¹ The judicon research project has been supported by the Incubator Grant of the Hungarian Academy of Sciences (Centre for Social Sciences) and the Pázmány Péter Catholic University (Faculty of Humanities and Social Sciences KAP16-1.1-1.6 Grant). Homepage: www.judicon.tk.mta.hu/en Google Scholar

² See Sweet, Alec Stone, Governing with Judges: Constitutional Politics in Europe (2000); Ran Hirschl, Towards Juristocracy: The Origins and Consequences of the New Constitutionalism (2004); Tom Ginsburg, The Global Spread of Constitutional Review, in The Oxford Handbook of Politics and Law 81, 81–99 (Whittington, Keith E., R. Daniel Kelemen & Caldeira, Gregory A. eds., 2008).Google Scholar

³ See Schwartz, Herman, The Struggle for Constitutional Justice in Post-Communist Europe (2000); Radoslav Prochäzka, Mission Accomplished: On Founding Constitutional Adjudication in Central Europe (2002); Wojciech Sadurski, Rights Before Courts: A Study of Constitutional Courts in Postcommunist States of Central and Eastern Europe (2014); Benjamin Bricker, Visions of Judicial Review: A Comparative Examination of Courts and Policy in Democracies (2016).Google Scholar

⁴ See Wojciech Sadrurski (ed.), Constitutional Justice East and West (2002); Otto Luchterhandt, et al. (Hrsg.), Verfassungsgerichtsbarkeit, in Mittel- und Osteuropa (2007).Google Scholar

⁵ But see Bricker, supra note 3. As Bricker's work, however, focuses on the constitutionality and unconstitutionality of legislative acts, it cannot answer the question in the focus our own research: To what extent is the legislature constrained by the constitutional court?Google Scholar

⁶ See Kis János: Alkotmányos demokrácia (2000); Győrfi Tamás, Az alkotmánybíráskodás politikai karaktere (2001); Tóth Gabor Attila: A szövegen túl. Értekezés a magyar alkotmányról (2009).Google Scholar

⁷ See Boulanger, Christian, Hüten, richten, gründen. Rollen der Verfassungsgerichte in der Demokratisierung Deutschlands und Ungarns (2013); Halmai Gábor, In memoriam magyar alkotmánybiráskodás, in Fundamentum 36–64 (2014); Szente Zoltán, The Political Orientation of the Members of the Hungarian Constitutional Court between 2010 and 2014, in Constitutional Studies 123–49 (2016).Google Scholar

⁸ Selected Central and Eastern European countries included in the investigation are Albania, the Czech Republic, Germany, Hungary, Poland, Romania, and Slovakia.Google Scholar

⁹ When presenting the methodology, we concentrate on the case of Hungary to be able to demonstrate all nuances of the methodological questions. Although some differences in selecting and appointing judges—or in determining the clear-cut cleavages between right-wing and left-wing political camps and governments—exist within the Central and Eastern European region, these variations might not prevent the adaptation of the methodology to other Central and Eastern European countries. There might perhaps be more grey zones in some cases, but they do not attenuate the relevance of the method.Google Scholar

¹⁰ The JUDICON project involves researchers from seven countries: Albania, the Czech Republic, Germany, Hungary, Slovakia, Poland, and Romania. More details are available on the project's webpage: www.judicon.tk.mta.hu/en.Google Scholar

¹¹ Our research focuses on the relationship between constitutional courts and legislatures. Consequently, legal regulations that are relevant for the present research are laws adopted by parliaments. This does not preclude, however, the application of the methodology presented below in another study to examine—for example, the extent to which constitutional courts have limited the scope of governmental activity. Judicial decisions on government decrees might be also examined by using the methodology presented in this paper.Google Scholar

¹² Although this article is limited to the attitudinal model, we think the methodology may be applied to other models. We consider other ones, especially the strategic model as a possible next step for the wider research.Google Scholar

¹³ See Dyevre, Arthur, Unifying the Field of Comparative Judicial Politics: Towards a General Theory of Judicial Behaviour, Eur. Pol. Sci. Rev. 297–327 (2010).Google Scholar

¹⁴ See Christoph Hönnige, Beyond Judicialization: Why We Need More Comparative Research About Constitutional Courts, Eur. Pol. Sci. 346–58 (2010).Google Scholar

¹⁵ For an illustrative overview over the differentiation of judicial decisions, see Maartje de Visser, Constitutional Review in Europe: A Comparative Analysis (2014).Google Scholar

¹⁶ See Allan Brewer-Carías, Constitutional Courts as Positive Legislators: A Comparative Law Study (2013).Google Scholar

¹⁷ See Tsebelis, George, Veto Players: How Political Institutions Work (2002).Google Scholar

¹⁸ See Volcansek, Mary, Constitutional Politics in Italy (2000).Google Scholar

¹⁹ See Christoph Hönnige, Verfassungsgericht, Regierung und Opposition. Die vergleichende Analyse eines Spannungsdreiecks (2007).Google Scholar

²⁰ See Ganghof, Steffen, Promises and Pitfalls of Veto Player Analysis, Swiss Pol. Sci. Rev. 1–25 (2003).Google Scholar

²¹ de Visser, supra note 15, at 309.Google Scholar

²² Bricker, supra note 3, at 62.Google Scholar

²³ See Sólyom László & Georg Brunner, Constitutional Judiciary in a New Democracy: The Hungarian Constitutional Court (2000).Google Scholar

²⁴ After detailed consideration, within our framework, we only consider laws, while lower level regulations —for example, decrees—or court decisions are excluded because they would exceed the scope of the research.Google Scholar

²⁵ Refusals will be coded only if there is a dissenting opinion—for example, the judge claims that the constitutional court should have decided the case on its merits—because refusals are mainly based on technical arguments—for example, the court could not deal with the case due to some formal deficiencies. In controversial cases, there will certainly be a dissenting opinion arguing that the court should have dealt with the case on its merits.Google Scholar

²⁶ One might argue that finding an omission is a more severe interference with the competencies of legislation, therefore some clarification is needed for why we conceive unconstitutionality by legislative omission as a mild form of ruling. First of all, we have to stress that a decision based on legislative omission does not concern the legal system in force—for instance, it does not annul any of the existing legal norms. Furthermore, even if the court assessed an omission, but does not propose any solution to how unconstitutionality might be corrected, the legislature is not under direct pressure to act at all. If the court's decision does not contain any threat or possible sanction—which is usually the case in Central and Eastern European countries—the legislature might simply neglect or disregard the ruling of the court, and status quo might be preserved. In this case, the court simply suggested legislative action, but if it doesn't want to change the status quo nothing will change. This is why we argue that a ruling which found a statute unconstitutional by referring to legislative omission does not really compel the legislature to act. By contrast, constitutional requirements and, certainly even more, substantive unconstitutionality means that status quo will be automatically changed by the decision of the court.Google Scholar

²⁷ We know that sometimes it is quite difficult to make an unambiguous distinction between procedural/formal unconstitutionality and substantive unconstitutionality. The term rule of law might serve as a transition point between formal unconstitutionality and substantive unconstitutionality. We have considered that we should include any principle and connotations derived from the rule of law that do not exclude a repeated adoption of the legislative regulation with the same substance as a procedural unconstitutionality. Eventually, we decided, however, to code all references to the principle of rule of law or any derivation of it—like accessibility, clarity, calculability, necessary time provided for preparation to comply with the new law, legal certainty and taboo of retroactivity—as substantive unconstitutionality. At the end of the coding process, we will return to these special cases and re-evaluate them one by one in order to clarify whether the decision based on the principle of rule of law excluded a repeated adoption of the legal regulation with the same substance. A further problem emerges concerning the taboo of retroactive legislation in the context of democratic transitions. Because sensible questions of transitional justice definitely collide with this fundamental principle of rule of law, procedural/formal unconstitutionality based on the taboo of retrospective legislation might have a highly restrictive effect on legislation in times of democratic transition. This is why decisions implying conflicts between the principles of the transitional justice and the rule of law should be very carefully analyzed.Google Scholar

²⁸ Because our research aims to analyze judicial decisions quantitatively, we have simply focused on the presence or absence of constitutional requirements and have not evaluated the strength of these requirements one by one. Further qualitative investigation will be needed to refine the picture regarding the strength of these constitutional requirements. At this phase of the project, we are focusing on the formal presence of constitutional requirements.Google Scholar

²⁹ Beyond the German Federal Constitutional Court, no court in Western Europe has the competence to declare an abstract and binding interpretation of constitution without reviewing a law filed to the court. By contrast, this kind of constitutional adjudication is not unknown in Central Europe although even courts in Central and Eastern Europe have rarely been asked to exercise it. For this, see Hönnige supra note 19, at 132; Sadurski supra note 3, at 23. Binding constitutional interpretations are not only advisory notes in the justifications of a decision, but they are included into the operative part of judicial decisions and they shouldn't be confounded with preliminary reference procedure of the ECJ. For an example, see Alkotmánybíróság (AB) [Constitutional Court] 21/1996 (Hung.), http://hunconcourt.hu/letoltesek/en_0021_1996.pdf.Google Scholar

³⁰ At first glance, it might not be obvious why we are arguing that constitutional interpretation in abstracto corresponds to constitution-making or writing. The idea, and problem, that constitutional courts might be, not only positive legislators, but actors of constitution writing processes, has been developed in the legal scholarship in connection to the heavily discussed problem of unconstitutional constitutional amendments. Constraining the constituent power by judicial decisions means certainly that the court vindicates the right to be an integrative part of a composite body assumed as pouvoir constituant. Also, the concept of post-sovereign constitution-making, as presented by Andrew Arato, hints to the constitutional courts emerged as powerful actors to fill in the gaps and deficiencies of a transitory constitution. In several countries—and especially in Hungary and in South Africa —prominent actors of these post-sovereign constitution making processes were constitutional courts. Both of these directions of constitutional theories consider, however, either all forms of the activity of constitutional courts in general, or they narrow down the problem to the concept of unconstitutional constitutional amendment. In contrast to these trends, we argue that including lengthy and detailed interpretation of constitutional norms and concepts into the operative part of a judicial decision is clearly an expansion of the text of the constitution. Because the operative part of a decision is undoubtedly legally binding, while the status of justifications is contested in this regard, we argue that binding interpretation of the constitution included into the operative part is equivalent to constitution-making. To this question, see Arato, Andrew, Post Sovereign Constitution Making: Learning and Legitimacy, (2016); Alec Stone Sweet, Constitutional Courts, in The Oxford handbook of Comparative Constitutional Law, 826 (Michael Rosenfeld & András Sajó eds., 2012); Hans Vorländer: Deutungsmacht-Die Macht der Verfassungsgerichtsbarkeit, in Die Deutungsmacht der Verfassungsgerichtsbarkeit, 20 (2006).Google Scholar

³¹ As for the term qualitative/quantitative partial annulment, see de Visser, supra note 15, at 314.Google Scholar

³² The initial results of the research point out that in Hungary, as opposed for example to Germany, qualitative partial annulment is not a regular decision in the Constitutional Court's practice, which usually considers whether the law should be completely or only partially annulled irrespective of possible interpretations. The comparative aims of the project, however, required the introduction of the category of qualitative partial annulment.Google Scholar

³³ Binding prescription (IVd) and constitutional requirement in the operative part of the decision (Id) are essentially the same. The only difference is that while the constitutional requirement does not formally annul the law, a binding prescription is a decision that annuls the law.Google Scholar

³⁴ The comparative aims of the research required the introduction of the headnote category—or, in case of the German constitutional court, Leitsatz—although in some cases, such as in Hungary, decisions of the constitutional court do not include such summaries.Google Scholar

³⁵ See Table 2.Google Scholar

³⁶ The strongest form of procedural unconstitutionality consists of procedural unconstitutionality (Ic) + complete annulment (IIc) + ex tunc timing (IIIc) + binding prescription (IVd), which gives 5 points in sum (1+1+1+2). The weakest form of substantial unconstitutionality consists of substantive unconstitutionality (Ie) + qualitative partial annulment (IIa) + pro futuro timing (IIIa) + no prescription (IVa), which means 6 points in sum (6+0+0+0)Google Scholar

³⁷ The strongest form of substantive unconstitutionality consists of substantive unconstitutionality (Ie) + complete annulment (IIc) + ex tunc timing (II Ic) + binding prescription (IVd) which means 10 points in sum (6+1+1+2).Google Scholar

³⁸ In cases of utmost permissibility, the constitutional court might explicitly broaden the legislature's room for maneuvering. As our research, however, focuses on the extent to which the constitutional court constrains the legislature, we decided not to use negative values and a dedicated scale to evaluate decisions that broaden the legislature's room for maneuvering. Because constitutional interpretation in abstracto is a weapon only rarely used by constitutional courts, qualitative analyses might help to determine what broadening in these cases actually means.Google Scholar

³⁹ As noted earlier, we regard binding prescription (IVd) and constitutional requirement in the operative part of the decision to have the same weight.Google Scholar

⁴⁰ See Halmai, supra note 7; Szente, supra note 7.Google Scholar

⁴¹ See generally Hönnige, supra note 19.Google Scholar

⁴² it must be noted that it was impossible to identify political positions related to the laws passed by the pre-1990 communist parliament.Google Scholar

⁴³ Alkotmánybíróság (AB) [Constitutional Court] 2009.IV.21. 47/2009 (Hung.).Google Scholar

⁴⁴ We posited as a general principle that rulings—for example, units of observations—will be defined by the court. For instance, if the court embraces various legal regulations into one ruling and decides on this bunch of legal regulations by declaring them unconstitutional in one ruling, we do not disaggregate this ruling. We assume that the court had good reasons to deal those legal regulations in a bunch.Google Scholar

⁴⁵ Alkotmánybíróság (AB) [Constitutional Court] 2013.XI.22 AK 33/2013 (Hung.)Google Scholar

⁴⁶ Alkotmánybíróság (AB) [Constitutional Court] 2013.XI.22. AK 33/2013 (Hung.) had one ruling, which referred to nine different legal regulations. The HCC rejected the petition aimed at establishing the unconstitutionality and the annulment of Section 2 paragraph (1) and (2), Section 6 paragraph (3) item c) and d), Section 267 paragraph (1) item j), Section 332 paragraph (1) item d), Section 373 paragraph (1) item c), Section 399 paragraph (1), and Section 416 paragraph (1) item c) of 1998 Act XIX on Criminal Procedure (Hung).Google Scholar

⁴⁷ Alkotmánybíróság (AB) [Constitutional Court] 2013.XI.22. AK 33/2013.Google Scholar

⁴⁸ It should be noted that in some cases a law can be linked to more legislative acts when the law in question consists of parts that require simple majority and other parts that require a supermajority. Appropriately, these cases need to be disaggregated into separated units of observation.Google Scholar

⁴⁹ It should be noted that four members of the Sólyom-court were elected by the outgoing communist parliament. There are three minor practical difficulties we faced which should be mentioned before proceeding to the evaluation of the data. In 2004 one judge, nominated by the left-wing parties, did not participate in any decisions which concerned right wing legislation. Both in 2004 and 2013, there was one judge who had been nominated originally in a complete consensus of all parties, which makes it impossible to bind them to any political groups. In 1995 there were four judges who had been elected right before the first free election in 1990, which makes their party affiliation more ambiguous than any others. Beyond these difficulties, the data for all three years—1995, 2004, and 2013—are complete and the evaluation can be executed.Google Scholar

⁵⁰ During the test phase of our research, we did not separate rulings of the Constitutional Court that are politically relevant from other, politically less significant cases. Certainly, highlighting decisions that are politically significant can be a valid research strategy that might serve as a suitable starting point for further research.Google Scholar

⁵¹ 1995: 42 relevant cases—10 right wing legislation and 32 left wing legislation; 2004: 45 relevant cases —11 right wing legislation and 34 left wing legislation; 2013: 80 relevant cases—68 right wing legislation and 12 left wing legislation.Google Scholar

⁵² It should be noted that at the start of a parliamentary term, the Constitutional Court decides on legislations adopted by previous parliaments and by earlier majorities. The number of decisions of the CCs concerning the bills adopted by a previous parliamentary term decreases by advancing within the parliamentary term. In turn, the number of cases concerning legislation of the actual parliamentary term increases approaching the end of the actual term.Google Scholar

⁵³ Dissenting opinions in the years analyzed: 1995: 2; 2004: 51; 2013: 100.Google Scholar

⁵⁴ The strength of the decisions of the HCC in the respective years, weighed average of left- and right-wing decisions taken together: 1995: 3.62; 2004: 3.28; 2013: 2.66.Google Scholar

⁵⁵ See Figure 1.Google Scholar

⁵⁶ See Brouard, Sylvain, The politics of constitutional veto in France: constitutional council, legislative majority and electoral competition, WEST EUR. POL. 383–403 (2009); see generally HÖNNIGE, supra note 19; Pedro Magalhes, The limits of judicialization: legislative politics and constitutional review in the Iberian Democracies (2003) (unpublished Ph.D. dissertation, Ohio State University) (Retrieved from: http://rave.ohiolink.edu/etdc/view?acc_num=osu1046117531).Google Scholar