Do Preliminary References Increase Public Support for European Law? Experimental Evidence from Germany

Abstract

Explanations for the successful expansion and consolidation of the European Union and its legal system have long emphasized the importance of domestic courts’ sending preliminary references to the Court of Justice. Key to many of these theoretical accounts is the claim that domestic courts are better equipped than the Court of Justice to compel national governments to comply with EU law. Integrating insights from the comparative judicial politics literature into the context of the EU's preliminary references system, we provide a theoretical and empirical foundation for this claim by arguing that incorporating domestic courts into the EU legal process enhances public support for expansive judicial interpretations of EU law. We go on to argue, however, that this transfer of legitimacy depends on citizens’ views of the national and European courts. We support our argument with evidence from a preregistered survey experiment fielded in Germany.

Even with their seemingly ever-expanding role in modern politics, the fundamental question remains of how international courts can successfully compel national governments to abide by their international legal commitments.¹ Perhaps no international court has navigated this challenge as successfully, and come to wield as much influence and authority, as the Court of Justice of the European Union (CJEU).² From the regulation of trade and economic activity to the enforcement of democratic norms like the rule of law, the CJEU has a profound impact on the development of EU law and the policymaking of European governments.³ Beyond the EU, the Court has served as a template for many of the growing number of international and regional courts.⁴

Central to the CJEU's influence as a driver of European integration is the preliminary reference system. Indeed, this procedure, which allows national courts to first refer a case to the CJEU for a ruling on a question of EU law and then make a final decision, is credited with fundamentally transforming the breadth and scope of European law.⁵ By incorporating domestic courts, preliminary references bring the power of domestic courts into the EU legal process. This connection between the CJEU and domestic courts, in turn, enhances the efficacy of EU law and CJEU decisions by raising the political stakes for any member-state government that would consider challenging or otherwise refusing to comply with EU law. As Helfer and Alter note, “governments could not defy the CJEU without also calling into question the independence and authority of their own courts.”⁶

Such empowerment of EU law, however, crucially depends on the ability of domestic courts to overcome the inherent limitation of judicial power: the inability to directly implement decisions.⁷ A domestic court must itself be well positioned to overcome this challenge and compel compliance if it is to be an effective ally for the CJEU. Much like their international counterparts, however, domestic courts cannot presume the efficacy of their decisions but must rely on governments for their implementation.⁸ As a result, even national courts may find themselves constrained if compliance is not forthcoming, an issue which may be particularly likely in precisely those cases for which the CJEU most needs assistance in compelling government behavior. How domestic courts can overcome their own implementation problem, then, is a critical factor in the efficacy of preliminary references as conduits for national courts to promote the successful exercise of the CJEU's judicial authority.

We argue that judicial legitimacy is one pathway by which domestic courts can, through the preliminary reference procedure, enhance the efficacy of CJEU jurisprudence. Combining the institutional context of the CJEU with theoretical insights from the judicial politics literatures on judicial legitimacy and legitimation,⁹ we contend that preliminary references allow domestic courts to “lend” their legitimacy to CJEU decisions, thereby increasing public support for rulings. Using data from an experiment embedded in a survey fielded in Germany, we provide evidence that a domestic court—in our vignette the German Constitutional Court—deciding an issue of EU law based on the CJEU's ruling via a preliminary reference increases public support as compared to the CJEU's directly deciding the case. This legitimizing relationship, however, is limited to citizens who have more trust in their domestic courts than in the CJEU. Taken together, these results provide novel empirical evidence that the actions of domestic courts can directly affect citizens’ acceptance of EU law and contribute to our understanding of why the EU's preliminary reference procedure was so transformational: it increased citizens’ support for EU law.

The remainder of the paper is organized as follows. The next section discusses the preliminary references literature, focusing on the theoretical accounts linking the procedure with the empowerment of EU law. We then present our theory, followed by the research design. After presenting the results, we conclude with a discussion of their potential implications, both for the EU and for international legal regimes more generally.

Preliminary References and the CJEU's Empowerment

The preliminary reference procedure is the cornerstone of the EU legal integration process. Through referrals from national courts, the CJEU established fundamental doctrines such as direct effect (in Van Gend en Loos, allowing national courts to apply EU law themselves) and supremacy (in Costa v. ENEL, establishing EU law's precedence over the conflicting national laws of member states). These doctrines, in tandem, gave individual litigants standing before their national courts to assert their rights under EU law and provided the means for the CJEU to create new protections under the EU treaties. Although whether national governments were willing partners in the expansion of EU law is the subject of an extensive debate,¹⁰ scholars agree that national courts’ referrals were critical for the CJEU's promotion of legal integration. For example, a popular early explanation for its success was the “judicial empowerment thesis.” As Weiler explains, lower courts across Europe were eager to use the preliminary reference procedure because they were “given the facility to engage with the highest jurisdiction in the Community and thus to have de facto judicial review of legislation.”¹¹ Furthermore, the efforts of organized interests to leverage the preliminary reference procedure in national courts pushed EU law forward by providing the CJEU the opportunity to rule on a wide variety of issues.¹²

Additionally, preliminary references not only empowered lower courts but also allowed national courts to empower CJEU decisions—and by extension EU law. In their canonical article on the subject, Burley and Mattli argue that the CJEU “simultaneously strengthens its own legal legitimacy by making it appear that its own authority flows from the national courts. It is the national courts, after all, who have sought its guidance; and it is the national courts who will ultimately decide the case, in the sense of issuing an actual ruling on the facts.”¹³ This mechanism, the “empowerment of the [CJEU] with respect to the national court,” as they term it, critically depends on national courts’ having the ability to compel government compliance with their rulings. Alter, for example, assumes such compliance: “Once national judiciaries had accepted European law supremacy, national courts would not let politicians ignore unwanted [CJEU] decisions … National court support of [CJEU] jurisprudence effectively closed the option of exit through non-compliance with [a CJEU] decision.”¹⁴ Similarly, Weiler claims, “a state, in our Western democracies, cannot disobey its own courts.”¹⁵

While such accounts address how the CJEU can obtain compliance from reluctant member-state governments, they largely neglect that domestic courts have the same fundamental compliance challenge.¹⁶ Much like their international counterparts, domestic courts cannot presume the faithful implementation of their decisions, especially when it is contrary to the preferences of the government tasked with carrying out the decision. Crucially, this fundamental weakness of courts is present for even the most powerful courts in the most consolidated of democracies, including those of the EU.¹⁷ Thus explanations of the CJEU's success that depend on the efficacy of national courts must explain how domestic courts can further acceptance of EU law in the face of domestic political opposition.

One such explanation is the capacity of domestic courts to use their public support to legitimize, and thereby enhance public acceptance of, international legal decisions. Domestic courts, owing to factors such as socialization,¹⁸ the use of judicial symbols,¹⁹ or simply prudent decision making over time,²⁰ can develop a “reservoir of goodwill” with citizens.²¹ Even skeptical citizens with little support for the court can come to support its decisions. As Carrubba explains, “If the public observes enough good outcomes, the public will decide that violations of the regulatory regime's rules are not in its interests, and it will punish its government for defying the court … Simply put, if the public knows that compliance is usually in its interests, its government defying a court ruling usually is not in its interests.”²² This legitimacy makes noncompliance a politically costly strategy for governments by breeding citizens’ expectations that court decisions are followed and thus fostering compliance even with contentious or unpopular decisions.

Critically, this public support may empower courts to alter public opinion on the policies at issue in their decisions. This “legitimation” of policy, whereby public opinion becomes more favorable toward the position taken by a highly regarded court, makes domestic courts potentially influential opinion makers.²³ Importantly, scholars have recognized the linkage between citizens’ reactions to international court decisions and the political responses that follow them. Madsen and colleagues, who directly study public attitudes toward European court decisions, write that “the question of whether people are inherently more acceptant of decisions that are made by national rather than international institutions is more broadly relevant to understanding the public backlash against international institutions.”²⁴ Such shifts in public support of court rulings have direct consequences for government compliance. Gonzalez-Ocantos and Dinas note that “mass acceptance of rulings provides an important source of leverage because it complicates efforts by political entrepreneurs wishing to engineer backlash against judicial institutions.”²⁵ In short, citizens’ reactions to international legal decisions like those rendered by EU courts have direct and important implications for governments’ responses. Specifically, stronger public support for a decision resulting from domestic court involvement encourages compliance by both lowering the cost of compliance, as citizens are more accepting of the result, and raising the cost of noncompliance, as the domestic judiciary's legitimacy contributes to what Weiler refers to as the “enforcement value that national law will have on that occasion.”²⁶

Importantly, theory and corroborating empirical evidence exist for this connection between public attitudes, domestic courts’ legitimacy, and compliance. While shifts in the costs of compliance may induce compliance only to the extent that the matter is publicized and politicized, research has shown that courts, civil society, and political actors can give governments “a bloody nose” for not complying by affecting public attitudes.²⁷ Moreover, as Vanberg notes, the small electoral margins found in much of democratic politics makes the loss of even a small set of voters potentially damaging. Indeed, Krehbiel provides cross-national empirical evidence that noncompliance decreases government vote share in contexts with a strong norm of compliance, suggesting that such costs can in fact be realized.²⁸ Given the difficulty of directly evaluating a government's calculus for compliance,²⁹ scholars have provided a wealth of evidence that courts act strategically precisely to induce compliance through public support for their decisions. For example, Vanberg provides evidence that the German Constitutional Court is more likely to confront the government when it expects public awareness of its ruling, while Staton and Krehbiel demonstrate that the Mexican Supreme Court and the German Constitutional Court, respectively, pursue strategies to enhance that awareness.³⁰

To further illustrate this linkage, consider the litigation by the United Kingdom's Equal Opportunities Commission (EOC) on the issue of gender discrimination. Alter and Vargas argue that the EOC adopted a strategic-litigation strategy to convince domestic courts—specifically, industrial tribunals—to refer cases of gender discrimination in the British workplace to the CJEU after it “could not muster the support necessary to counter the Conservative government's antagonism toward equality policy.”³¹ After it won the Drake v. Chief Adjudication Officer case—brought by a woman denied government benefits designed for men who leave the workforce to care for a disabled person—at the CJEU and the domestic court applied the CJEU's ruling, the EOC “launched a large media campaign with grassroots meetings and conferences” that made it “politically too embarrassing for the supposedly family-oriented Conservative government to deprive [female] caretakers and the disabled from benefits.”³² By “exciting public attention with the successful expansion of the legal protections available to women workers and creating the potential of large costs for employers who discriminated on the basis of gender,” the EOC elicited policy change.³³ Specifically, it was the decision making of domestic courts—by referring cases and following CJEU rulings—that enabled the EOC to galvanize public support for policy change “when its other efforts at influencing the national political agenda failed.”³⁴

This union of the judicial legitimacy and legitimation scholarship, therefore, suggests that domestic courts should—through the preliminary reference procedure—have a unique capacity to affect public support for EU law. By incorporating national courts into the EU legal process, preliminary references allow the CJEU to effectively tap into national courts’ legitimacy and, thereby, both increase support for a policy position and raise the cost of noncompliance with the now-combined CJEU and national court decision. This dynamic is particularly advantageous for the CJEU, because it tends to have less legitimacy than its national counterparts.³⁵ While data on attitudes toward the CJEU are limited, the most recent Eurobarometer that asked about trust in both national legal systems and the CJEU found slightly higher trust in the former (53 percent versus 48 percent).³⁶ Simply put, the CJEU alone may not have sufficient public support, but it can rely on a national court that does.³⁷ If preliminary references have such a legitimizing effect, then we may hypothesize that citizens are more supportive of decisions on EU law made by their domestic court via the preliminary reference process than decisions made directly by the CJEU without the national court's involvement:

H1 Public support is higher for preliminary reference rulings issued by a national court than for rulings issued directly by the CJEU.

While largely at a disadvantage compared to their domestic counterparts, international courts like the CJEU may similarly develop a basis of public support.³⁸ This diffuse support functions much like that of domestic courts, in that it facilitates public acceptance of decisions, including those with which citizens disagree. If domestic courts legitimize EU legal decisions through the preliminary reference process, the degree of their impact likely depends on whether citizens are already willing to support a CJEU ruling or need to be convinced by their domestic courts to do so. Those who view the CJEU positively are already likely to support the Court's decisions, irrespective of national court involvement, thus limiting the impact of a preliminary reference. The legitimizing impact of a preliminary reference is likely greatest among those who hold their domestic courts in high regard but do not feel the same about the CJEU. For these individuals, their misgivings about a decision to expand EU law might be assuaged by the participation of a domestic court they trust.

H2 The effect of the preliminary reference procedure on public support for a decision increases as trust in the national court surpasses trust in the CJEU.

Experimental Evidence from Germany

The data we use to test these hypotheses come from an experiment embedded in a nationally representative survey of 1,205 Germans fielded from 3 to 20 September 2021 in partnership with YouGov.³⁹ Germany is an appropriate context for testing our hypotheses for several reasons. First, support for the EU and EU institutions, including the CJEU, tends to be strong in Germany, particularly in comparison to other Western European countries.⁴⁰ This higher baseline of support should make it harder to find evidence of a legitimizing effect from a preliminary reference, as respondents may be predisposed to support EU law irrespective of the court interpreting it. Second, public trust in the German judiciary is similarly high, making our theoretical mechanism of judicial legitimation of policy more plausible than it would be in a context with a highly politicized and maligned judiciary, such as Hungary.⁴¹ Such a context raises important theoretical questions that we consider in the conclusion. Third, Germany is the largest member state of the EU and particularly influential when it comes to the scope and depth of EU integration,⁴² making it a substantively important case for understanding the dynamics of public support for EU law.

Our experimental approach is centered on a vignette in which respondents are told about a hypothetical court decision that used EU law to respond to a hypothetical government policy.⁴³ Each respondent first read the following:

Imagine the federal government passed a law restricting the ability of unvaccinated citizens of other EU countries to enter Germany. The federal government contends that the policy is important to avoid a spike in coronavirus cases.

Respondents then were randomly assigned to one of two treatments designed to reflect either the direct involvement of the CJEU or the preliminary reference procedure, by which the CJEU's ruling is carried out through the decision of Germany's constitutional court, the Bundesverfassungsgericht. The former treatment read:

In response, the European Court of Justice ruled that the federal government's policy violates EU law guaranteeing the freedom of movement. As a result of the European Court of Justice's decision, the federal government must allow both vaccinated and unvaccinated EU citizens to enter the country.

The latter read:

In response, the Bundesverfassungsgericht asked the European Court of Justice whether the policy violates EU law. Based on the European Court of Justice's opinion, the Bundesverfassungsgericht ruled that the federal government's policy violates EU law guaranteeing the freedom of movement. As a result of the Bundesverfassungsgericht's decision, the federal government must allow both vaccinated and unvaccinated EU citizens to enter the country.

Four aspects of the vignette merit a more detailed discussion. First, we designed the vignette to be realistic by intentionally selecting a policy that was both salient and relevant for EU law. At the time of our survey, COVID-19 vaccines were widely available but still generating political controversy, particularly when it came to vaccine mandates and travel restrictions.⁴⁴ Moreover, such restrictions are directly relevant for EU law, particularly the freedom of movement, making it conceivable that such a policy would be challenged as in violation of EU law.

Second, our choice of the German Constitutional Court as the referring court, while a challenge to the realism of our vignette due to the Court's well-known reluctance to refer cases to the CJEU,⁴⁵ presents a “best case” scenario for demonstrating the plausibility of our hypothesized legitimizing effect of preliminary references due to the Court's uniquely high public support and familiarity.⁴⁶ In short, if any domestic court is likely to legitimize EU law, it is the German Constitutional Court. That said, even with recent research finding that apex courts are increasingly engaging in the preliminary reference process,⁴⁷ we acknowledge this limitation of our vignette's generalizability.⁴⁸

Third, we note that the legal decision is the same regardless of which court issues it. Respondents face the same policy outcome regardless of treatment; the only aspect that differs is the involvement of the national court in the preliminary references treatment.

Fourth, the preliminary references treatment is designed to reflect the nature of that procedure by explicitly informing respondents that the national court requested the CJEU's ruling and based its decision on that ruling. In doing so, the vignette makes clear the roles of both the national court and the CJEU. We label the treatment variable PreliminaryReferenceDecision, with a value of 1 for those respondents who read about the Bundesverfassungsgericht deciding in concert with the CJEU's ruling and 0 for respondents who read about the CJEU issuing the decision directly.

Immediately following the vignette, respondents were asked to answer three questions (each on a four-point scale) about their reaction to the court decision—questions that reflect our theoretical argument's emphasis on the linkage between public attitudes toward judicial decisions and government compliance. First, we asked respondents whether they agreed or disagreed with the decision. Second, we asked whether they accepted the decision as the final word on the matter or believed it should be challenged.⁴⁹ The third question was: “Imagine the federal Government refused to comply with the [ECJ's / Federal Constitutional Court's] decision and continued restricting the ability of unvaccinated citizens of other EU countries to enter Germany. Would you support this effort?” Overall, these three items form a reliable scale, with α = 0.70. The items all load on a single dimension, with three loading on this first factor at 0.73, 0.64, and 0.52, respectively. We use as our outcome variable the scores from this factor analysis, scaled from 0 to 1.⁵⁰ Higher values of the outcome variable, which we label SupportForEULawDecision, indicate a more positive attitude toward the decision. We expect more such support for decisions issued by the Bundesverfassungsgericht through the preliminary reference procedure (PreliminaryReferenceDecision = 1).

For our second hypothesis, we compare respondents’ confidence in the CJEU with their confidence in the German Constitutional Court. For both courts, respondents were asked whether they have “a great deal of confidence,” “only some confidence,” or “hardly any confidence.”⁵¹ Since our theoretical expectation is that the effect of a preliminary reference will occur among respondents who trust the domestic court—here the Bundesverfassungsgericht—more than the CJEU, we create a dichotomous variable (HigherConfidenceInBundesverfassungsgericht) that takes a value of 1 if the respondent has more confidence in the Bundesverfassungsgericht than in the CJEU, and 0 otherwise. Our expectation is that the effect of PreliminaryReferenceDecision will be greater among those with more confidence in the German court than in the CJEU.⁵²

Results

We analyze the results of our experiment using ordinary least squares. SupportForEULawDecision is the outcome variable and PreliminaryReferenceDecision the key explanatory variable, which we interact with HigherConfidenceInBundesverfassungsgericht to analyze the second hypothesis. The left pane of Figure 1 presents the direct treatment effects.⁵³ As predicted by our first hypothesis, a decision rendered by the Bundesverfassungsgericht, with the aid of a CJEU opinion, engenders significantly higher public support than the same decision issued directly by the CJEU. The difference is substantial: nearly one-fifth of a standard deviation.

FIGURE 1. Preliminary references and support for European Union law

The impact of a preliminary reference is even clearer when analyzing the conditional effect of respondents’ relative trust in the German court versus the CJEU. These results are presented in the right pane of the figure. Consistent with our second hypothesis, a reference increases support for the decision only among those who have more confidence in the German Constitutional Court than in the CJEU. These respondents’ support increased significantly, from 0.39 for a CJEU decision to 0.51 for a preliminary reference decision issued by the Constitutional Court; the difference is close to half of a standard deviation. In short, a preliminary reference helped win the support of those most skeptical of the CJEU.

We also note two other observations from these results. First, lack of confidence in the CJEU corresponds, unsurprisingly, to less support for a decision when it is made directly by the CJEU. Importantly, this relationship appears specific to the CJEU and not reflective of a broader distrust of courts, as respondents who distrust the CJEU did not exhibit significantly less support for a decision by the Bundesverfassungsgericht via a preliminary reference. Second, a preliminary reference had no statistical effect on support for the decision among those who trust the CJEU at least as much as the German Constitutional Court. In other words, referring a case appears to neither harm nor help the CJEU's standing among this subset of the population.

Discussion

Few aspects of the EU's legal system garner as much attention as the preliminary reference system. While this work largely focuses on the role of judges and political elites,⁵⁴ our account's incorporation of the public into the EU legal system's success reflects a growing recognition that citizens’ willingness to support and accept the EU's growing role will be necessary for the bloc's continued vitality.⁵⁵ Arguing that references enhance the efficacy of CJEU jurisprudence by allowing domestic courts to lend their legitimacy to the process, we provide a theoretical foundation for the procedure's impact that accounts for the political realities of both the CJEU and national courts. This advance, coupled with evidence from a survey experiment fielded in Germany, broadens our conceptualization of how preliminary references have influenced, and continue to influence, the successful expansion and development of EU law.

These findings have several implications. While comparative courts scholars have long recognized the role of public opinion in shaping judicial politics,⁵⁶ it is only more recently that they have begun to consider the pathways by which public support for international law and the decisions of international courts develops.⁵⁷ Our results suggest that, contrary to other recent research,⁵⁸ process can matter in determining how citizens react to a ruling on international law. Even if citizens generally care more about outcomes than about specific judicial procedures, they may be responsive to processes that integrate institutions with which they are more comfortable and familiar, such as domestic courts. This dynamic suggests that the best tactic available to international courts may not be legal doctrines or precedents,⁵⁹ but procedures that allow them to recruit allies from the domestic judiciary. With the EU serving as a model for many international legal systems,⁶⁰ other preliminary reference systems have the potential to provide a similar benefit to other international courts.

Second, our argument and results contribute to the evolving literature on preliminary references in the EU.⁶¹ By bringing the procedure's impact on public opinion into the literature, we highlight the continued relevance of references for an EU—and CJEU—facing numerous political, economic, and social challenges.⁶² As the EU's legal system has matured, so too has the preliminary reference system. For example, apex courts appear to be considerably more willing to engage with the CJEU through preliminary references than they were earlier in the EU's development.⁶³ To the extent that high courts enjoy greater public support than their counterparts lower in the judicial hierarchy, our argument suggests that this development could promote even greater public acceptance of an expanding EU legal order. However, our argument and empirical evidence do not represent the end of potential theoretical development on this topic. For instance, preliminary references’ legitimacy-conferring effect may depend on citizens’ awareness of decisions,⁶⁴ an issue scholars have recognized in the context of both international and domestic courts.⁶⁵ Thus courts’ and other legal actors’ communication strategies,⁶⁶ and citizens’ sources of information, may condition references’ ultimate efficacy. While we find that references can allow the CJEU to borrow national courts’ legitimacy, further research might adjudicate the precise conditions under which they do so.

Third, by focusing on judicial legitimacy, our argument raises the question of what, if any, impact preliminary references have when domestic courts lack public support. Where the domestic courts are just as distrusted as their international counterparts, references may have little capacity to grow public support. If, however, citizens hold an international court in higher regard than their national judiciary, then involving a domestic court in the legal process might weaken public support for international law by tainting the process. With the politicization of courts part of the global trend of democratic backsliding,⁶⁷ including in the EU,⁶⁸ such a dynamic could significantly undermine the capacity of international courts to enforce international law and uphold democratic norms.⁶⁹ The potential for such a relationship highlights the need for further research on the use and consequences of preliminary references in contexts like less consolidated and backsliding democracies.

Fourth, while we apply the study's theoretical framework to the EU, its focus on judicial legitimacy suggests an applicability to a broader set of institutional contexts. With similar, if not identical, preliminary reference procedures available to many international courts and tribunals,⁷⁰ a similar dynamic may come into play in systems beyond the EU. For example, Alter and Helfer note that the Andean Tribunal of Justice has a procedure that effectively mirrors the CJEU's preliminary reference.⁷¹ Our theory would predict that this could enhance support for Andean law, to the extent that member states have highly esteemed domestic judiciaries. More generally, with most of the non-European international courts that have preliminary reference procedures located in Africa and the Americas, a dynamic such as what we find here may often be more aspirational than currently observable, as support for many of those regions’ domestic courts is becoming increasingly conditional on partisan preferences.⁷² Nonetheless, to the extent that domestic courts have gained diffuse support in comparison to their international counterparts, our findings suggest that references should allow international legal systems to leverage that domestic judicial legitimacy. With this in mind, we hope that future research will expand beyond the EU to evaluate the extent to which theories related to preliminary references, including our own, are robust to the different political, economic, and social contexts in which other international courts operate.

We close by noting three important limitations of our study. First, our results come from a single country. While Germany's importance in EU politics, and its unique political culture, make it an appropriate case for our study, it is still only one of twenty-seven member states.

Second, our results are based on a vignette describing a single policy area, so we must be cautious in extrapolating the findings to other policy areas. On the other hand, we find strikingly similar results in an alternative experiment (presented in the appendix) that focuses on a core area of EU law, taxes. These additional experimental results give us some confidence that our findings are not merely a function of the specific policy presented in the vignette.

Third, the external validity of our findings is necessarily affected by the nature of our experimental design. By construction, the vignette informs all respondents of the EU law decision. As we mentioned earlier, public awareness of CJEU decisions and EU law is often modest, which places something of a scope condition on our argument. In this sense, our results might be described as representing a “best-case scenario” for the legitimacy-enhancing effect of preliminary references. But we do not see this as undermining the significance of our finding. Even the potential for such an effect can influence the behavior of governments concerned about political blowback. Moreover, awareness of cases is itself a function of strategic behavior on the part of political elites,⁷³ interest groups or other entities with a stake in a case,⁷⁴ and the media,⁷⁵ all of whom may have, under the right circumstances, the incentive to activate the legitimacy-borrowing property of references that we identify here. While these are theoretical and empirical extensions we are unable to grapple with here, we see our central findings as potential stepping stones toward understanding the impact of institutional features on citizens’ support for international law.