the party – we want to push the boundaries of legislation and interpretation’.16 Using the impending decision on Kreil allowed the FDP to do exactly that, to achieve a goal in parliament they had not been able to achieve in the past. The FDP's behaviour here demonstrates that as the ideological distance between the majority and the minority increases, the minority is more likely to attempt to anticipate ECJ decisions.

What is interesting in this case is the disinterest in the ECJ decision displayed by the majority parties; indeed, it does not conform to expectations about governing party behaviour. In research on legislators and their relationship with the German Federal Constitutional Court (FCC), the leaders of the governing coalition are generally quite concerned with how the FCC might rule so that they can write legislation to avoid having to face the court in the future. That behaviour is not displayed here – in general, the governing coalition preferred to wait for the actual decision of the ECJ to come down, to not do anything too rash beforehand. This mirrors the behaviours of the CDU/CSU in the access to employment cases. The reasons for the difference in legislator behaviour towards the ECJ and the FCC became more clear during the course of my interviews in Berlin: first, several members of the Bundestag's staff (both in the administration and in the various Fraktionen) indicated that the ECJ and its decisions felt very distant from the Bundestag, very much on the outskirts of the work the Bundestag does. Although the ECJ has been around and handing down decisions that affect Germany almost as long as the FCC has, there are ‘simply not enough data points’17 for legislators to have a tight relationship with the ECJ and be familiar with the intricacies of its jurisprudence. This leads to the second point, which is that parties in the governing coalition do not have the time to spend wading through complex ECJ-related material when they are responsible for the entirety of the national government, especially considering that ECJ decisions, first, do not always impact national legislation; and, second, have a circuitous path to the Bundestag when they do impact legislation. This points to a potential advantage in anticipating ECJ decisions for smaller parties in the opposition, who can focus their research energies where they wish, and on issues that are of the greatest concern to them.

From these few cases, then, it is clear that while governing majorities might indeed have a theoretical incentive to think about the ECJ as they revise old legislation or consider new legislation, the actual incentive is lacking. The government simply does not feel threatened enough by the ECJ to take proactive measures, very unlike the relationship it has with the German FCC. The motivation to consider the ECJ is further hampered by the difficulty in determining how the Court might rule; indeed, it might even be the case that the ECJ is significantly ‘less predictable’ than other courts given the vast array of cases it decides as well as the multitude of unique national legal provisions with which it must contend.


The ECJ has significant power to turn national legislative provisions on their heads, be it by requiring a full-scale reorganization of a national military or by condemning certain types of hiring quotas as not in conformance with Community law and upholding others. If the government, as displayed in the case studies presented here, perceives a disconnect between the parliament and the ECJ, then it is ill-equipped to predict what the Court might do and react in a meaningful way. This in turn essentially means an even more decreased role for national parliaments in the European Union, since they are not willing or able to fully assess the ramifications of an ECJ ruling before it is already upon them. This finding echoes the conclusions on the British parliament illustrated by Nicol (2001).

In this study, legislators are not so much unaware of the power of the ECJ, however, they are simply unwilling to engage it. This unwillingness to engage the Court has been noted elsewhere. Mancini and Keeling (1994), for example, note the anxiety about the Court that seems to pervade the German political elite: ‘[. . .] their [German authorities] irritation is fuelled by [. . .] the impression that the Court is one-sided; that, faced with a choice between the interests of the Member States and the interests of the Community, the Court systematically favours the latter’ (Mancini and Keeling, 1994, p. 186). The fact that the German government always waited for decisions from its national courts before taking any action on ECJ decisions is further evidence of this.

More heartening is the evidence about the persistence of the minority to inject the ECJ and its past or potential decisions into parliamentary proceedings. Even though the legislative minority uses the Court for its own gain, it nonetheless serves as a reminder of the Court's power and the need for Member States to act on it. This was displayed quite clearly by one legislator in the access to employment cases. Incensed by the Court's decision on Kalanke, MdB Kerstin Müller (Greens) charged the government to ‘act now on both the federal and European levels’ to work around the Court's decision (64. Sitzung des Deutschen Bundestages, 26 October 1995, 5517). Müller's colleague MdB Irmingard Schewe-Gerigk (Greens) echoed her sentiment: ‘Ladies and Gentlemen of the Government, we have here before us a call for action!’ (64. Sitzung des Deutschen Bundestages, 26 October 1995, 5525). The legislative minority, by monitoring the government's responses to ECJ decisions and using those decisions to further their own policies, play a meaningful role in keeping the government at least somewhat tuned in to ECJ matters.

Linking the national parliaments with the European Court of Justice is a task only Nicol (2001) has undertaken to date. Studying European politics of any kind without taking into account the European level of governance means missing an important piece of the political process. Through study of the relationship between national parliaments and the ECJ, we can make new inquiries into the importance (and severity) of the democratic deficit in the EU,18 contribute to the ongoing supranationalist v. intergovernmentalist debate, and also make better predictions about the future of parliaments under the proposed EU constitution.19 As Nicol (2001, p. 265) writes, ‘we may well be at the beginning of a fascinating story’ as we discover the new strategies, incentives and costs that come with adding an extra level to parliamentary behaviour as we know it.


  • 1

    Any court may refer cases to the ECJ for a preliminary ruling, but only courts of last instance have an obligation under the Treaty to do so.

  • 2

    See, for example, Landfried (1984, 1985, 1988, 1995), Stone (1992), Stone Sweet (2000), Volcansek (2001) and Vanberg (2001, 2005).

  • 3

    See, for example, Slaughter et al. (1998), Weiler (1999), Alter (1998, 2001), Garrett (1992) and Garrett et al. (1998).

  • 4

    Autolimitation effects are not only found between legislature and court, however. Scholarship by Manow and Burkhardt (2004), for example, shows how divided government (differing majorities in the Bundestag and Bundesrat in Germany) leads to frequent autolimitation by the governing majority.

  • 5

    Stone notes that governments may give a bill's ‘jurisprudential history’, or a statement to the legislature that indicates that ‘the government had already been through a significant, if private, autolimitation process[. . .]’ (1992, p. 125).

  • 6

    The Federal Constitutional Court is located in Karlsruhe.

  • 7

    An analysis of an additional sequence of cases related to equal pay is not included in this article, but is available upon request.

  • 8

    Admittedly, this is not a case that originated in a German court, but it has such significance for the later cases that did begin in German courts that I felt it was appropriate to include it here.

  • 9

    The interviews took place in Berlin during November 2005.

  • 10

    BGB =Bürgerliches Gesetzbuch, or the German Civil Code.

  • 11

    When a decision on the Kalanke case was imminent, German newspapers spoke of the ‘death of the Bremen women's quota in Luxembourg’, noting that when ‘the European Court of Justice rules in September, the chances [for the quota] will be slimmer than expected’ (Ase, 1995). One reporter noted that the Kalanke decision seemed to take everyone by surprise: ‘The enthusiastic engagement at the national level, where the theme of the women's quota was discussed [. . .] stood in stark contrast to the proceedings of the ECJ, of which few seemed to take notice[. . .]’ (Clever, 1995).

  • 12

    Personal conversation, 11 November 2005.

  • 13

    See Gibson and Caldeira, 1998. Indeed, when I asked my interview subject about which legislators he thought were most attuned to ECJ decisions, he said, ‘I can think of [. . .] five people’. Five legislators out of more than five hundred! (Personal conversation, 11 November 2005)

  • 14

    Personal conversation, 18 November 2005.

  • 15

    Personal conversation, 18 November 2005.

  • 16

    Personal conversation, 18 November 2005.

  • 17

    Personal conversations, 11 November 2005 and 18 November 2005.

  • 18

    See, for example, Moravscik (2002).

  • 19

    See Cooper (2006) for a discussion of the possible role of national parliaments in the Constitutional Treaty.