A behavioural approach to judicial independence

Burchardt, Dana Article — Published Version A behavioural approach to judicial independence Maastricht Journal of European and Comparative Law Provided in Cooperation with: WZB Berlin Social Science Center Suggested Citation: Burchardt, Dana (2025) : A behavioural approach to judicial independence, Maastricht Journal of European and Comparative Law, ISSN 2399-5548, Sage, Thousand Oaks, CA, Vol. 32, Iss. 5-6, pp. 552-579, https://doi.org/10.1177/1023263X251363881 This Version is available at: https://hdl.handle.net/10419/333782 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc/4.0/ A behavioural approach to judicial independence Dana Burchardt* , ** Abstract This article develops a behavioural approach to the principle of judicial independence. It argues that implementing the proposed approach can help to safeguard judicial independence more effectively. It addresses the question of what a behaviour-related concept of judicial independence should look like and what state obligations arise from this. It takes into account empirical findings which were not yet available when the legal principle was initially shaped by constitutional and regional human rights jurisprudence. First, the article outlines certain basic premises of a behavioural concept of judicial independence. It then takes a closer look at the factors that determine the behaviour of judges and of such third actors that might attempt to unduly influence judges. In addition, the article engages with the objective of the principle of judicial independence to inspire trust in the judiciary and outlines potential behavioural effects relevant for this trust dimension. Finally, it discusses whether certain behavioural effects that can shape judicial decision-making should be regarded as normatively undesirable in the context of judicial independence. Keywords Judicial independence, behavioural concept, psychological factors, observer perspective, social perspectives 1. Introduction 1 Judicial independence is a behavioural concept. This is a fact that is not always adequately reflected in the way the legal principle of judicial independence is addressed on the domestic and regional * University of Bremen, Bremen, Germany ** WZB Berlin Social Science Center, Berlin, Germany Corresponding author: Dana Burchardt, University of Bremen, Bremen, Germany; WZB Berlin Social Science Center, Berlin, Germany. Email: [email protected] 1. This article is based, in part, on the author’s following article in German: D. Burchardt, ‘Der Grundsatz richterlicher Unabhängigkeit: eine verhaltensbezogene Perspektive’,70Jahrbuch des öffentlichen Rechts der Gegenwart. Neue Folge (JöR) (2022) © Mohr Siebeck Tübingen. Article Maastricht Journal of European and Comparative Law 2025, Vol. 32(5-6) 552–579 © The Author(s) 2025 Article reuse guidelines: sagepub.com/journals-permissions DOI: 10.1177/1023263X251363881 maastrichtjournal.sagepub.com level. It is intuitive that judicial independence affects the behaviour of both judges and third actors who seek to influence the former in normatively undesirable ways. However, the legal principle is often not sufficiently shaped in light of this behavioural dimension. In fact, the principle of judicial independence finds itself in a conceptual tension between a legal formalist and a legal realist understanding of judicial decision-making. This has resulted in the principle of judicial independence being somewhat disconnected from the reality of judicial behaviour. However, this disconnect makes it difficult to implement this principle in a practically effective manner. For the principle of judicial independence to be shaped in a convincing manner and, on this basis, to be implemented effectively, the behavioural nature of judicial independence must be examined more closely. One must consider whom this principle aims to protect; what determines the behaviour of judges and third actors; how these behavioural factors can be influenced to promote independence; and what should constitute normatively desirable or undesirable behaviour in the context of judicial independence. A closer look at these questions reveals the complexity of judicial independence, which is a significant challenge to conceptualizing this principle. This becomes particularly clear when one takes into account recent empirical findings, which were not yet available when the legal principle was initially shaped especially by constitutional and regional human rights jurisprudence. Such developments concern insights into the factors that contribute to normatively undesirable behaviour as well as into human cognition in general. It also relates to the perception of judicial decision-making by the addressees of judicial decision-making. An understanding of the principle of judicial independence that is based on an oversimplified idea of judicial decision-making does not do justice to this complexity. Effective implementation requires anchoring this principle in the actual behaviour of the actors involved. Against this background, this article develops a behavioural approach to the principle of judicial independence. It addresses the question of what a behaviour-related concept of judicial independence should look like and what state obligations arise from this. First, the article outlines certain basic premises of a behavioural concept of judicial independence (section 2) It then takes a closer look at the factors that determine the behaviour of judges and of such third actors that might attempt to unduly influence judges, highlighting the behavioural effect of factors that have traditionally been addressed as relevant for judicial independence as well as other behavioural aspects such as psychological framework conditions that go beyond these factors (section 3). In addition, the article engages with the objective of the principle of judicial independence to inspire trust in the judiciary. This trust dimension of judicial independence addresses actors that are external to the judiciary such as those seeking justice and the general public. The article outlines potential behavioural effects relevant for this trust dimension (section 4). Finally, the article discusses whether certain behavioural effects that can shape judicial decisionmaking should be regarded as normatively undesirable in the context of judicial independence (section 5). 2. Judicial independence as a behavioural concept: Preliminary remarks Before delving into the behavioural nature of judicial independence, a preliminary remark is in order. The term judicial independence is understood in this article in a broad manner. It includes, on the one hand, the relationship between judges and actors in the executive, legislative and judicial branch of government: judges are expected to be independent from these public actors. On the other hand, it also includes the relationship between judges and parties to the dispute: judges are expected Burchardt 553 to be independent from the parties in the case at hand to be decided by the judge. This impartiality dimension is integrated into the broad concept of judicial independence referred to in this article, rather than being separated from the notion of judicial independence. 2 This broad understanding makes sense for the present context because the behavioural factor for judges and third actors are similar in all of these settings. On this terminological basis, this section now turns to judicial independence as a behavioural concept. The section first outlines certain terminological shortcomings that contribute to blurring the behavioural dimension of judicial independence (a). This is followed by a model of judicial decision-making that helps situating the various behavioural factors relevant for judicial independence (b). A. Behavioural dimension and terminological shortcomings The behavioural dimension of judicial independence is often not sufficiently reflected in the doctrinal and jurisprudential engagement with this topic. This is particularly apparent in the notion of ‘guarantee’, which is very widely used in constitutional and human rights law and jurisprudence in relation to judicial independence. 3 The following two aspects will highlight the shortcomings of this terminology in relation to judicial independence as a behavioural notion. First, the term ‘guarantee’of judicial independence contributes to obscuring the behavioural dimension of judicial independence as it does not appropriately reflect the role of the state with regard to judicial independence. In fact, the state cannot ‘guarantee’that judges will behave in line with judicial independence. In other words, the state cannot guarantee de facto independence. The state can only influence the factors that determine the behaviour of judges and third actors. These factors can only influence the probability of the respective behaviour. 4 However, the probability can be maximized by designing the factors that determine behaviour. Accordingly, the legal obligation of the state that is at the core of the principle of judicial independence should be primarily understood as a structural obligation to maximize this positive behavioural setting. The state can design the structural framework in such a way that it makes it more likely that the judge will behave according to the expectations of judicial independence. When we talk about ‘guarantees’in the context of judicial independence, this terminology does not make sense for de facto independence as such. It can only be conceivable with regard to certain factors that influence judicial behaviour. The existence of some of these factors can be legally and 2. This broad understanding of judicial independent is also used by some courts in their jurisprudence on judicial independence. See e.g. the approach taken by the CJEU: Case C-192/18 Commission v. Poland, EU:C:2019:924, para. 108–110; Case C-274/14 Banco de Santander, EU:C:2019:802, para. 61. 3. See for a small selection on the use of the term ‘guarantee’e.g.: ECtHR, Mutu and Pechstein v. Switzerland, Judgment of 2 October 2018, Applications nos. 40575/10 and 67474/10; ECtHR, Guðmundur Andri Ástráðsson v. Iceland, Judgment of 1 December 2020 [GC], Application No. 26374/18); CJEU, Joined Cases C-585/18, C-624/18 and C-625/18 A. K. and Others v Sąd Najwyższy, CP v Sąd Najwyższy and DO v Sąd Najwyższy, EU:C:2019:982; Case C-791/19 European Commission v. Republic of Poland, EU:C:2021:596; German Federal Constitutional Court, 22 March 2018, 2 BvR 780/16; French Conseil Constitutionnel, 1 March 2007, 2007-551 DC; Canadian Supreme Court, [1985] 2 SCR 673; Swiss Federal Court, 10 June 2010, 4A_458/2009. See also UN Basic Principles on the Independence of the Judiciary (6 September 1985); UN Special Rapporteur on the Independence of Judges and Lawyers, Mr Leandro Despouy, UN Human Rights Council, document A/HRC/11/41 of 24 March 2009; Venice Commission, Report on the Independence of the Judicial System Part I: The Independence of Judges, CDL-AD(2010)004 (16 March 2010). 4. See also C. Schütz, Der Ökonomisierte Richter (Duncker & Humblot, 2005), p. 220. 554 Maastricht Journal of European and Comparative Law 32(5-6) factually guaranteed. For example, the organizational and personal separation of the judiciary and executive, and certain legal requirements regarding elections, appointments and promotions, length of office, irremovability and remuneration of judges can be guaranteed by the state. These legal requirements contribute to de jure independence. However, other structural factors cannot be guaranteed. For example, it cannot be guaranteed that the legal requirements regarding judicial status (election, appointment, etc.) are implemented in practice in all instances. Further, it is not possible to guarantee that there is no influence on judges by third actors, including the indirect and often psychological influence that regional courts, international instruments as well as domestic constitutional courts consider to be part of the judicial independence concept. 5 In sum, this means that neither de facto independence nor all the factors that determine behaviour can be guaranteed. To speak of a ‘guarantee’of judicial independence is thus misleading in several respects. Second, using the ‘guarantee’terminology comes with the idea that the principle of judicial independence aims (only) to protect the judge from normatively undesirable influence by third actors. 6 Yet this understanding is not in line with the behavioural dimension of judicial independence. If one thinks of judges as subjects that are protected by the principle of judicial independence, this leads to focusing on the behaviour of third actors that potentially could unduly influence judges. The idea is that these actors’behaviour is a danger from which judges must be protected. 7 The primary objective is then to prevent this normatively undesirable behaviour by third actors. 8 However, this approach leads to not addressing sufficiently the behaviour of judges as a point of reference. Instead of primarily focusing on the behaviour of third actors, the concept of judicial independence must aim at ensuring that judges behave in a normatively desirable manner (see section 3). The judges’behaviour should be the primary point of reference. The behaviour of third actors of course also plays a role in this respect, but it only matters alongside the behaviour of the judge. Both groups of subjects and their respective behaviour must be considered separately with regard to the factors that determine this behaviour. It should also be kept in mind that the principle of judicial independence ultimately aims to protect societal interests and related constitutional values. The constitutional values that are implemented through judicial independence and serve the interests of the addressees of the respective legal system include, in particular, separation of powers, the binding effect of law and procedural fairness. The broader societal interests that have been shown to correlate with judicial independence include long-term economic growth, 9 high protection of political and economic freedoms, 10 5. E.g. Case C-619/18 Commission v. Poland, EU:C:2019:531, para. 112; IACtHR, Case of the Supreme Court of Justice, Quintana Coello et al. v. Ecuador, decision of 23 August 2013, para. 146; UN Basic Principles on the Independence of the Judiciary, 6 September 1985. For an example of domestic jurisprudence, see e.g. German FCC, decision of 22 March 2018, 2 BvR 780/16, para. 57; FCC, decision of 22 June 2006, 2 BvR 957/05. 6. On this ‘protection’terminology in German constitutional law, e.g. C. Schütz, Der Ökonomisierte Richter, p. 117 et seq. 7. See e.g. W. Meyer, in I. von Münch, P. Kunig, J.A. Kämmerer and M. Kotzur (eds), Grundgesetz Kommentar, vol. 2 (CH Beck, 7th edn., 2021), Art. 97 GG, para. 101. 8. Speaking of a ‘prohibition’addressed to third actors, C.D. Classen, in H. von Mangoldt, F. Klein and C. Starck (eds.), Grundgesetz Kommentar, vol 3I (CH Beck, 7th ed., 2018), Art. 97, para. 7. 9. L.P. Feld and S. Voigt, ‘Economic Growth and Judicial Independence: Cross-Country Evidence Using a New Set of Indicators’,19European Journal of Political Economy (2003). 10. R. La Porta, F. López de Silanes, C. Pop-Eleches and A. Shleifer, ‘Judicial Checks and Balances’, 112 Journal of Political Economy (2004). Burchardt 555 resilience of democratic systems against autocratic tendencies, 11 the acceptance of the results of democratic elections 12 as well as a related system-stabilizing effect. 13 This makes the addressees of a legal system the ultimate subjects of protection of the principle of judicial independence. As a result of this objective, the principle of judicial independence aims at de facto independence and not merely de jure independence. 14 Only actually independent behaviour by judges serves the constitutional values mentioned. It is thus necessary to induce normatively desirable behaviour by judges in practice rather than merely creating formal legal requirements. In order to do so effectively, it is crucial to understand the behavioural factors that contribute to judges acting in an independent manner (see section 3). Secondly, it is important that the addressees of a legal system perceive the judges’behaviour as independent in practice. These addressees should have reason to trust that judges are very likely to behave in line with judicial independence. This latter objective, its implications for the concept of judicial independence as well as potential behavioural effects related to it are outlined in more detail in section 4. B. Judicial decision-making process and behavioural factors As a reference point for the analysis in sections 3 to 5 and as a framework for better understanding and conceptualizing the factors that shape independent judicial behaviour, a closer look at the process of judicial decision-making is in order. This makes it possible to classify the factors that determine judicial behaviour and thus create a basis for assessing these factors normatively as well as for assessing how effectively certain structural conditions promote independent decisionmaking. The following overview thus aims to situate the various behavioural influences at the levels of judicial decision-making. It shows that behavioural factors that promote or hinder judicial independence can be relevant on two distinct levels of judicial decision-making. At the first level, the intentional level, judges intentionally decide whether they want to act according to the normative standards set forth by the principle of judicial independence and their legal obligation that follows from this principle. For example, they decide whether to give in to a third actor’s attempt to exert undue influence on them. Judges thus intentionally choose between behavioural alternatives that are in line with judicial independence or not. If they resist an attempt by a third actor to exercise undue influence on them, they decide to act in line with judicial independence; if, in contrast, they decide to give in, they decide to act in a normatively undesirable manner from the perspective of judicial independence. Which alternative judges choose at this 11. D.M. Gibler and K.A. Randazzo, ‘Testing the Effects of Independent Judiciaries on the Likelihood of Democratic Backsliding’,55American Journal of Political Science (2011). See also C. Reenock, J.K. Staton and M. Radean, ‘Legal Institutions and the Democratic Survival’,75Journal of Politics (2013). 12. S. Chernykh, ‘When Do Political Parties Protest Election Results?’,47Comparative Political Studies (2014). 13. See also C.M. Cameron, ‘Judicial Independence: How Can You Tell It When You See It? And, Who Cares?’, in S.B. Burbank and B. Friedman (eds), Judicial Independence at the Crossroads: An Interdisciplinary Approach (Sage, 2002), p. 134. 14. In the scholarly literature, the notions of de jure independence and de facto independence are often used (unlike in the present context) as indicators for empirical measurements, see e.g. L.P. Feld and S. Voigt, 19 European Journal of Political Economy (2003); G. Dimitropoulos, ‘Measuring Judicial Independence in International Law: Putting Together the Pieces of the Puzzle’,24Maastricht Journal of European and Comparative Law (2017); J. Rios-Figueroa and J.K. Staton, ‘An Evaluation of Cross-National Measures of Judicial Independence’,30Journal of Law, Economics & Organization (2013). For a normative rather than empirical use of these terms, see R. Spano, ‘The Rule of Law as the Lodestar of the European Convention on Human Rights: The Strasbourg Court and the Independence of the Judiciary’,27European Law Journal (2021), p. 217–219. 556 Maastricht Journal of European and Comparative Law 32(5-6) first level depends on a variety of behavioural factors. These factors can concern the cost-benefit assessment of behaviour that is in line with judicial independence or not; and it can also include other factors such as ethos and fairness perceptions or group practice. The details of these factors are addressed in section 3. For the present overview, it suffices to point to the fact that the totality of these factors create a behavioural setting, which, depending on how it is designed, makes it more or less likely that judges intend to behave in a manner that is in line with judicial independence. The classical (theoretical and constitutional) concept of judicial independence focuses on this first level and the factors relevant to it. The aim is to design structural conditions that minimize the risk of behaviour that is not in line with judicial independence. This first level is, however, not the only phase of the judicial behaviour that should be taken into account for judicial independence. On a second level, the implementation level, judges translate their intention from level 1, that is, the intention to act in line with the principle of judicial independence, into concrete action. This is done by taking the actual judicial decision in the legal sense. These can be judgments, procedural decisions, decisions regarding interim measures, etc. At level 2, the judicial decision can be determined by various factors. First and foremost, of course, the legal norms applied by the judge shape the concrete content of the judicial decision. However, in addition to law as determining factor, other behavioural factors can play a role as well. These can be factors that judges do not consider normatively undesirable, that judges are not aware of and/or that are, at least in the judges’ perception, not under their control. Such factors include, inter alia, cognitive effects such as ‘heuristics and biases’and behavioural factors related to the social perspective of the judges. An example of heuristics and biases in judicial decision-making 15 is the ‘anchoring effect’which can influence numeric decisions by judges on damages, fines, length of prison sentences etc. 16 Examples of the social perspective include cognitive effects related to gender or ethnic background of judges. 17 The fact that such factors influence the content of judicial decision-making has been widely demonstrated by empirical research on judicial behaviour. So far, such behavioural factors are however rarely discussed as an aspect of judicial independence. Whether such factors should be included in the concept of judicial independence is discussedinsection5forfactorsrelated to social perspectivity. The interaction of the two decision-making levels is shown in Figure 1. The behaviour of third actors is relevant at both decision-making levels. On the one hand, third actors can influence the decision for or against an independent behaviour at level 1. For example, third actors can influence the judge’s interests by promising advantages or threatening judges with disadvantages. On the other hand, third actors can also influence behavioural effects that take place on level 2 of the decision-making process. This would primarily concern behavioural factors that take place outside the judge’s control and/or perception. On this level, third actors can influence 15. For an overview in the context of judicial decision-making: D. Teichman and E. Zamir, Behavioural Law and Economics (OUP, 2018), chapter 15; C. Guthrie, J.J. Rachlinski and A.J. Wistrich, ‘Inside the Judicial Mind’,86Cornell Law Review (2000); E. Peer and E. Gamliel, ‘Heuristics and Biases in Judicial Decisions’,49Court Review (2013). 16. On the anchoring effect in judicial decision-making, see e.g. J.J. Rachlinski, A.J. Wistrich and C. Guthrie, ‘Can Judges Make Reliable Numeric Judgments? Distorted Damages and Skewed Sentences’,90Indiana Law Journal (2015); B. Englich, T. Mussweiler and F. Strack, ‘Playing Dice With Criminal Sentences: The Influence of Irrelevant Anchors on Experts’Judicial Decision Making’,32Personality and Social Psychology Bulletin (2006). On the question of whether the anchoring effect should be considered problematic in relation to judicial independence, see D. Burchardt, ‘Der Grundsatz richterlicher Unabhängigkeit: eine verhaltensbezogene Perspektive’,70Jahrbuch des öffentlichen Rechts der Gegenwart. Neue Folge (JöR) (2022), p. 502–506. 17. For an overview, see S. Basabe-Serrano, L. Epstein and K. Weinshall, ‘Backgrounds, Attributes, and Identities’,in L. Epstein et al (eds), The Oxford Handbook of Comparative Judicial Behaviour (OUP, 2024). Burchardt 557 judges by taking advantage of, or even triggering, certain cognitive effects such as the anchoring effect mentioned above. Whether third actors try to influence judges also depends on behavioural factors. These may be partly congruent with the factors that determine judicial behaviour but they can also include specific factors. Section 3 outlines these factors. The link between the factors that determine the behaviour of third actors and the factors that determine judicial behaviour can lead to a chain-like interaction. This interplay of behavioural factors is shown in Figure 2. 3. Behaviour of judges and third actors: Behavioural factors for judicial independence After the preliminary remarks on the behavioural dimension of judicial independence, this section turns to the factors and related measures that promote behaviour that is in line with judicial Figure 1. Levels of decision-making. Figure 2. Interplay between the behavioural factors for judges and third actors. 558 Maastricht Journal of European and Comparative Law 32(5-6) independence. It shows how the state can generate structural conditions that are conducive to behaviour of judges and third actors that is in line with judicial independence. The factors addressed here are only those that can be influenced by the state and that are therefore relevant to the state obligation ensuing from the principle of judicial independence. This overview engages with the first level of judicial decision-making as outlined in section 2.B. It highlights three points. First, the elements of the behavioural setting influence the behaviour of judges and third actors in different ways. An element can have several, sometimes even contradictory, effects. To maximize a behavioural setting that is favourable to judicial independence, it is important to understand the behavioural functioning of the respective elements. Second, in addition to the aspects that are regularly mentioned in the context of judicial independence, other factors –in particular psychological framework conditions –are also relevant for whether judges and third actors behave in line with judicial independence. Some of these factors are discussed below. Third, it is crucial to take into account the interaction of the various behavioural factors. Instead of looking at individual factors in isolation, the overall behavioural setting must be kept in mind when it comes to assessing and promoting the likelihood of normatively desirable behaviour. A. Behaviour of judges The factors that shape judicial behaviour and that can contribute to promoting independence are manifold. The following overview starts with a first set of factors that follows from a rationalist account of human behaviour; a second set of behavioural factors complements this approach. These sets of factors interact and, together, form the overall behavioural framework by which judicial behaviour is determined. To start with, a rationalist account of human behaviour focuses on the relationship between the expected benefits and possible costs of normatively undesirable behaviour. Judges are addressed as rational actors who behave in a way that maximizes utility. From this perspective, judges are expected to behave independently if the expected benefits of this behaviour are greater than the possible costs associated with it, taking into account the probability with which these costs occur. 18 Using this utility-based lens, the state can influence the behavioural setting for judicial independence by targeting the benefits and costs of behaviour that is not in line with judicial independence as well as the probability with which, according to the expectation of the relevant actor, benefits and costs will materialize. To promote judicial independence, the expected benefits of normatively undesirable behaviour can be minimized. The design of judicial career opportunities is particularly relevant in this regard. The following overview shows how career benefits act as behavioural factors relevant for judicial independence. This overview aims to highlight how career-related measures that states use to foster judicial independence operate from a behavioural perspective. Afirst crucial interest is the judges’interest in keeping their professional position. This interest is affected by the legal provision on the irremovability of judges from their office. When judges can be (easily) removed from office, there is a risk that the judges’current professional existence determines the cost-benefit analysis with regard to normatively desirable or undesirable behaviour. In a setting in which actors can remove judges from office as a (de facto) sanction for decisions 18. On these elements of the cost-benefit analysis, A. Mishra, ‘Corruption, Hierarchies and Bureaucratic Structure’,in S. Rose-Ackerman (ed.), International Handbook on the Economics of Corruption (Edward Elgar, 2006), p. 192. Burchardt 559 The state obligation resulting from the principle of judicial independence is not only aimed at promoting behaviour on the part of judges and third actors that is in line with judicial independence. It also aims to establish and maintain the trust in the independence of the judiciary (section 4.A). This objective brings an additional behavioural dimension to the principle of judicial independence. It relates to the group of actors who perceive and evaluate the behaviour of judges and third actors who might attempt to influence these judges. This perception and evaluation are influenced by behavioural factors. These factors and their implications for the principle of judicial independence are outlined in section 4.B. A. Objective: Trust of those subject to the law Under the principle of judicial independence, the state must aim at creating structural conditions in which those subject to the law can expect the judiciary to decide the cases brought before it in an independent manner. Observing actors within a legal order should be able to trust that judges will behave in line with judicial independence. This includes the expectation that third actors do not seek to unduly influence judges. Observing actors are both those involved in court proceedings and the general public. Promoting the trust of these actors is a requirement of the principle of judicial independence. The ECtHR has repeatedly emphasized the importance of trust in the courts. According to the ECtHR, judicial independence is about ‘confidence which the courts must inspire in a democratic society’. 43 This formulation, which was more recently also adopted by the CJEU, 44 refers to the function of an independent judiciary for the principle of democracy. The normative requirement for trust as set out by the above courts is grounded on empirical findings. In fact, the social relevance of the trust that actors have in the independence of judges is empirically underpinned. This includes findings regarding the influence of procedural fairness on how persons perceive judicial decisions and how this shapes their subsequent behaviour. A large number of studies have investigated the question of how procedural justice affects how both parties to proceedings and members of a wider audience assess decisions; and how this assessment can influence the future behaviour of these actors towards the decision-makers. 45 Procedural fairness has been shown to be a crucial factor that can lead to a favourable evaluation of the decision maker even if the content of a decision is negatively 43. ECtHR, Sramek v. Austria, Judgment of 22 October 1984, Application No. 8790/79, para. 42. The reference to the importance of trust for a democratic society is a regular component of the ECtHR’s jurisprudence on judicial independence, see e.g. ECtHR, Ramos Nunes de Carvalho e Sá v. Portugal, Judgment of 6 November 2018, Application Nos. 55391/13, 57728/13 and 74041/13, para. 149; ECtHR, Sacilor Lormines v. France, Judgment of 9 November 2006, Application No. 65411/01, para. 60; ECtHR, Kleyn and Others v. the Netherlands, Judgment of 6 May 2003, Application Nos. 9343/98, 39651/98, 43147/98 and 46664/99, para. 191. 44. Case C-585/18, C-624/18 and C-625/18 A.K. (Independence of the Disciplinary Chamber of the Supreme Court), para. 153. 45. See e.g. T.R. Tyler and Y.J. Huo, Trust in the Law: Encouraging Public Cooperation with the Police and Courts (Russell Sage Foundation, 2002); T.R. Tyler, J.D. Casper and B. Fisher, ‘Maintaining Allegiance toward Political Authorities: The Role of Prior Attitudes and the Use of Fair Procedures’,33American Journal of Political Science (1989); J. Jackson et al, ‘Why Do People Comply With The Law? Legitimacy and the Influence of Legal Institutions’,52British Journal of Criminology (2012); M.D. Ramirez, ‘Procedural Perceptions and Support for the U.S. Supreme Court’,29Political Psychology (2008); S.C. Benesh, ‘Understanding Public Confidence in American Courts’,68The Journal of Politics (2006). 566 Maastricht Journal of European and Comparative Law 32(5-6) evaluated. 46 Both for those involved in the proceedings who were unsuccessful in a legal dispute and for external actors whose preferences ran counter to a court decision, it is significant or even decisive whether the procedure that led to this decision was perceived as fair. And with regard to the future behaviour of these actors, a decision that is perceived as fair seems to increase the willingness to cooperate with judicial authorities or other institutional actors. 47 In light of these empirical findings, the objective of guaranteeing trust in courts must therefore be considered key to the principle of judicial independence. 48 Based on this understanding, the state has an obligation to maximize this trust. This is an obligation that some courts have explicitly recognized as being part of the principle of judicial independence. 49 This obligation should generally be taken into account when applying the principle of judicial independence in the legal context. The following section explains this trust dimension of judicial independence and how potential behavioural effects play a role in this regard. B. Trust and the observer perspective Although there is no standardized definition of trust, 50 there are several basic features of trust that are relevant to trust in the context of judicial independence. First, trust is to be understood as a relationship between a trustor and a trustee. 51 In this relationship, trust can be unilateral or mutual. And trust relates to a specific behaviour of the trustee. Trust does not exist abstractly between two actors, but actor A trusts in a certain behaviour of actor B. The point of reference here is an expected (desired) future behaviour of the latter. 52 Trust is therefore defined by three elements: actor A (the trustor), actor B (the trustee) and the latter’s behaviour x. 53 In the judicial independence context: the subjects of the law (A) should trust that judges and indirectly also third actors (B) behave in line with judicial independence (x). 46. T.R. Tyler and Y.J. Huo, Trust in the Law: Encouraging Public Cooperation with the Police and Courts; J. Brockner and B.M. Wiesenfeld, ‘An Integrative Framework for Explaining Reactions to Decisions: Interactive Effects of Outcomes and Procedures’,20Psychological Bulletin (1996); T. Stahl, J.-W. Van Prooijen and R. Vermunt, ‘On the Psychology of Procedural Justice: Reactions to Procedures of Ingroup vs. Outgroup Authorities’,34European Journal of Social Psychology (2004); P.C. Magalhães and L. Aguiar-Conraria, ‘Procedural Fairness, the Economy, and Support for Political Authorities’,40Political Psychology (2019). 47. See in particular T.R. Tyler and Y.J. Huo, Trust in the Law: Encouraging Public Cooperation with the Police and Courts. 48. This function of securing trust and acceptance is also emphasized in legal scholarship, see e.g. F. van Dijk, Perceptions of the Independence of Judges in Europe (Palgrave, 2021); J. Ferlemann, ‘Vorübergehend unabhängig? –Richter auf Zeit zwischen Unabhängigkeit und Flexibilität’, in M. Mülder et al. (eds), Richterliche Abhängigkeit –Rechtsfindung im Öffentlichen Recht (Nomos, 2018), p. 49, 65. 49. See on this obligation under domestic constitutional law, e.g. German FCC, decision of 14 January 2020, 2 BvR 1333/ 17, para. 91. 50. For an overview of the discussion on the concept of trust, see e.g. L.M. PytlikZillig and C.D. Kimbrough, ‘Consensus on Conceptualizations and Definitions of Trust: Are We There Yet?’, in E. Shockley et al. (eds), Interdisciplinary Perspectives on Trust (Springer, 2016). 51. Ibid. 52. On this dimension of expectation, G. Möllering, Trust: Reason, Routine, Reflexivity (Emerald Group Publishing, 2006), p. 7. 53. R. Hardin, ‘Conceptions and Explanations of Trust’, in K.S. Cook (ed.), Trust in Society (Russell Sage Foundation, 2001). Burchardt 567 Second, trust is to be understood as a reaction to uncertainty. 54 The behaviour of actor B cannot be determinatively predicted. There is a risk of a negative outcome, i.e., that actor B does not behave as actor A would like them to behave. If there were no uncertainty in this respect, trust would be irrelevant because the probability of negative behaviour on the part of the trustee is zero. 55 In the context of judicial independence, this uncertainty consists in the fact that it cannot be ruled out that the decision-making of judges is determined by normatively undesirable factors and that there is normatively undesirable influence by third actors. Observers can thus merely make a prognosis regarding the behaviour of judges and third actors. Third, trust is subjective in nature. It is not an objective risk assessment. 56 To start with, the knowledge base is subjective in a trust setting. Trustors will most likely only have knowledge of some of the available information, so that they do not know the objective probability of the trustee’s behaviour. 57 In addition, the available information may be (partially) generated by the trustors themselves, for example because it is based on personal experience. Moreover,the risk assessment as such is subjective. Factors that are specific to the trustor can influence this assessment. Such factors can be of a cognitive nature, including cognitive distortions –as will be outlined below. 58 In addition, actors can link the specific risk assessment to a larger personalized trust context, for example if a binary trust relationship is embedded in a network of overlapping social relationships 59 or is part of an institutional structure. 60 In addition to cognitive and affective factors, 61 a general inclination to trust as a personality trait can also play a role in the development of trust. 62 This subjective nature of trust is also characteristic of trust in the context of judicial independence. The perception of the behaviour of judges and of third actors is subjective, as is the risk assessment regarding independent behaviour. In fact, the subjective nature of trust is of particular importance for the legal concept of judicial independence. If this is not taken seriously, the objective to ‘inspire trust’in courts cannot be realized. In the following, the remainder of this section thus shows how the legal principle of judicial independence should integrate the subjective nature of trust, taking into account the potential behavioural effects related to it. 54. L.M. PytlikZillig and C.D. Kimbrough, in E. Shockley et al. (eds), Interdisciplinary Perspectives on Trust; D.M. Rousseau, S.B. Sitkin, R.S. Burt and C. Camerer, ‘Not So Different after All: A Cross-discipline View of Trust’,23 Academy of Management Review (1998), p. 395. 55. J. Jackson and J. Gau, ‘Carving Up Concepts? Differentiating Between Trust and Legitimacy in Public Attitudes Towards Legal Authority’, in E. Shockley et al. (eds), Interdisciplinary Perspectives on Trust (Springer, 2016), p. 53. 56. G. Möllering, Trust: Reason, Routine, Reflexivity,p.8. 57. J. Jackson and J. Gau, in E. Shockley et al. (eds), Interdisciplinary Perspectives on Trust, p. 53. 58. On cognitive heuristics as a factor of trust, M.D. Baer and J.A. Colquitt, ‘Why Do People Trust? Moving toward a More Comprehensive Consideration of the Antecedents of Trust’, in R.H. Searle, A.-M.I. Nienaber and S.B. Sitkin (eds), The Routledge Companion to Trust (Routledge, 2018), p. 174–177. 59. G. Möllering, Trust: Reason, Routine, Reflexivity,p.9. 60. R. Bachmann, ‘Institutions and Trust’, in R.H. Searle, A.-M.I. Nienaber and S.B. Sitkin (eds), The Routledge Companion to Trust (Routledge, 2018). See also on the concept of trust in a system, N. Luhmann, Vertrauen –Ein Mechanismus zur Reduktion sozialer Komplexität (F. Enke Verlag, 1968). 61. E.g. D. Lewis and A. Weigert, ‘Trust as a Social Reality’,63Social Forces (1985); D. Dunning, D. Fetchenhauer and T. Schlösse, ‘Why People Trust: Solved Puzzles and Open Mysteries’,28Current Directions in Psychological Science (2019); D. Dunning and D. Fetchenhauer, ‘Understanding the Psychology of Trust’, in D. Dunning (ed.), Social Motivation (Taylor & Francis, 2010). 62. R.C. Mayer, J.H. Davis and F.D. Schoorman, ‘An Integrative Model of organizational Trust’,20The Academy of Management Review (1995). 568 Maastricht Journal of European and Comparative Law 32(5-6) Not all courts that adjudicate about questions of judicial independence take the subjective nature of trust sufficiently serious. For example, the German constitutional court, although recognizing in principle that trust in the independence of courts is an important element, only engages in an objective assessment of risks for judicial independence. It does not take into account the subjective perspective of the external observers within the German legal order by asking, for example, how the general public would perceive certain situations such as public servants acting as temporary judges. 63 In comparison, the CJEU and the ECHR refer more intensely to trust and the subjective perspective of the trustees. The CJEU explicitly examines whether there are legitimate doubts about judicial independence on the part of the general public. According to the CJEU, the judiciary must be structured in a way ‘to dispel any reasonable doubt in the minds of individuals as to the imperviousness of that body to external factors and its neutrality with respect to the interests before it’. 64 The ECtHR also examines whether there is a legitimate fear or doubts by a party to a dispute. 65 This standard is used for assessing both judicial independence in the narrower sense and impartiality. 66 For the ECtHR, the ‘appearance of independence’is a criterion when assessing whether there has been a violation of Article 6 ECHR. 67 It bases this on the legal principle ‘justice must not only be done; it must also be seen to be done’, which has been part of the ECtHR’s jurisprudence for a long time. 68 In this jurisprudence, the ECtHR takes a two-step approach to assessing judicial independence –an approach that is also reflected in the present article. An objective standard of review (step one) is supplemented by a subjective dimension that reflects the perspective of the external observers (step two). For example, in Belilos v. Switzerland, the ECtHR affirmed that, with regard to certain elements of judicial independence such as the appointment of judges, their irremovability and freedom from instructions, there was no objective risk but it nevertheless rejected the independence of the court in question, referring to how an ‘ordinary citizen’perceives the judges. 69 This example also highlights that the subjective review standard is not about the subjective perspective of an individual but rather about the subjective perspective of a group of actors. An aspect that merits particular attention due to its potential behavioural implications is the difference between internal and external perspectives. The internal perspective refers to situations in which judges assess the independence of other judges. Judges are part of the judiciary and therefore assess the relevant behaviour of other judges from a particular perspective. This perspective not 63. FCC, decision of 22 March 2018, 2 BvR 780/16, para. 61. 64. Joined Cases C-585/18, C-624/18 and C-625/18, A.K. and Others v Sąd Najwyższy, CP v Sąd Najwyższy and DO v Sąd Najwyższy, para. 123; Case C-192/18 Commission v. Poland, para. 111; Case C-619/18 Commission v. Poland, para. 74. See also Case C-274/14 Banco de Santander, para. 63; Case C-272/19 VQ v Land Hessen, EU:C:2020:535, para. 52. 65. ECtHR, Sacilor Lormines v. France, Judgment of 9 November 2006, Application No. 65411/01, para. 63. 66. For the latter formally only in the context of objective and not subjective impartiality, but subjective impartiality only plays a minor role in the jurisprudence of the ECtHR. On the relationship between these two elements, see also L.F. Müller, Richterliche Unabhängigkeit und Unparteilichkeit nach Art. 6 EMRK, p. 116–117. 67. ECtHR, Campbell and Fell v. the United Kingdom, Judgment of 28 June 1984, Application Nos. 7819/77, 7878/77, para. 78. 68. Ibid., para. 81. See also similarly in ECtHR, Delcourt v. Belgium, Judgment of 17 January 1970, Application No. 2689/ 65, para. 31. 69. Namely as a member of the police and not as an independent member of the police court, see ECtHR, Belilos v. Switzerland, Judgment of 29 April 1988, Application No. 10328/83, para. 67. A further example of this approach is a case in which the ECtHR considered the mere presence of a person during court deliberations as a sufficient factor for potential influence: ECtHR, Martinie v. France, Judgment of 12 April 2006, Application No. 58675/00, para. 53–55. Burchardt 569 only comes with specific knowledge that is not directly available to the parties to proceedings or to the general public, but it also can generate certain cognitive effects, as outlined below. Inversely, the parties to proceedings or the general public have an external perspective on the judiciary –a perspective that comes with less knowledge about internal structures and practices as well as with cognitive effects related to the external standpoint. This difference in perspectives is important when judges –including the judges at constitutional and supreme courts as well as regional courts –must decide whether judicial independence is affected in a case at hand. For such a decision, judges must assess the risk of normatively undesirable behaviour by other judges. If one does not take into account the difference in perspectives, it is likely that these courts’assessment will be influenced by cognitive effects related to the internal perspective of the judges who make the assessment. To do justice to the trust dimension of judicial independence, the perspective of external observers should be considered. Differences between internal and external perspectives can manifest themselves in various ways. One important aspect is the group affiliation of the evaluating actor. Behavioural science research has shown that people assess situations differently depending on whether they belong to a certain group or not and whether a question concerns their own group or another group. For example, there is a tendency to evaluate a person’s behaviour more positively if this person belongs to one’s own group than if the person belongs to a different group. 70 If judges are asked to assess other judges in terms of judicial independence, this phenomenon can have an impact on the assessment. 71 The effect may be that judges evaluate the independence of judges as members of their own group more favourably than they would have done for non-group members under the same circumstances. Another relevant cognitive effect relates to the way in which people assess themselves. The perception that one has of one’s own decision-making processes can be influenced by cognitive biases. When people assess themselves, they tend to see themselves as less susceptible to unconscious behavioural factors than other people. 72 In addition, there is a general tendency to evaluate one’s own actions more favourably than the actions of others. This effect has been shown to be part of a so-called self-serving bias. This bias can affect the assessment of both practical skills, which tend to be rated as above average, 73 and normative questions such as ethical or fair behaviour. 74 The normatively desirable behaviours with regard to judicial independence concern both practical and normative aspects so that these two dimensions of the self-serving bias can potentially be engaged when assessing judicial independence. Moreover, when the standard used for assessment are rather vague, this seems to cause an even more positive self-assessment. 75 This effect may also 70. See e.g. M. Hewstone, M. Rubin and H. Willis, ‘Intergroup Bias’,53Annual Review of Psychology (2002); M.B. Brewer and J.G. Weber, ‘Self-Evaluation Effects of Interpersonal versus Intergroup Social Comparison’,66Journal of Personality and Social Psychology (1994); B. Mullen, R. Brown and C. Smith, ‘In Group Bias as a Function of Salience, Relevance, and Status: An Integration’,22European Journal of Social Psychology (1992), p. 103. 71. See also A. Higgins and I. Levy, ‘Judicial Policy, Public Perception, and the Science of Decision Making: A New Framework for the Law of Apprehended Bias’,38Civil Justice Quarterly (2019), p. 390. 72. See e.g. E. Pronin, D.Y. Lin and L. Ross, ‘The Bias Blind Spot: Perceptions of Bias in Self Versus Others’,28 Personality and Social Psychology Bulletin (2002); E. Pronin, T. Gilovich and L. Ross, ‘Objectivity in the Eye of the Beholder: Divergent Perceptions of Bias in Self versus Others’, 111 Psychological Review (2004). 73. O. Svenson, ‘Are We All Less Risky and More Skillful Than Our Fellow Drivers?’,47Acta Psychologica (1981). 74. See e.g. L. Babcock et al, ‘Biased Judgments of Fairness in Bargaining’,85American Economic Review (1995); R. Baumhart, An Honest Profit: What Businessmen Say About Ethics in Business (Holt, Rinehart and Winston, 1968). 75. D. Dunning, J.A. Meyerowitz and A.D. Holzberg, ‘Ambiguity and Self-Evaluation: The Role of Idiosyncratic Trait Definitions in Self-Serving Assessments of Ability’,57Journal of Personality and Social Psychology (1989). 570 Maastricht Journal of European and Comparative Law 32(5-6) be significant in the context of judicial independence because actors can perceive the standard for ‘independent behaviour’as relatively vague. As a result of such effects, there is a risk that judges have a different view of judicial independence than external actors. Internal and external perspectives are not congruent. In fact, such diverging perspectives on judicial independence appear to manifest themselves in two surveys conducted by the European Network of Councils for the Judiciary in 2019. 76 In these surveys, both judges and lawyers from 25 European countries were asked about several aspects of judicial independence. 77 When one compares the answers that both groups of persons have given to identically formulated questions, it is apparent that the internal perspective of the judges and the external perspective of legal professionals outside the judiciary often differ significantly. The external perspective often is more critical than that of the judges themselves. Table 1 contains an extract from this survey. It shows the results for the judiciary in Germany, the Netherlands, the Czech Republic and Greece. For these countries, this effect is very pronounced. Although not all countries in the survey show such a clear pattern for all the relevant questions, the listed results are nonetheless indicative of an effect of the internal and external perspective. Of course, one cannot directly infer from this data set which of the possible cognitive effects and knowledge-related differences has led to the diverging evaluation of the above questions. However, this data set shows that the evaluation tends to diverge in the same manner: external observers are often more critical than judges from the internal perspective. Yet the survey does provide an indication as to the effects that might have caused these divergent evaluations. If one also considers further data collected as part of the survey, this data indicates the relevance of the above-mentioned cognitive effects relating to a positive self-assessment. 78 It is noticeable that the judges in almost all of the 25 participating countries in 2019 rate their own personal independence higher than the independence of the judiciary in their respective country as a whole. 79 This is also the case for the 27 countries assessed in the 2022 ENCJ survey of judges. 80 These results –which are shown in Table 2 –emphasize the relevance of the subjective perspective on judicial independence. If, in addition, one takes into account surveys on the assessment of judicial independence by the general population, the tendency towards a more critical assessment from an external perspective compared to an internal perspective is also evident. For example, while the German judges in the above-mentioned ENCJ survey rate the independence of the German judiciary generally as high, with only 0.87% of the judges surveyed rating this independence as less than 5 points (on 76. ENCJ Survey on the independence of Judges 2019 –Independence and Accountability of the Judiciary, https://pgwrkwebsitemedia.s3.eu-west-1.amazonaws.com/production/pwk-web-encj2017p/Reports/Data%20ENCJ%202019% 20Survey%20on%20the%20Independence%20of%20judges.pdf and ENCJ/CCBE Survey among lawyers on the independence of Judges 2018–2019, https://pgwrkwebsitemedia.s3.eu-west-1.amazonaws.com/production/pwk-webencj2017p/Reports/ENCJ%20Survey%20on%20Independence%20Accountability%20of%20the%20Judiciary% 20among%20lawyers%20%202019.pdf. 77. The more recent survey of 2022 only included judges; a survey among lawyers was not conducted at this occasion. A direct comparison is thus not possible for 2022. 78. See also F. van Dijk, Perceptions of the Independence of Judges in Europe, p. 34. 79. Exceptions: Denmark with identical scores; and UK judges in England, Scotland and Wales with slightly more favourable overall scores compared to self-assessment. 80. ENCJ Survey on the independence of Judges 2022, https://pgwrk-websitemedia.s3.eu-west-1.amazonaws.com/ production/pwk-web-encj2017-p/GA%2022/Report%20ENCJ%20Survey%202022.pdf. Burchardt 571 a scale of 0–10), the assessment by the general public is more differentiated. The EU Justice Scoreboard for the same year 2019 showed that a total of 18% of the respondents in Germany rated it as ‘fairly poor’or ‘very poor’. 82 Similarly, while only 0,16% of the Dutch judges in the ENCJ survey rated the independence in the Netherlands as less than 5 points, the general perception in the Netherlands according to the EU Justice Scoreboard was 15% for ‘fairly poor’or ‘very poor’. While only 1.24% of the Czech judges rated the independence in the Czech Republic as less than 5 points, the general perception according to the EU Justice Scoreboard was 36% for ‘fairly poor’or ‘very poor’. And while 2.46% of the Greek judges rated the independence in Greece as less than 5 points, the general perception according to the EU Justice Scoreboard was 34% for ‘fairly poor’or ‘very poor’. 83 Table 1. Comparison ENCJ Survey 2019 results: judges vs lawyers 81 Question Judge (% of approval) Lawyers (% of approval) First appointment as a judge not only on the basis of ability and experience GE: 8% GE: 31% NL: 2% NL: 23% CZ: 15% CZ: 48% GR: 17 GR: 48% Appointment to the Supreme Court not only on the basis of ability and experience GE: 34% GE: 61% NL: 1% NL: 19% CZ: 16% CZ: 44% GR: 27% GR: 65% Promotion not only based on skills and experience GE: 28% GE: 38% NL: 8% NL: 27% CZ: 23% CZ: 45% GR: 23% GR: 62% Case allocations not according to established rules or procedures GE: 2% GE: 18% NL: 1% NL: 13% CZ: 3% CZ: 38% GR: 15% GR: 57% Judgments that went against the interests of the government were not implemented/enforced GE: 13% GE: 23% NL: 8% NL: 23,5% CZ: 10% CZ: 32% GR: 31% GR: 29% Decisions or actions of individual judges inappropriately/directly influenced by the media GE: 14% GE: 48% NL: 3,5% NL: 27% CZ: 16% CZ: 55% GR: 35% GR: 68% 81. Only the results of questions that were formulated identically for both groups of actors are listed here. 82. The data overview of the EU Justice Barometer 2019 is available at https://ec.europa.eu/info/sites/default/files/justice_ scoreboard_2019_quantative_data_factsheet_en.pdf. The specific question was: ‘From what you know, how would you rate the justice system in (our country) in terms of the independence of courts and judges? Would you say it is very good, fairly good, fairly bad or very bad?’ 83. For an in-depth assessment of further countries, see also F. van Dijk, Perceptions of the Independence of Judges in Europe, p. 36. 572 Maastricht Journal of European and Comparative Law 32(5-6) With regard to the trust of external actors, it is therefore not possible to refer solely to the internal perspective of judges. As the above examples show, it is very likely that the evaluation of independence issues from an internal perspective does not fully reflect that of external observers. It is not Table 2. Comparison of the assessment by judges of the overall independence in their country vs their own independence. Country Overall independence in the country 2019 Own independence 2019 Overall independence in the country 2022 Own independence 2022 Austria 8.9 9.3 8.8 9.3 Belgium 8.4 9.1 8.7 9.3 Bosnia and Herzegovina 7.1 8.4 7.0 8.6 Bulgaria 6.8 8.5 7.2 8.9 Croatia 6.6 8.1 7.4 8.8 Cyprus n/a n/a 9.5 9.8 Czech Republic 8.8 9.4 9.1 9.5 Denmark 9.8 9.8 9.8 9.9 Estonia n/a n/a 8.8 9.2 Finland 9.2 9.4 9.3 9.5 France n/a n/a 8.3 9.0 Germany 8.6 8.9 8.9 9.1 Greece 8.0 8.9 8.2 9.1 Hungary 7.3 8.8 8.1 9.1 Ireland 9.4 9.5 9.4 9.7 Italy 8.4 9.3 8.2 9.1 Latvia 6.5 7.1 7.2 7.8 Lithuania 7.7 8.3 7.5 8.0 Montenegro 8.4 9.3 7.8 8.8 Netherlands 9.0 9.2 9.2 9.4 Norway 9.2 9.5 9.3 9.5 Portugal 7.5 8.2 8.3 9.0 Romania 7.9 8.9 n/a n/a Slovakia 8.1 9.1 7.1 8.3 Slovenia 7.6 8.4 7.5 8.3 Spain 7.8 9.0 8.0 9.1 Sweden 8.6 9.1 8.8 9.4 United Kingdom: England and Wales 9.4 9.3 9.5 9.6 United Kingdom: Northern Ireland 9.3 9.4 9.7 9.6 United Kingdom: Scotland 9.5 9.3 9.4 9.5 Total 8.3 8.3 9.0 8.5 9.1 Burchardt 573 possible to equate the judges’perspective with the perspective of the external actors. The differences in perspective make an evaluation from the internal perspective alone an inadequate substitute. It is thus problematic when judges assess issues of judicial independence based solely on their own perspective as this perspective alone cannot do justice to the particularities of the external perspective. This problem exists both with regard to assessments of judicial independence in the narrower sense and with regard to impartiality. 84 In view of the objective of ensuring trust of those seeking justice and of the general public in judicial independence, the legal assessment of issues relating to judicial independence should take into account the perspective of this targeted group. In particular, this means considering a more critical stance than the one taken from an internal perspective. This would lead to a stricter application of the principle of judicial independence. Moreover, when taking into account the external perspective, the term ‘doubt’must be used as a yardstick. Taking into account the more critical tendency and the limited knowledge that is structurally linked to the external perspective, doubts must suffice in this respect. In line with this observation, the CJEU has emphasized that even ‘appearances may be of a certain importance’. 85 Other courts should also embrace this approach when assessing independence issues. To take the external perspective into account, empirical findings can be incorporated into the legal assessment of judicial independence issues. It is possible to refer, where applicable, to studies on how external actors assess specific organizational features relevant to judicial independence. 86 This applies both to courts when examining possible violations of the principle of judicial independence and to the legislator when designing behavioural settings that promote judicial independence in the first place. 87 In sum, a behaviour-related understanding of the principle of judicial independence requires a ‘double’empirical basis: on the one hand –as set out in section 3 –with regard to the factors that promote normatively desirable behaviour by judges and third actors; and on the other hand with regard to the subjective assessment of these factors from the perspective of external observers. In this way, the behaviour of all actors involved can be taken into account when implementing the objectives of the principle of judicial independence. 5. Extending the behaviour considered normatively undesirable under the standard of judicial independence? This final section relates back to the two levels of judicial decision-making outlined in section 2, focusing on the second level. As mentioned above, there are several behavioural factors that can influence the concrete content of a judicial decision such as heuristics and biases and effects 84. On this issue in connection with impartiality and the standard of review in English law: A. Higgins and I. Levy, 38 Civil Justice Quarterly (2019). 85. Joined Cases C-585/18, C-624/18 and C-625/18, A.K. and Others v Sąd Najwyższy, CP v Sąd Najwyższy and DO v Sąd Najwyższy, para. 128. 86. For example, J. Gutmann and S. Voigt, ‘Judicial Independence in the EU: A Puzzle’,49European Journal of Law and Economics (2020). The EU Eurobarometer also contains data on the factors that the population takes into account when assessing the independence of the judiciary, such as the status of the judge, influence by actors with economic interests and influence by the government and politicians. 87. Arguing that, due to the necessary empirical basis, these questions are better dealt with at the level of legal policy than by judges, because judges ‘have no special insights into public perception or training in the relevant science on human decision making’, A. Higgins and I. Levy, 38 Civil Justice Quarterly (2019), p. 394. 574 Maastricht Journal of European and Comparative Law 32(5-6) linked to the judges’social perspective. Often, these are factors that judges are not aware of and/or that are not under their control. Despite their well-established impact on judicial behaviour, such factors have so far rarely been linked to the concept of judicial independence. This section discusses whether and, if so, to what extent such factors should form part of this concept, that is, whether they should be considered as normatively undesirable with regard to judicial independence. This section focuses, by way of example, on one selected behavioural factor for judicial decision-making: the different social perspectives that judges bring to the decision-making process depending on the social group(s) to which they belong. This is a behavioural dimension of judicial decision-making that has been empirically researched for decades. 88 It bears mentioning that the influence of the judges’social perspective as discussed here is not about interest-based or ideology-driven decision-making. 89 It is not argued that judges consciously promote the interests of the social group to which they themselves belong and therefore decide accordingly. Rather, the focus is on more subtle phenomena that are usually not consciously perceived by the judges as decision factors. These may, for example, be related to the judges’personal experiences or to the unconscious effects of group identity. These types of effects will be addressed here. This reference to social perspectives also corresponds to an approach in the general diversity debate. 90 This approach is about integrating different social perspectives into social and political discourse and decision-making processes. The underlying assumption is that the fact that different social groups have different experiences leads to specific social perspectives that can, inter alia, contribute to better understanding the experiences and positions of fellow group members. This is also relevant regarding diversity (or lack thereof) in the judiciary. For example, the South African Constitutional Court has emphasized the importance of diverse social perspectives in the judiciary: ‘the sound belief behind the encouragement of greater diversity in judicial appointments was that women and visible minorities would bring an important perspective to the difficult task of judging’. 91 Further, scholars also emphasize the special perspective of certain groups that judges can bring to the task of judicial decision-making for judges in specialized courts such as commercial courts and for honorary judges who often have the function of bringing relevant non-legal perspectives to the court. 92 Studies have provided empirical evidence for the relevance of social perspectives for judicial decision-making. To start with, gender is a potential factor for such diverging social perspectives. Research on the decision-making behaviour of female and male judges has shown differences, at least in certain areas. For example, a study on the gender composition of the bench has shown that decisions in sexual harassment proceedings differ depending on whether or not there were female judges on the bench. 93 Different decision-making patterns were also identified for decisions 88. See S. Basabe-Serrano, L. Epstein and K. Weinshall, in L. Epstein et al. (eds), The Oxford Handbook of Comparative Judicial Behaviour. 89. For an overview on such factors: R. Solberg and E.N. Waltenburg, ‘Attitudinal Judging: Partisanship and Ideology’,in L. Epstein et al (eds), The Oxford Handbook of Comparative Judicial Behaviour (OUP, 2024). 90. See e.g. I.M. Young, Inclusion and Democracy (OUP, 2002); C. Röhner, ‘Von Repräsentation zu demokratischer Gleichheit –Politische Teilhabe und gesellschaftliche Ungleichheit’,59Der Staat (2020). 91. Constitutional Court of South Africa, decision of 4 June 1999 (President of the Republic of South Africa and Others v. South African Rugby Football Union and Others), CCT16/98) [1999] ZACC 9; 1999 (4) SA 147; 1999 (7) BCLR 725, para. 42 (emphasis added). 92. E.g. W. Meyer, in I. von Münch, P. Kunig, J.A. Kämmerer and M. Kotzur (eds), Grundgesetz Kommentar, para. 44. 93. J.L. Peresie, ‘Female Judges Matter: Gender and Collegial Decision Making in the Federal Appellate Courts’, 114 Yale Law Journal (2005). Burchardt 575