The Case for Judicial Councils as Fourth-Branch Institutions

Judicial councils – separation of powers – fourth-branch institutions – judicial independence – four ideal types of judicial councils – a judge-controlled, politician-controlled, inter-branch and fourth-branch judicial council – danger of politicisation and corporativism of the judiciary

these rationales, commonly associated with the 'fourth branch', 4 judicial councils have rarely been classified as 'fourth-branch' institutions. 5By 'fourth-branch institution' we mean an institution that is a standalone body which derives its legitimacy independently of all three classical branches of power.In contrast, in Europe, judicial independence has been prioritised over judicial accountability 6 and it has been taken for granted that judicial councils belong to the judicial branch because they concern the judiciary. 7But no one has provided any theoretical underpinning for this assumption.
The main reason is probably the fact that while judicial councils, as we show below, can be conceptualised as part of the fourth branch, they are unique among the fourth-branch institutions as they regulate one of the original three branches. 8oreover, judicial councils can be captured by any of the three original branches with similar fervour.This is not a mere theoretical possibility.Several judicial councils in fact have been captured by both political branches (Poland 9 being the 4 M. Tushnet, The New Fourth Branch: Institutions for Protecting Constitutional Democracy (Cambridge University Press 2021); T. Khaitan, 'Guarantor Institutions', 16 Asian Journal of Comparative Law (2021) p. 40.We explore the rationales of the fourth branch below in the first section.5 For a rare example of a discussion of judicial councils as a part of the fourth branch see International IDEA, Constitution-Building Primer 19, 'Independent Regulatory and Oversight (Fourth-Branch) Institutions', 24 November 2019, https://www.idea.int/sites/default/files/publications/independent-regulatory-and-oversight-institutions.pdf,visited 8 February 2024, p. 26, 47 and 50-51.See also B. Iancu, 'Government and Governance: The Constitutional Politics of Institutional Neutrality', in B. Iancu and E. Tănăsescu (eds.),Governance and Constitutionalism: Law, Politics and Institutional Neutrality (Routledge 2019) p. 27-29 (characterising judicial councils as autonomous administrative agencies and arguing that their position is problematic from the perspective of the principle of separation of powers).6 More recent documents of the Council of Europe, Venice Commission and other supranational bodies rhetorically acknowledge that judicial councils should ensure both independence and accountability, but they reconceptualise judicial accountability in such a way that it becomes toothless, and according to these standards judges should still dominate judicial councils.

7
In some European countries there have been scholars who questioned that assumption, but these voices have been sidelined by the growing number of supranational standards and, more recently, by the case law of the European Court of Justice and the European Court of Human Rights, which follow these standards whole-heartedly (see e.g.ECtHR 15 March 2022, No. 43572/ 18, Grzęda v Poland; and infra nn. 12 and 64).The Case for Judicial Councils as Fourth-Branch Institutions prototypical example) and the judiciary (e.g. in Slovakia 10 or Georgia 11 ) and used for sinister goals.Even in some established democracies in Western Europe judicial councils do not operate smoothlyto put it diplomaticallyand have become heavily politicised (e.g. in Spain 12 ) or subject to battles between various factions within the judiciary represented by judicial associations (e.g. in Italy 13 ).It is thus unsurprising that judicial councils dominated by judges have witnessed academic pushback as well as political backlash. 14y discussing these specific features of judicial councils from the separation of powers viewpoint, conceptualising them with the fourth-branch logic in mind and analysing their interaction with the other three branches, we hope that we can illuminate some of the more general debates concerning the fourth-branch institutions (sometimes referred to as 'guarantor institutions' 15 or the 'integrity branch' institutions 16 ).In doing so, we aim to contribute to two strands of scholarship: to the literature on the separation of powers and to the judicial studies literature.
From the perspective of the separation of powers, judicial councils remain undertheorised.Little or no attention has been paid to how the delegation of governance competence, previously held by the executive power, has affected the architecture of state power within political systems.Moreover, there is a growing understanding that strong judicial councils in the hands of judges, favoured by supranational actors, can yield suboptimal results and may invite backlash from politicians.Yet, a return to the previous executive-and ministry-led models of judicial governance is not possible, since these were often even more problematic. 17Our aim is to break the impasse.
Our argument is twofold.First, we show that, depending on their position within the separation of powers, judicial councils can follow four ideal types: judge-controlled; politician-controlled; inter-branch; and fourth branch.Second, we argue that redesigning judicial councils as fourth-branch institutions may best protect them against the two greatest dangerscorporativism and politicisation.In doing so, we swim against the tide and question the majority of existing European standards concerning judicial councils, and potentially also some case law of both European supranational courts. 18However, we provide thorough theoretical arguments for our position and believe that supranational actors would benefit from our study as well.
We want to add four caveats here.First, the idea of differentiating between understandings of the placement of the judicial council within the branches of power builds on 130 interviews with judges, politicians and legal practitioners on issues of judicial self-governance, which we conducted in six European jurisdictions in 2016-2019.It was in the in-depth elite interviews that the discrepancy in actors' understandings of judicial councils emerged for the first time and where we found the initial inspiration to write this article, which, however, is theoretical and normative. 19Second, for the purposes of this article we do not engage with scholarship discussing whether the post-tripartite separation of powers institutions are best framed as fourth-branch qua separate branch or, for example, guarantor institutions simply standing beyond the trinity of state branches.
Third, in this article we narrow down our argument to judicial councils alone.By a judicial council we mean a constitutionally mandated body endowed with significant decision-making powers concerning the careers of judges as well as court administration.We are aware that in many countries the court services, 20 17 One can clearly see the consequences of the re-empowerment of the Minister of Justice in Poland, where the functions of Minister of Justice and General Prosecutor were merged and the Minister of Justice fired dozens of court presidents at a whim: F. Zoll and L. Wortham, 'Judicial Independence and Accountability: Withstanding Political Stress in Poland ', 42 Fordham  International Law Journal (2019) p. 875.18   Grzęda v Poland, supra n. 7; Lorenzo Bragado and Others v Spain, supra n. 12. 19   We analyse the interviews from the sociological point of view elsewhere.See D. Kosař et al.,  European Perceptions of Judicial Self-Governance (Cambridge University Press forthcoming).merit commissions21 and judicial appointment commissions 22 play a similar role and we believe that our argument is applicable also to them.But these bodies have narrower competences and are often not constitutionally entrenched, and thus their conceptualisation raises additional questions we cannot discuss here. 23In other words, this article has some relevance to these alternative models, but it focuses directly only on judicial councils.Finally, the question of what place judicial governance occupies in the separation of powers has a global outreach, 24 but in this article we focus on Europe and the policy debates therein.That explains why we engage primarily with the guidelines of the European supranational bodies and leave aside the global standards developed by the United Nations 25 and other agencies operating worldwide.
This article proceeds as follows.The first section sketches the core principles of the separation of powers and introduces the nature of and relationships among individual branches.It transcends Montesquieu's trinity and explains the rationales behind the so-called fourth-branch institutions.The section after that situates judicial councils in the broader political development and explains their rise as well as the more recent pushback and backlash against them.The following section then defines four ideal types of judicial councils from the separation of powers point of view.The penultimate section identifies the drawbacks and limits of all four ideal types and provides the case for reconceptualising judicial councils as fourth-branch institutions.The last section concludes.

S  :         
The principle of the separation of powers is a notoriously vague and contested concept. 26At the most general level, it requires power in constitutional systems to be divided in some sense and seeks two, often competing, aims. 27The first aim is to temper power and secure that the institutions in the system do not slide into accumulating too much power, potentially turning governance into tyranny (the negative rationale).The second aim, which stresses collaboration between branches (joint enterprise of governance), is to facilitate 'good' governance, serving core constitutional values, such as vindication of the expression of political voice or the protection of individual human rights (the positive rationale).This section first briefly describes the classical tripartite model of the separation of powers principle built around the idea of three branches.Then it introduces a more recent conception, adding to the structure of governance the possibility of a fourth category, the so-called new fourth branch.
At least since Montesquieu, constitutional theory has been dominated by the notion that all power entrusted to the government should be divided into three branches: the legislative, executive and judiciary. 28As classically stated by M.J.C. Vile, the pure version of this conception requires the division of government: into three branches or departments, the legislature, the executive, and the judiciary.To each of these three branches there is a corresponding identifiable function of government, legislative, executive, or judicial.Each branch of the government must be confined to the exercise of its own function and not allowed to encroach upon the functions of the other branches.Furthermore, the persons who compose these three agencies of government must be kept separate and See Kilbourn v Thompson, 103 US 168, 190 (1881) ('It is believed to be one of the chief merits of the American system of written constitutional law, that all the powers entrusted to government, whether State or national, are divided into the three grand departments, the executive, the legislative, and the judicial'); Möllers, supra n. 26; Waldron, supra n. 26.
The Case for Judicial Councils as Fourth-Branch Institutions distinct, no individual being allowed to be at the same time a member of more than one branch. 29 practice, the three elements of separation (institutional, functional and personal) have been complemented by the fourth element of checks and balances, which empowers each of the three branches 'to modify positively the attitudes of the other branches of government'. 30In this way, the branches should be able to check each other for abuse of functions not assigned to them and through that 'keep each other in their proper place'. 31hile the tripartite model might have stood the test of time, in recent years its adequacy as a tool for analysing and assessing governmental structure has been repeatedly questioned. 32Around the world, modern constitutional systems entrust some governmental functions to institutions which may not easily be categorised as legislative, executive or judiciary.Chapter 9 of the South African Constitution, for example, provides for State institutions supporting constitutional democracy, among which it lists seven institutions, including a human rights commission, an auditor-general and an electoral commission. 33The Constitution describes these institutions as 'independent, or subject only to the Constitution and the law' and provides that their members are appointed by the National Assembly and have security of tenure. 34The Constitution of the Slovak Republic establishes an office of The Public Protector of Rights, defining it as 'an independent body of the Slovak Republic', charged with 'protecting basic rights and freedoms of natural and legal persons in proceedings before : : : bodies of public authority'. 35  (Hart Publishing, 2023).See also Iancu and Tănăsescu, supra n. 5, framing an analysis of neutral independent agencies as a challenge to the classical triad of powers or branches of government.
for a criminal conviction or long-term health incapacity. 36And virtually every country has a central bank, a body that manages the state's monetary policy, is free to choose its own means of pursuing it without political interference, and whose members have security of tenure. 37espite their apparent diversity, the functions undertaken by these 'misfit institutions' share some common features.Drawing on the recent constitutional law scholarship, it is possible to identify three constitutive features of these institutions. 38irst, the functions which these institutions undertake consist in protecting or realising a constitutional norm, in the sense that the basic idea of how the function should be exercised should not be negotiable through the normal political process. 39For example, we might argue that regardless of who the current political majority is, there is a shared acceptance that there should be monetary stability, that elections should be fair or that human rights should be protected.This is to be distinguished from issues like taxation or criminal policies, where the specifics of these policies are determined through the usual process of political debate.
Second, the functions described are distinct in the sense that their proper exercise according to the shared constitutional norm might go against the short-term interest of the current political majority. 40The governing political force might be tempted, for example, to interfere with elections, print new money to secure reelection or sweep human-rights violations under the carpet.The political branches (the executive and the legislature) are therefore often in a 'conflict of interest', and hence not ideally placed to execute such functions. 41inally, the third characteristic of those functions is that while there is an established constitutional norm prescribing how they should be exercised, concrete realisation of the norm still requires a great degree of political judgement. 42Securing fairness of the election, for example, necessarily requires political judgements on how to draw electoral district boundaries with an eye for 'representativity' and management of monetary policy necessarily has a political component (it is still a policy), potentially impacting on the distribution of wealth.Not being entirely determined by the overarching constitutional norm, the realisation of all these tasks necessarily requires the delivery of a political judgement, and thus also a dose of democratic legitimacy.For this reason, courts, tasked in general with 'applying the law' and not being well designed to deal with polycentric problems, may not be well placed for the task. 43ith none of the three classical branches being in an ideal position, the solution adopted by many constitutional systems is to put in charge an institution that is separated from the classical branches.What this entails is that these institutions are systematically, functionally and personally separate from the three classical branches.At the same time, just like the three classical branches, these institutions also enter into the relationship of mutual checks and balances vis-à-vis the other branches, ideally resulting in a balance between independence and (political) accountability, which is generally considered as an ideal for executing the entrusted task.It is this constitutionally protected degree of independence that secures that the institutions are not reducible to the will of any of the other governmental branches and allows them to be conceptualised as forming part of a distinctfourthconstitutional branch.
The idea of adding a fourth category to the separation of powers model is not without controversy.Three lines of criticism have emerged in the literature.First, some might be hesitant to put the described institutions on an equal footing with the three classical branches as, unlike those classical branches, fourth-branch institutions are much more diverse and less universal.One may speak of the 'fourth branch' as a type of constitutional archipelagoan expanse of water with many islands. 44Fourth-branch institutions thus do not serve a particular unifying function common across political systems.Instead, the mere creation of these institutions may be a reaction to a pathology of the respective constitutional system, where none of the three classical branches is in a position properly to carry out a given function. 45In this sense, the fourth branch differs from the three classical branches, which seem to be accepted and, in some form, adopted in mostif not allmodern constitutional systems.
Second, it is not entirely clear whether fourth-branch institutions deliver the promised goods. 46Their track record is mixed.Some seem to work well, while 43 For the notion of polycentric problems and their distinction from 'legal' problems see L.L. Fuller, 'The Forms and Limits of Adjudication ', 92 Harvard Law Review (1978) p. 353.others have become prone to concentration of interests or capture. 47Recently, Tarun Khaitan defended the normative claim that in order to function effectively (to guarantee the relevant norm) the design of guarantor institutions should optimise: (1) sufficient expertise and capacity to perform their functions effectively; (2) sufficient independence from political, economic or social actors with an interest in frustrating the relevant norm it is meant to guarantee; and (3) sufficient accountability to bodies with an interest in upholding the relevant norm. 48Whether these three conditions suffice must be tested empirically though, and we will not explore this route in this article.
Finally, fourth-branch institutions may be controversial from a normative standpoint.If we think that there are functions that should be insulated from the three classical branches it may be useful to create a fourth category for them.Yet, deciding which functions should be shielded in this way may be a highly contentious issue, depending ultimately on deep normative convictions concerning values which should be achieved in a constitutional democracy.
To sum up, traditionally the principle of the separation of powers structures the government into three branches.Recent constitutional practice, however, has come up with institutions which are independent of the three classical branches and which could usefully be conceptualised as part of a separate, fourth branch.Despite their diversity, these 'fourth-branch institutions' share three featuresthey are tasked with functions (1) which consist of the protection or realisation of a constitutional norm not negotiable through the regular political process; (2) whose proper exercise according to the shared constitutional norm might go against the short-term interest of the current political majority; and (3) the realisation of which requires political judgement and a greater degree of democratic legitimacy.With this theoretical background, the rest of the article explores the function and constitutional place of one of the most discussed institutions in recent constitutional law and political science scholarshipjudicial councils.

R     
This section briefly explains the motivation behind establishing judicial councils and contextualises their emergence as the shift in competences between the three state powers.Subsequently, we analyse the recent political backlash in some countries against judicial councils dominated by judges and the arguments invoked in these reversal policies.We also engage with scholarly criticism of such judicial councils that identifies different problematic features and, unlike some of the political backlash, does not advocate the capture of the judiciary.This prologue is necessary for contextualising judicial councils and conceptualising them as potential fourth-branch institutions in the sections that follow.
Historically, judicial governance was initially in the hands of sovereign kings and emperors and their representatives in the colonies.Later on, republicanism transferred these powers from the sovereign rulers to the executive branch.In most countries, the Presidential administration, the Ministry of Justice or a similar body usually played a key role.For instance, in Europe, judicial governance was vested with the Lord Chancellor in England, the Oberste Justizstelle in Austria, Le Ministère de la Justice in France, the Reichskanzler in Germany and Il Ministero della Giustizia in Italy. 49In some countries, the judicial career part of judicial governance was regulated by a different body from the court administration part.
It was only in the wake of the Second World War that we could start witnessing a gradual transfer of judicial governance from political branches to bodies containing a significant number of judges. 50In this period, the first two judicial councils as we understand them todayas independent and autonomous bodies endowed with significant powers regarding judges' careers and the  We are aware that in only a few countries was judicial governance in the hands of the Supreme Court judges even before the Second World War.For instance, Mexico granted the Supreme Court the power to handpick lower court judges and oversee their careers as early as in 1917 (J.R. Figueroa, 'Populism and Democratic Erosion: The Role of the Judiciary Evidence from Mexico, 2018-2021', 198 Revista de Estudios Politicos (2022) p. 187).In Europe, for instance, Italy established its first Consiglio superiore della magistratura in 1907, but this judicial self-governance body had far less power than judicial councils have today (see S. Benvenuti, 'The Politics of Judicial Accountability in Italy: Shifting the Balance', 14 EuConst (2018) p. 369 at p. 373-374).

92
David Kosař et al.EuConst (2024)     management of the judiciary 51appeared.In 1946, the French Constitution of 27 October 1946 entrenched the Conseil supérieur de la magistrature.A year later, the Italian Constitution of 27 December 1947 laid the foundations of the Italian Consiglio superiore della magistratura.These two judicial councils resulted from different historical, cultural and social contexts and they sought different aims, 52 but they were the pioneers.The rise of judicial councils reflected the need to formalise the growing importance of courts and design a new institutional set-up that would ensure a proper balance between judicial independence and judicial accountability. 53ventually, judicial councils spread across the world. 54In the late 1970s, they were established in Spain, Portugal and Greece, coinciding with the fall of their authoritarian regimes.Judicial councils, like the newly-created constitutional courts, were supposed to facilitate the transition to democracy and separate judiciaries from the influence of the executive power.Judicial councils also found their way into many Latin American countries. 55After the end of the Cold War, most countries in post-Communist Europe happily embraced judicial councils as well, sometimes voluntarily, sometimes under the pressure of the Council of Europe and the European Union. 56Soon, judicial councils started to find their This is often forgotten, and these two judicial councils are put in the same basket.For further details see Benvenuti, supra n.50 and T.S. Renoux (ed.), Les Conseils supérieurs de la magistrature en Europe : actes de la table ronde internationale du 14 septembre 1998 (La Documentation Française 1999).way also beyond the post-authoritarian or post-totalitarian context, in consolidated democracies.Most recently, Ireland set one up in 2022. 57Apart from Europe and Latin America, a similar development has taken place in Africa, 58 Asia 59 and Caribbean States. 60n Europe, the shift in competences expressed by the establishment of judicial councils was typically justified by four motives: to strengthen judicial independence; to insulate judges from potential political interferences; to achieve a more efficient judiciary; and to increase the social legitimacy of, as well as public confidence in, the courts. 61The reasoning behind the introduction of judicial councils demonstrates that judicial councils were expected to be better suited both to secure systemic insulation of the judicial power from the influence of the other two political branches and to bring more expertise to the substantive governance of the judicial branch.
It was believed that judges were better suited to decide on the governance of the judiciary as they possessed the necessary skill, understood best the internal functioning and the needs of courts, and were not dependent on partisan policies and electoral preferences.There was also an unstated accompanying assumption that judges will not have an interest in frustrating the major norm (judicial independence) the judicial council was meant to guarantee. 62No one expected that judges in some countries might use judicial insulation via a judicial council for personal gain, for nepotism and for settling the score with their critics within 57 the judiciary. 63There was thus no need to ensure that the judicial council was sufficiently independent of judges.
During all this development, little to no attention has been paid to the question of how the delegation of competence from the executive to judicial councils has affected the separation of powers.Instead, supranational organisations, the advisory bodies of the European Union and the Council of Europe in particular, pushed forward an understanding of a judicial council as an autonomous judicial body, detached from the two other branches. 64The fact that the adjective judicial does not mean that judicial councils were meant to exercise the function of the judicial branchthat belongs to the courts onlywas sidelined.Judicial councils were simply meant to be composed of and controlled by judges, and execute the governance of the judicial branch, which was delegated to them from the executive power, in order to secure a better insulation of the judiciary from politics.
Assuming that judicial councils are necessarily judicial bodies raises at least three problems.First, from the institutional perspective, the characterisation of a judicial council as a judicial body ignores the complexity of the reality.The composition and powers of judicial councils do not correspond to the idea of autonomous self-governance within the judicial branch.In fact, the majority of judicial councils are actually built on a principle of parity between judges and 63 We now know that this was a rather naïve assumption which backfired especially in Central and Eastern Europe.See K. Šipulová and S. Spáč, 'No Ghost in the Shell', German Law Journal (forthcoming); N. Tsereteli, 'Constructing the Pyramid of Influence', German Law Journal (forthcoming).

64
These standards are so numerous that we cannot do justice to all of them here.The Case for Judicial Councils as Fourth-Branch Institutions non-judges or will at least include several non-judges. 65Countries like France 66 and Slovakia 67 intentionally reduced the number of judges in their judicial councils to reduce judges' control within these bodies.Even supranational organisations are slowly abandoning the idea of judge-filled judicial councils and recommending a mixed composition, counter-balancing the voice of judges within them with that of lay members. 68Considering judicial councils simply as judicial bodies thus seems to be hardly defensible.
Second, what judicial councils do is still governance, even if it might directly influence the functioning and independence of the courts.In this respect, necessarily, the perspective and position of those administering the courts should be different from those tasked with applying the law, the judges.Governing is simply not judging.
Third, from the theoretical point of view, the transfer of judicial governance powers to the judiciary itself not only empowers and emancipates the courts from the political branches but also removes some of the key checks and balances existing in the system. 69Judges who elect and are accountable to themselves, with little or no involvement of the political branches, lose important sources of democratic legitimacy and oversight.As a result, the establishment of judicial councils has not led to a mere increase in judicial independence and efficiency as narrated by supranational organisations.Instead, we have witnessed the emergence of a whole new institutional set-up that has shifted the balance among the three branches of power and strongly empowered the judicial branch.Surprisingly, the repercussions of such an institutional earthquake have remained largely undertheorised and thus necessarily driven by many misconceptions on both the theoretical and policy levels.
The lack of theoretical grounding was eventually reflected also in practice.While many actors do support the introduction or existence of judicial councils, they also understand their role, composition and overall placement within the separation of powers differently.Moreover, within the last decade judicial councils have received significant academic pushback as well as political backlash. 70

96
David Kosař et al.EuConst (2024)     to retaliate against the rise of those bodies and against strong judicial councils in particular.This pushback and backlash against and threats of withdrawal from the judicial council model showed clear disagreement among the key actors on how to conceptualise the judicial council from the separation of powers point of view.Many judges in Poland and Hungary objected to the reforms adopted by Viktor Orbán and Jarosław Kaczyński, arguing that they violated the separation of powers as judicial governance belongs to the judicial branch.Orbán and Kaczyński clearly disagreed and pushed for their own, largely unarticulated, vision of the separation of powers.Their vision rejects complete separation of powers and leaves room for the political branches to have an influence on the judicial branch. 71According to their rhetoric, checks and balances should work both waysjudicial checks upon the political branches and vice versa.
However fascinating this theoretical disagreement may be, it has been severely undertheorised in the literature.In a way, the rise and fall of judicial councils represents a good example of the changing domestic architecture of separation of powers without an architect.The next section remedies this problem and identifies four understandings of judicial councils and their respective place within the theory of the separation of powers.

F           
As we demonstrated in the previous section, the conceptualisation of judicial councils as autonomous judicial bodies endowed with judicial governance competences is too simplified and is not theoretically grounded.The key scholarly works discuss primarily second-order questions of composition, selection and competences without a proper understanding of what consequences the choice of a particular type of judicial council has for the balance between the three governmental branches.
But if judicial councils do not necessarily represent the judiciaries and their self-governance function, then where might they sit in the three branches of state The Case for Judicial Councils as Fourth-Branch Institutions power?In this section we show that, depending on their relationship to the executive, legislature, and judiciary, judicial councils can follow one of four conceptualisations: (1) a judge-controlled judicial council, which is a selfgoverning representative of a judicial branch; (2) a politician-controlled judicial council, which formally represents the judiciary but is de facto controlled by political actors; (3) an inter-branch judicial council, which works as a coordinating institution representing all three traditional branches of power; and (4) a fourth-branch judicial council, which is autonomous of any of the three powers (Table 1).
The core element that differentiates the four Weberian ideal type 72 models of the judicial council is their position within the separation of powers.The models are built on a presumption that the position of the judicial council within the separation of powers reflects the behavioural expectations from relevant actors, which in turn are reflected in the institutional design of the councils. 73In this sense, the models are grounded in reality, and reflect behaviour and existing institutional elements.At the same time, as ideal types, they are necessarily We build loosely on the agent/principal dilemma, which discusses a conflict between the representative (principal) who delegates a power and authorises an individual (agent) to act on their behalf.If the interests of agents and principals do not coincide, agents may deviate and stop acting in the best interests of the principal.Typically, this risk increases if an agent represents several principals.For the purposes of our models, we work with institutional expectations the interests of which agents (i.e. the members of the judicial council) are supposed to represent and suggest main features that should lower agency costs (risks of deviation).For the application of the dilemma in the judiciary see K.

98
David Kosař et al.EuConst (2024)     analytical constructs formed through accentuation and generalisation of selected institutional features or existing behaviour expectations so that they create a rational whole.Therefore, while in practice existing judicial councils usually lean towards one of the types, most councils are designed with features belonging to more than one model.
In our analysis, we zoom in on the composition 74 of the judicial council rather than on its competence. 75We focus primarily on the questions of who selects the members of the judicial councils and who can be selected as members of that body.Depending on the position of the judicial council in the separation of powers, these two questions outline whose interests the judicial council should represent and how free or constrained the judicial council's members are in the execution of their mandates.This logic also reflects the presumption that principals responsible for selecting judicial council members pursue a mix of preferenceseither related directly to judicial governance policies, or to substantive policy outcomes they wish to impact through the means of judicial governance.The balance between these two preferences impacts their choices and the expectations laid on members judicial councils.Besides that, however, we analyse also other institutional elements of judicial councils which may manifest a certain understanding of the position of these bodies within the separation of powers, such as who the actors responsible for the judicial council's budget are, who assumes the presidency of the judicial council, and whoand under what conditionsmay remove the judicial council's members.
In what follows we explore each of the models in more detail and discuss its rationale and the consequential institutional features.One caveat must be added here.The four ideal types of judicial councils are constructed on the basis of the separation of powers principle in democracies with a unitary form of government.We are aware that our models may have limited transferability to federal states where there is more nuanced interplay between the federation and its units.

A judge-controlled judicial council
The understanding of a judicial council as a body belonging under the umbrella of judicial power comes the closest to the judicial council model advocated by international organisations and other supranational bodies, which continuously push transitioning countries to tilt judicial governance in favour of the judges.The 74 For the political importance of personal choices see Parau, supra n. 56, p. 125.75   This does not mean that the competences of the judicial councils are not important.They are.But it is not a criterion that distinguishes our ideal types from the separation of powers standpoint.Moreover, we define judicial council narrowly so as to exclude court services as well as mere judicial appointment commissions, and hence we assume that judicial councils have roughly the same powers.
Venice Commission, Consultative Council of Judges and European Network of Judicial Councils traditionally understood judicial councils as bodies 'independent of the executive and legislature, and : : : responsible for the support of the Judiciaries in the independent delivery of justice', 76 and 'free from any subordination to political party considerations'. 77hese supranational bodies eventually abandoned the idea of judicial councils composed solely of judges 78 in favour of mixed composition and acknowledge that lay members may secure a better diversity in opinions and independence of the judiciary. 79However, they still lean heavily towards the idea of judicial councils subsumed under the judicial branch. 80The latest recommendations of European supranational bodies therefore still stress a 'substantial judicial representation chosen democratically by other judges', 81 who act as representatives of the entire judiciary. 82The conceptualisation of judicial councils as the loudspeakers of judges frequently appears both among elites of the post-Communist area 83 and in Western Europe. 84he judge-controlled model thus understands the judicial council as a part of the judicial branch.Judicial councils take over a function in judicial governance previously held by the executive power 85 and in its implementation represent the 76 European Network of Judicial Councils (ENCJ), Information provided by the ENCJ, July 2014, p. 4, https://e-justice.europa.eu/fileDownload.do?id=bbd51f97-e402-4f98-86ad-0456297a9fd4, visited 8 February 2024.
interests of the judiciary.The delegation of the function is not absolute.The other branches (e.g. the executive via the Minister of Justice or the President, the legislature often via the upper chamber of the Parliament) still take part in the governance, but the judicial council is the primary representative channel of the judiciary in the judicial governance arena.Its main role is to guarantee judicial independence by taking the selection and appointment of judges of general courts away from the political branches, and to ensure that political or partisan considerations do not prevail. 86he subsumption under the judicial branch also translates into control over the composition of the judicial council and the autonomy of its members.The majority of members of the judge-controlled judicial council are selected by the judicial branch, while they are also expected to act on behalf of the judiciary and represent its interests. 87It is worth noting that a judge-controlled model presumes that the interests of the judiciary automatically overlap with the interest in protecting judicial independence, maintaining the rule of law and the effective delivery of justice.
The tilting of a judicial council towards the judicial branch can be further strengthened by a set of other institutional features.The first is the domination of the membership by judges, who are expected automatically to share the interests of the judiciary by reason of both their professional role conception 88 and loyalty to the principal (their colleagues on the bench) who selected them.This presumption leads to a model which counts on the majority of members being judges selected by their peers, as the best guarantee of insulation from political and partisan interests.This creational logic is manifested in virtually all the judicial councils established in countries of Central and Eastern Europe. 89he second design feature strengthening the judge-controlled model is that the presidency is vested in the hands of a judge or a chief justice.A judge-controlled judicial council typically welcomes judicial authorities as a source of expertise, particularly in those jurisdictions where court presidents play an important role in judicial governance and enjoy many formal and informal powers. 90In such scenarios, however, judges and court presidents inside the judicial council play dual roles.First, they act as a safeguard against partisan interests being reflected in the appointment of judges.Second, their presence is considered vital for their expertise in other areas of judicial governance. 91he strongly pro-judicial nature of this model reflects also in the accountability of the members of the judicial council.While the pure theoretical model might advocate control of principals over the accountability of judicial council members if they underperform, the presence of judges in the judicial council makes this principle difficult to achieve.In practice, judges in the judicial-controlled model are protected by the principle of judicial independence, one that relates primarily to their role in judicial decision-making, which may however spill over also to their other roles.In fact, the European Court of Human Rights recently held that judges should receive similar protection from their membership of a judicial council as from their normal adjudicatory role. 92Consequently, the judgecontrolled model conditions the removal of judges from the judicial council with stringent criteria.This principle generally decreases the political accountability of the whole body. 93n conclusion, the judge-controlled model of judicial council is a body that belongs to the judicial power and represents the voice of the judiciary in judicial governance.It takes over the competence previously exercised by the executive.Although the delegation of power is not absolute, it still shifts the balance between the three branches by strengthening the judiciary.

A politician-controlled judicial council
The second ideal type, the politician-controlled judicial council, at first sight resembles the judge-controlled judicial council but in fact stands in stark contrast to it.While it seemingly transfers power previously held by the executive to a new independent body, it retains the political branches' control over judicial governance.How is that possible?The political branches select the members of the judicial council.The legislature and the parliament may select politicians as their representatives and expect them to represent their interests in governing the judiciary.With the judicial council being a manifestation of self-governance, however, in practice this model often morphs into a version in which the political branches select judges and expect them to act as their agents (i.e.bound by the preferences of politicians as their principals).
While the politically controlled model may appear counterintuitive, the rationale behind it relies on a need to link the legitimacy of a body responsible for judicial governance to a directly-elected actor.The politician-controlled judicial council can therefore still be dominantly composed of judges, but the political branches provide them with the democratic legitimacy to govern (i.e. to execute a public function) via the act of election. 94This leads to a model with high political accountability but relatively low independence.
Regarding the manifestation of accountability, similar observations to those raised in the case of judge-controlled judicial councils still apply.The pure theoretical model would support the imperative mandates of judicial council members, including removability at will.Nevertheless, the presence of judges inside this model in practice complicates this logic due to a clash with principles of judicial independence.The political branches will thus typically control the composition of a judicial council only via the act of selection.
The closest model we can find in practice to the politician-controlled judicial council is the Polish National Council of the Judiciary since the controversial 2017 reform. 95The reform abandoned the previous guiding principle of the composition of the Polish judicial council, which was to include representatives of different branches of power and to ensure that the judicial members of the council were elected by judges.In the revised council, 15 judges previously elected by other judges are now elected by the Sejm (the Polish Parliament).The reform of the National Council was heavily criticised by both academia 96 and supranational bodies 97 as a politicised legal overhaul that interfered with the independence of 94 F. Wittreck, 'German Judicial Self-Government -Institutions and Constraints', 19 German Law Journal (2018) p. 1931.

95
Śledzinska-Simon, supra n. 9. 96 E.g.Zoll and Wortham, supra n. 17, Sadurski, supra n. 9.The Case for Judicial Councils as Fourth-Branch Institutions the Polish judiciary.The same logic stood behind the reform of judicial governance in Hungary.Viktor Orbán just chose a different pathhe created a brand-new judicial governance body and disempowered the existing judicial council.
However, politician-controlled judicial councils also appear outside the countries experiencing democratic regression or populist attacks against the judiciary.A typical example is the Spanish General Council of the Judiciary, which is composed of 20 members, 12 of whom are judges and eight are lawyers (nonjudges).As the 12 judges are elected by two chambers of the Spanish parliament from lists of candidates proposed by judges and judicial associations, the Spanish judicial council is also relatively strongly controlled by politicians. 98

An inter-branch judicial council
The inter-branch type of judicial council functions as a platform that represents the interests of all three branches in governing the judiciary.It is composed of judges and politicians, however, with equal representation of each of the three powers.Even more importantly, the three powers are also equally represented in the selection process (each power selects one-third of the members). 99This model therefore rests on the principles of equality and balance.Members of the interbranch judicial council are expected to represent the interests and act on behalf of their principals (e.g. the branch that selected them).In practice, the easiest way to secure the corresponding role conception of the judicial council's members is to allow each of the three powers to select only its own members as its representatives within the council.The manifestation of this model can be found for example in Slovakia, which established a judicial council whose composition is determined (although not equally) by the judiciary, legislative and executive power.After an initial informal practice of political actors nominating judges, Slovakia changed the wording of the Constitution to clarify that political branches cannot select judges as their representatives inside the judicial council.
This model therefore closely follows the tripartite separation of powers and can also be described as a platform or a channel where representatives of the three powers negotiate their preferences on judicial governance.Its essence is therefore deeply political, as the judicial council reflects the polarisation of society and interest groups inside the original three branches.The coordinating character of a 98 Torres Pérez, supra n. 12; Lorenzo Bragado and Others v Spain, supra n. 12.

99
In this respect, the model comes rather close to the latest opinion of the Consultative Council of European Judges which still requires the Council to have a majority of judges elected by their peers, but also recommends the participation of both the executive and legislative powers in selecting the rest of the members.See Consultative Council of European Judges, supra n. 64.

104
David Kosař et al.EuConst (2024)     judicial council can be further strengthened by the strategic choice of who presides over the councilunder this model presidency is assigned to an authority standing beyond the partisan politics.The identification of such authority is context-dependent.It can be someone akin to the Lord Chancellor in the United Kingdom prior to the 2005 constitutional reform.Another example is the President of the Republic in Italy, who presides over the Italian judicial council. 100lthough in practice the Italian President only rarely attends the meetings of the judicial council, his presidency bestows on the council a high degree of political legitimacy that transcends partisan politics. 101Another option would be a rotating presidency, in which representatives of each branch regularly rotate, or sit in a joint triadic presidency.
Interestingly, the political nature of the inter-branch judicial council may not protect its members against removals in the same way that judge-or politiciancontrolled councils do.An extreme example would be a model allowing the removal of members at will by their principals.A similar example can be found in Slovakia which, to the displeasure of international organisations, attempted to construct a rule which would give the principals (judges and politicians) the discretion to remove their agents at any time.The motivation behind the principle can be tied to electoral terms.If the judicial council members represent their principal in the narrowest sense of the word, the newly elected principal may wish to remove members selected by the previous authority and appoint his or her own agent to the judicial council.In other words, if there is a new president or a new government, it may recall 'presidential' and 'governmental' members of the judicial council and replace them with its own new representatives.This means that the mandates of judicial council members are not only imperative but personally tied to the bodies that appointed them.
The personal incompatibility of judicial council members in the inter-branch type is not presumed.On the contrary, judges are still considered to be a source of expertise, as well as legitimacy.However, politicians are also expected to play a certain expert role and should secure those roles which are political in nature.A typical example would be the negotiation of the budget.Importantly, in contrast to the 'fourth-branch' type introduced in the following section, the criterion of expertise present in the inter-branch judicial council is still overwhelmingly governed by the idea of representation and imperative mandates.This means that while the three branches may select nominees who are knowledgeable and skilled in judicial governance, they still expect them to act in line with the interests of the respective branch.
Like the first judge-controlled model, the inter-branch judicial council also takes over the function of judicial governance.However, combined with the logic of its creation and composition, an inter-branch type of judicial council does not create a new centre of power but merely a platform through which the standard three governmental powers channel their interests and grievances.Like the first model, too, this judicial council still does not have a fully independent budget and depends for that on the Ministry of Justice or Ministry of Finance.
Besides institutional design, the logic of this model may also be strongly reflected in the self-perception of judicial council members, that is their understanding of whom they represent in the body.Although we have argued that the dominant rule is that each branch selects its representatives from among its ranks, the imperative mandate can overrule this standard.What is of the essence is how bound the judicial council's members feel by the principal who selected them.The inter-branch model presumes that the members behave as agents, i.e. understand their roles as representing the interests of the principal who selected them.Again, an example comes from Slovakia, where several judges sitting in the judicial council openly admitted that they would never vote in the council in apparent conflict with the interests and views of the political actor who nominated them.102

A fourth-branch judicial council
The fourth-branch type of judicial council is an autonomous body, independent of any of the other three branches, including the judiciary.Conceptually, this model stands on an equal footing with the three powers.It shares some similarities with the previous three models in its broad contours, but it goes significantly beyond the triadic interplay of the separation of powers in each of its aspects.
The most important feature of the fourth-branch judicial council is that it is not directly controlled by the three classical branches, either jointly (thus differing from the inter-branch model) or separately (thus differing from the judgecontrolled and politician-controlled models), but its members are capable of determining their own institutional will, 103 meaning they set their own agency and form of governance.This core underlying principle is reflected both in the form of control and in the composition of the judicial council.The fourth-branch judicial council can in theory be composed of judges, politicians and non-judges. 104Non-judges can be either members of other legal professions or representatives of civil society, bringing a different type of expertise (human resources, economics, psychology, IT, governance, management, and other executive and leadership skills) inside the judicial council, which might be necessary for the financial or digital dimension of judicial governance.
The lay element is absolutely crucial for a fourth-branch judicial council.First, it is this lay element inside the council's membership which should act as a counterbalance for judges who might be burdened by their own self-interests or conflicts of interest inherently tied to their role in the third branch of state power, 105 as well as for politicians who might want to skew the judiciary towards their partisan preferences or short-term goals.Lay members are not directly associated with interests of any of the three state powers, hence they strengthen the principle of separation of powers and the autonomy of the judicial council from any of the other powers.They represent the purest version of a trustee agent, with agency towards the society.The Venice Commission, for example, suggested that lay members are an important part of the judicial council composition and should be involved in all 'decisions that safeguard against cronyism and cloning among judges'. 106Second, the importance of the lay element of the membership also relates to the unusual breath of the field of judicial governance, which goes way beyond the personal dimension.Non-judges can secure vital expertise in areas of governance related to the regulation of budgets, digital governance (administration of IT clouds of case law, security protocols), and managerial skills. 107Lay members with expertise in human resources, economics and other fields, usually not possessed by the members of traditional three branches, are hence necessary for the council to fulfil its role in line with principles of good governance. 108Third, the diversity of the composition, and a formal requirement to observe this diversity in the selection of the judicial council's members raise additional checks against the political actor who might wish to capture the membership of the judicial council.To put it simply, an actor who might wish to 104 Ex definitio, politicians cannot be non-judges for the purposes of this article. 105See Kosař et al., supra n. 19; also Benvenuti and Paris, supra n. 13; Vauchez, supra n. 66. 106 Venice Commission, supra n. 64, § 20. 107For more on dimensions of judicial governance see Kosař, supra n. 23 and Šipulová et al., supra n. 85. 108 Venice Commission arrived at the same conclusion and explicitly recommends addition of "lay-persons" depending on the breadth of the functions of the judicial councils.See CCJE, supra n. 64, § 19.
The Case for Judicial Councils as Fourth-Branch Institutions politicise the majority inside the judicial council will need to pressure several different professional groups.
From this perspective, non-judges without legal training or education are also essential as checks in the selection and appointment processes.This principle has already been established in the United Kingdom, where the appointment procedure requires that at least two members with no legal qualifications be nominated to the selection commission for Justices of the UK Supreme Court, 109 and at least one non-legally qualified Commissioner in the Board of Commissioners that selects candidates for judicial office in England and Wales, 110 in the Judicial Appointments Board for Scotland 111 and the Northern Ireland Judicial Appointments Commission. 112Non-judges without legal education are required also in the Dutch Council for the Judiciary. 113The presence of lay members inside the judicial council does not mean that they have to necessarily decide on all dimensions of governance; some areas, as the disciplining of judges, may follow a different logic and allow the Judicial Council to sit in a specific formation. 114he significance of this impartial element is further underlined by the judicial council's presidency, which is typically held by the lay members.The presence of lay members impacts on the principles of separation of function (bringing in more expertise) as well as on the separation of persons, as it eliminates the holding of dual roles by the judicial and political authorities.Therefore, a fourth-branch judicial council has both democratic legitimacy and disinterested expert judgement.This in turn helps it to separate both its function and personnel from the rest of the three branches.
113 De Rechtspraak, 'Members of the Council for the Judiciary', https://www.rechtspraak.nl/English/The-Council-for-the-Judiciary/Pages/Members-of-the-Council-for-the-Judiciary.aspx, visited 8 February 2024, see also E. Mak, 'Judicial Self Government in the Netherlands: Demarcating Autonomy', 17 German Law Journal (2018) p. 1801. 114The Venice Commission recognised this principle in relation to the disciplining of public prosecutors, where it actually suggested that the High Council of Justice may sit in smaller formation to secure a higher proportion of prosecutors in matters related to selection and dismissal of prosecutors: CCJE, Opinion on the proposed Amendments to the Constitution of Ukraine regarding the judiciary, No. 803/2015, 26 Oct. 2015, CDL-AD(2015)027, § 18, https://www.venice.coe.int/webforms/documents/default.aspx?pdffile=CDL-AD(2015)027-e, visited 8 February 2024.

108
David Kosař et al.EuConst (2024)     Impartiality is also increased by the fact that non-judges sitting on such a council are selected neither by the political branches nor by the judiciary.They can be chosen by the Bar, professional organisations, the ombudsperson, law schools, other independent agencies or other fourth-branch institutions such as the audit offices, anti-corruption bodies or central banks.Although the traditional three branches may participate in the selection of some judicial council members, those members have a free mandate, meaning that they are not expected to act on behalf of their principals.Instead, members of the fourth-branch judicial council should behave as trustees, and have independent authority to make decisions according to their best judgement and in the public interest. 115A potential mechanism fostering these considerations might be a professionalisation of the role of judicial council members (including e.g.salaries).
Importantly, the members are protected against wilful removal by the principals and their removal is restricted by accountability mechanisms.Independent judicial councils also directly negotiate and administer their own budgets.These rules result in the significant strengthening of the checks independent judicial councils have against possible interferences by the three governmental powers, while preserving their accountability and expertise.
Similarly to the previous three models, we cannot find a pure form of the fourth-branch model in practice.However, some judicial councils already contain characteristics and safeguards allowing for a great autonomy and independence of their members and this trend has recently gained a significant entrenchment both at national and supranational level.A newly established Irish judicial council appropriated, for example, many elements leaning the model close to the fourthbranch ideal, including a focus on the role of lay membership.The Georgian Court Watch issued recommendations to change the composition of the High Council of Justice in such a manner that the majority of members will be nonjudges, and restrict the membership of ex officio chief justice. 116The need to limit the impact of judiciary and position the judicial council beyond any of the three powers is reflected also by latest recommendations of the Venice Commission. 117

J   - 
The four theoretical models of judicial councils we have introduced differ in their individual features as well as in their position within the three branches of governmental power.Each of the models leans in a different direction, resulting in different repercussions for its autonomy, independence and efficiency.
In this section we argue that if a constitutional system decides to entrust the function of judicial administration to a separate institution such as a judicial council, such an institution should be conceptualised as a fourth branch.We begin by showing that both theoretical considerations and practical experience with the first three models of judicial councils demonstrate that they are prone to strengthening one of the branches of power inadequately and, as a consequence, distorting the interaction between the three powers.After that, we provide the positive case for the fourth-branch model.

A critical assessment of the models
We start with the judge-controlled model.The empirical experience we have from many Central and Eastern European countries demonstrates that this model of judicial council suffers from several drawbacks. 118First, while judges have expertise in law and several factets of court administration such as case assignment, they often lack the necessary skills to plan budgets and allocate the finances or human resources training necessary for professionally hiring judges.This is particularly relevant in the career judiciaries that are prevalent in Europe, where judges have often not studied anything other than law at university and have entered the judiciary at a young age without sufficient experience of running a big law firm, acting as a minister or being the dean of a law school. 119econd, by exercising these governance functions, judges sitting on judicial councils blur one of the four components of the principle of the separation of powersnamely the separation of functionsas the main role of judges is to adjudicate and not to administer or govern, which is the domain of the executive.The potential conflicts of interests are even more apparent when court presidents or chief justices sit on or chair judicial councils.
Third, the fact that most members of a judicial council represent judges may hamper the efforts of the judicial members of the council to increase the performance of the judiciary, the quality of justice or judicial accountability, as such reforms might not be viewed positively by their colleagues who elected them 118  110 David Kosař et al.EuConst (2024)     and to whom they will return after the end of their term on the council. 120In other words, corporativism and professional solidarity might prevail over the public interest.At first sight, a harsh evaluation of the judge-controlled model, which loosely corresponds to the early 1990s recommendations of European supranational bodies that shaped many judicial councils introduced in the post-Communist region, may come as a surprise.Nevertheless, even the advisory bodies now recognise the risks of corporativism, 121 low accountability 122 and undue influence from within the judiciary, 123 in models that do not balance the insulation of judges from political branches with sufficient checks.The latest conclusions of Venice Commission openly state that 'unrestrained self-government of judges may result in self-serving judiciary, detached from the society', 124 and the overall leitmotif of the most recent supranational guidelines is to increase the presence of non-judicial members, 125 lay members, civil society or representatives of the political opposition to mitigate the risks associated with judge-controlled judicial councils.
While moving in the right direction, these slight tweaks to the recommended model only partly reduce the risks of corporativism and abusive behaviour of the judicial council vis-à-vis rank-and-file judges.The composition criterion on its own cannot eliminate the dangers present in the model that skew the balance of powers between the branches in such a significant way.Crucially, while current European policy recommendations stress the need for judicial councils to be autonomous, 126 they fail to achieve autonomy because they still subsume the council under the judicial branch, making it dependent on the intentions and the commitment of the judicial branch to the principles of the rule of law. 127his does not work because if judges who control the judge-controlled judicial council decide to pursue problematic goals such as personal gain or nepotism, or just fail to pursue the prescribed goals such as efficiency and quality of justice, there is no check or balance to return them to the right track.In the worst-case scenario, judges who control the judge-controlled judicial council may even have an interest in frustrating the main norm (judicial independence) the judicial council was meant to guarantee.As a result, in some countries judge-controlled judicial councils have been captured by judicial oligarchs who rewarded their allies and suppressed any criticism. 128Again, under the judge-controlled judicial council model, it is very difficult to hold such abusive judicial leadership to account.
In theory, the risk of capture could be at least partially resolved by a rule that would grant several seats on the judicial council to the opposition.For a judicial branch model, this would mean that judicial candidates must represent several interest groups within the judiciary.For example, instead of a single electoral district, the selection of judicial members would be structured according to regions, court level or membership of judicial associations (or a combination of these criteria)whichever stratification best represents different voices129 within the given judiciary.But even if this diffusion of power among different factions within the judiciary can be achieved, it will not address most of the abovementioned drawbacks of the judge-controlled model, namely the lack of expertise, corporativism and the blurring of the separation of functions component of the principle of the separation of powers.
The issue with a judicial council leaning towards one branch of power is even more obvious in the politician-controlled model.Politician-controlled judicial councils do not bring many benefits beyond the increased political accountability.They suffer from most of the drawbacks identified in the judge-controlled model, as the majority of seats on the politician-controlled council are still held by judges.But in addition they face a major risk of politicisation, which might be too tempting to resist.As a result, the politician-controlled judicial council will necessarily succumb to one or more of the following four modes of politicisation.
First, the selection of members of the judicial council, including its judicial members, vested dominantly in political branches necessarily opens channels of informal pressures on and politicisation of the council.This in turn reduces the external judicial independence (i.e.independence vis-à-vis third actors, in this case politicians) as well as the perceived independence of the council.It particularly holds for those models where other features of judicial council design do not support the individual independence of council members (such as tenure, irremovability or high salaries), but instead tie the mandates of members to a single political principal, in most cases the government.This is the scenario that unfolded in Poland after the 2017 revamping of the National Council of the Judiciary, where the new legislation forced the premature termination of the judge-members' term of office and vested the election of new members fully in the hands of the Polish Parliament. 130The new model was introduced as an attempt to increase democratic legitimacy and public participation in the election of judicial council members, while in fact it allowed a single political party that dominated the government to capture the council. 131Kaczyński's government indeed swiftly used the newly composed council to pack the new Disciplinary Chamber introduced at the Supreme Court, and to achieve complete control over processes of judicial accountability. 132he Hungarian redesign of judicial governance demonstrates the risks associated with a politician-controlled judicial council perhaps even better.After the 2011 constitutional overhaul, Orbán vested the power to select, promote and demote all judges in a newly created National Judicial Office.All personal competences related to the careers of judges have been newly concentrated in the hands of the head of the Office.At first sight, the system appeared to offer some safeguards, as the head was to be elected by a two-thirds majority in the Parliament and could select new judges solely on the basis of a list prepared by local councils of judges.Yet, Orbán's supermajority after the 2010 elections allowed him to control the election of the chairmanship and staff the National Judicial Office with people close to the Fidesz party. 133econd, politicians who control the council may also exercise informal pressure on 'their' judicial members if these members do not pursue their policy and may eventually force them to resign from the council if they do not follow their wishes.
Third, the selection of judicial members by politicians, and the accompanying lack of judicial members elected by judges, divides and politicises not only the council but also the judiciary.
Fourth, this model contributes to politicisation, because it confuses the professional role conception of judges and their individual independence with the expectation that members of the judicial council will act as the agents of the political branches of power.Empirical research has suggested that judges selected by political branches suffer from split role-conceptions. 134In other words, while 132 See also the judgments of the ECJ, which deemed political control over the Judicial Council to be one of the elements reducing the independence of newly appointed judges, and the Disciplinary Chamber of the Supreme Court in particular: A.K. and Others v Sąd Najwyższy, supra n. 64; Commission v Poland, supra n. 97. 133The Parliament elected Tunde Hando, the wife of a prominent party lawmaker.The Case for Judicial Councils as Fourth-Branch Institutions some continue to act independently, others understand their mandate in the judicial council as inevitably tied to the preferences of their political principals.Even more troublesome evidence shows that in politically polarised societies such as in Poland, the judicial members of the politician-controlled judicial council serve the politicians who elected them.
In fact, even systems that vest the control over the composition of a judicial council in political branches but require more proportional engagement of coalition and opposition forces might still lead to ineffective judicial governance.The first reason is that politicians might not be able to agree on the selection of the council members, which results in gridlock and paralysis of the judicial council. 135he second reason is that the fact that judges do not feel that they are represented136 may also hamper the implementation of any reform adopted by the politician-controlled judicial council.
This brings us to the inter-branch model of judicial council.In contrast to the first two models, this one affects the separation of powers to a lesser degree.It is designed as a platform for negotiation for the three governmental branches and, as such, should be ideally balanced.Nevertheless, the combination of imperative mandates, the possibility of their principals removing the members at will and the lack of a lay element, which would convey the expertise unburdened by political partiality, implants too many risks into the coordinating model.
First, this model is potentially viable only in presidential systems, as in most parliamentary systems the legislature and executive are usually controlled by the same political force.This brings with it the risk that the coordinating model will still be dominated by a single political force.Thus, all the pitfalls of politiciancontrolled judicial councils apply here as well and the risk of the politicisation of judicial governance is high.A potential solution for parliamentary systems might be a requirement of multipartisan formula, i.e. either the selection of judicial council members by (major) political parties including parties in the opposition, or joint voting between the coalition and opposition political parties.While such a condition makes the model and its representativeness more viable, the experience with similar multipartisan practices (judicial council in Spain, selection of constitutional court justices in Germany or Belgium) comes with further drawbacks and depends too much on the level of political and societal homogeneity.To put it simply, such a model expects a high level of coordinationeither between the three branches of power, or between individual political partiesand is prone to internal paralysis in highly fragmented societies (as we discuss below).
Second, this model is highly unstable, as each new government can easily dominate the judicial council and get rid of the members elected by the previous political group either immediately (imperative mandates) or at the end of their terms and then replace them with their own nominees.Even if the model includes an opposition criterion (requiring the representation of members of the opposition parties or the involvement of both parliamentary chambers in the selection of members by the legislative power), the government's dominance would be only reduced, not completely eliminated.This was very much visible after the 2023 return of the Slovak populist government, led by Robert Fico, to power.It was mere days after the election when the government removed the members of the judicial council appointed by the previous executive and selected its own appointees, imposing a significant chilling effect on the behaviour of the whole council.There are, of course, ways to mitigate the government's effect (such as creating a single collegial electorate).Such measures, however, already change the logic of members' agency and lean the judicial council closer to the fourthbranch model.
Third, this model deeply polarises the judiciary, as it often pitches judicial members of the council elected by judges against such members appointed by the political branches, particularly if the political preferences of the judicial and political branches differ on how to organise and what to achieve in judicial governance.While the public typically does not pay too much attention to judicial councils,137 the political conflicts inside the judicial councilwhose aim is to safeguard the independence of the judiciarymight negatively impact the general trust in courts.
Fourth, the complex rules of composition and strong control of mandates might make the model fragile in practice.On one hand, the model stands too close to politics.If the interests of the majority of principals (either both political branches or the political and judicial branches) deviate from the principles of the rule of law, the model will inevitably fail to secure an independent judiciary as it lacks any autonomy in acting.On the other hand, if the interests of the principals, the three branches, are too diverse, the model might be paralysed by the inability of the judicial council's members to reach a compromise.This brings us to the last model, the fourth-branch model.This model conceptualises the judicial council as an autonomous body, independent of any of the three branches, including the judiciary.At first glance, judicial councils might be considered an odd addition to the archipelago of fourth-branch institutions.Indeed, compared to the usual considerations related to the fourth-branch institutions which require a combination of particular expertise and certain political sensitivity, judicial councils are a specific case.Unlike other fourth-branch institutions, judicial councils regulate and govern one of the governmental powers.Therefore, they may, depending on their design, significantly shift the balance of power between the three branches.That said, we argue that this specificity of judicial councils does not weaken the case for conceptualising them as fourth-branch institutions.On the contrary, it is this specificity of judicial councils that makes the rationale for conceptualising judicial councils as fourthbranch institutions particularly strong.
The core of our argument stems from the nature of the function that judicial councils are supposed to assumejudicial governance.This function fulfils all three conditions of functions which, according to the fourth-branch theory, might be better discharged by an independent fourth-branch institution. 138First, the function of judicial administration lies in the protection or realisation of a constitutional norm not negotiable through the regular political process.The norm here lies in securing an independent, yet efficient and democratically accountable, judiciary.Such a norm corresponds to what Bruce Ackerman labels as fundamental governmental value, 139 Mark Tushnet as protection of democracy, 140 and Tarun Khaitan as respect for a constitutional norm 141 that should not be negotiable through a regular political process.The independence of courts should not depend on who won the latest elections.Second, securing an independent and accountable judiciary might not necessarily always be in the short-term political interest of the ruling majority.After all, it was exactly this fear of political interference with the judiciary that formed one of the initial motivations behind the establishment of judicial councils as an alternative model to the previous ministerial model of judicial governance. 142And, third, the realisation of the function at hand requires political judgement and a greater degree of democratic legitimacy.Key questions of judicial governance, such as who should become a judge or a court president and where to strike a balance between the quality and speed of judicial decision-making, raise issues with no right answer that necessarily require political judgement.

Towards the fourth-branch model of judicial council
Based on the analysis of individual models and their drawbacks, we argue that judicial councils are especially strong candidates for being conceptualised as fourth-branch institutions.If a constitutional system decides to entrust the function of judicial administration to a separate intuition such as a judicial council, such institution should be conceptualised as a fourth branch.In our view, such conceptualisation brings two main benefits.
First, the conceptualisation of a judicial council as a fourth branch ensures that judicial governance is exercised at the sufficient distance from politics.On the one hand, free mandates, open selection based on expertise and protection from wilful dismissal by principals, as well as certain budgetary competences, guarantee a degree of separation of the judicial council from day-to-day politics.In doing so, it distances judicial governance from political polarisation and the unbalanced influence of the political branches of government.On the other hand, the fourthbranch model does not entirely isolate the judicial council from the source of democratic legitimacy.For one thing, political actors, ideally both from the governing coalition and the opposition, can still participate in the creation of the body.Moreover, the fourth-branch model of judicial council also inherently includes members from civil society who bring an element of impartiality thanks to their free mandate and lack of dependence on any of the governmental branches.As a result, the fourth-branch model is not an entirely apolitical body, but one with a degree of democratic (but non-partisan) legitimacy of its own necessary for making political (yet non-partisan) decisions.
The second benefit of the fourth-branch model is that it shields the judicial council from the second major risk it facescorporativism.One of the biggest lessons which our analysis demonstrates regarding the existing scholarship on judicial councils is that the independence of judicial councils needs to be institutionally secured both from the political branches of power as well as from the judiciary itself.Judicial councils, conceptualised as a part of the judiciary, face the danger of functioning in the interests of a few judges (judicial leadership), instead of in the general interest.By not allowing judges to dominate the body, by including members from civil society and by installing independent mandates, the fourth-branch model severs the strong link between the judicial branch and the council, and thus removes the conflict of interest inherently present in the judgecontrolled model.At the same time, it still allows for a degree of expertise to be accumulated within the council, as it may still include some judges and it also includes experts from civil society, such as (non-judge) lawyers, specialists in human resources and economists.As a result, while being severed from the immediate, narrow interest of the judiciary, the fourth-branch model still provides the judicial council with sufficient expertise to make informed decisions.
All in all, we argue that conceptualising and constructing the judicial council as a fourth branch offers the best combination of political independence, expertise, and political accountability from all models.Compared to the first three models, through its composition, free mandates, and checks and balances in selection and removal processes, the fourth-branch model embeds most safeguards against the The Case for Judicial Councils as Fourth-Branch Institutions danger that a judicial council will eventually lean too far towards one of the three governmental powers.It provides the best protection against the Scylla of politicisation and the Charybdis of corporativism.As a result, this model has the greatest chance of securing judicial administration that will be conducted in a non-partisan, but still politically sensitive manner and with the necessary expert knowledge.Thus, the fourth-branch model erodes the separation of powers principle in the least (harmful) way.

C
Judicial councils may be a good fit for being considered fourth-branch institutions, both theoretically and empirically.They meet theoretical expectations of the fourth-branch institutions as they exercise a function which cannot be executed well by any of the classical branches and which requires a specific mix of independence, accountability and expertise.Moreover, their reconceptualisation as fourth-branch institutions makes sense not only from the perspective of the separation of powers theory, but also on the basis of empirical experience with judicial councils around the world.Judge-controlled judicial councils often suffer from corporativism and fail to ensure internal judicial independence.Politiciancontrolled councils are prone to politicisation and political capture.Finally, the inter-branch model is inherently unstable and leans either to judge-controlled or politician-controlled models with their abovementioned pitfalls.
The conceptualisation of judicial councils as fourth-branch institutions would bring several benefits and allow them to dodge the two main dangerspoliticisation and corporativismthat are associated with judicial councils and that typically deform the separation of powers.None of the three competing models properly addresses these dangers.The conceptualisation of judicial councils as fourth-branch institutions requires them to have certain design features.We have therefore sketched a model of how such an institution would ideally be designed, guided by the core idea of the fourth-branch institutionsthat it possesses the required combination of independence, accountability and expertise.
The shift towards the fourth-branch model must be not only institutional, but also mental, which might be the most difficult part of accepting our reconceptualisation of judicial councils, 143 because it requires a new understanding of and thinking about judicial councils as autonomous institutions which have their own legitimacy, not one derived from any of the classical trinity of branches of state power.The most recent guidelines of European supranational bodies are already heading in this direction and stress the importance of non-judges being on judicial councils.144But these guidelines have gone only half-way and have not yet internalised the idea that the real diversification of judicial council membership that secures the resistance to politicisation and corporativism can best be achieved only if we accept the fourth-branch understanding of a judicial council.We provide a theoretical underpinning for a fully-fledged shift that is needed in many corners of Europe where judicial councils underperform.

8
See B. Iancu and E. Tănăsescu, 'Introduction', in Iancu and Tănăsescu, supra n. 5, p. 2 (pointing out that judicial councils are especially impactful on the structure of the separation of powers).

20M.
Bobek and D. Kosař, 'Global Solutions, Local Damages: A Critical Study in Judicial Councils in Central and Eastern Europe', 15 German Law Journal (2014) p. 1257; P. Castillo-Ortiz, Judicial Governance and Democracy in Europe (Springer 2023).

49
For a succinct summary of this development see N. Picardi, 'La Ministère de la Justice et les autres modèles d'administration de la justice en Europe', in D.A. Giuffrè (ed.), L'indipendenza della giustizia, oggi.Liber amicorum in onore di Giovanni E. Longo, Milano (Giuffrè Francis Lefebvre 1999) p. 269.For a more historical account see M. Shapiro, Courts: A Comparative and Political Analysis (University of Chicago Press 1981) p. 126 or J.P. Dawson, The Oracles of the Law (University of Michigan Press 1968). 50

53
Garoupa and Ginsburg, supra n. 1.We leave aside other theories of judicial councils that have little bearing on Europe (see Brett, supra n. 22; N. Garoupa, 'Revisiting the Theory of Judicial Councils', in S. Turenne and M. Moussa (eds.),Research Handbook of Judging and the Judiciary (Edward Elgar forthcoming)).54 C. Guarnieri, 'Appointment and Career of Judges in Continental Europe: The Rise of Judicial Self-government', 24 Legal Studies (2004) p. 169; Garoupa and Ginsburg, supra n. 1; N. Gy, 'A Discipline of Judicial Governance?', 7 Utrecht Law Review (2011) p. 102; T. Bunjevac, 'Court Governance: The Challenge of Change', 20 Journal of Judicial Administration (2011) p. 201; J. Vapnek, '21 Cost-Saving Measures for the Judiciary', 5 International Journal for Court Administration (2013) p. 55. 55 See Hammergren, supra n. 24 and R.B. Chavez, 'The Appointment and Removal Process for Judges in Argentina: The Role of Judicial Councils and Impeachment Juries in Promoting Judicial Independence', 49 Latin American Politics and Society (2007) p. 33.
Alter, 'Agents or Trustees?International Courts in their Political Context', 14 European Journal of International Relations (2008) p. 33; A. Dyevre, 'Technocracy and Distrust: Revisiting the Rationale for Constitutional Review', 30 International Journal of Constitutional Law (2015) p. 30; A.P. Mediano, 'Agencies' Formal Independence and Credible Commitment in the Latin American Regulatory State: A Comparative Analysis of 8 Countries and 13 Sectors', 14 Regulation & Governance (2018) p. 120.

90A
typical example would be the position of court presidents in post-Communist judiciaries: see A. Blisa and D. Kosař, 'Court Presidents: The Missing Piece in the Puzzle of Judicial Governance', 19 German Law Journal (2018) p. 2031; D. Kosař and S. Spáč, 'Post-Communist Chief Justices in Slovakia: From Transmission Belts to Semi-Autonomous Actors?', 13 Hague Journal on the Rule Law (2021) p. 107.Generally speaking, the more powerful the court presidents, the more strongly they push for participation in the judicial council in order not to be left out of the loop.91 For more details regarding these dimensions see Šipulová et al., supra n. 85. 92 See Grzęda v Poland, supra n. 7.For more details see M. Leloup and D. Kosař, 'Sometimes Even Easy Rule of Law Cases Make Bad Law ECtHR (GC) 15 March 2022, No. 43572/18, Grzęda v Poland', 18 EuConst (2022) p. 753.93 Kosař and Spáč, supra n. 90; S. Spáč, 'The Illusion of Merit-based Judicial Selection in Post-Communist Judiciary: Evidence from Slovakia', 69 Problems of Post-Communism (2022) p. 528.
130 M. Szwed, 'Fixing the Problem of Unlawfully Appointed Judges in Poland in the Light of the ECHR', 15 Hague Journal on the Rule of Law (2023) p. 353. 131P. Filipek, 'The New National Council of the Judiciary and its Impact on the Supreme Court in the Light of the Principle of Judicial Independence', 16 Problemy Współczesnego Prawa Międzynarodowego, Europejskiego i Porównawczego (2018) p. 177.
Constitutionalism and the Separation of Powers, 2 nd edn (Liberty Fund 1998) p. 13; E. Carolan, The New Separation of Powers.A Theory for the Modern State (Oxford University Press 2009); C. Möllers, The Three Branches: A Comparative Model of Separation of Powers (Oxford University Press 2013); J. Waldron, Political Political Theory (Harvard University Press 2016).Halberstam, 'The Promise of Comparative Administrative Law: A Constitutional Perspective on Independent Agencies', in S. Rose-Ackerman and P.L. Lindseth (eds.),Comparative Administrative Law (Edward Elgar 2010) p. 139; B. Ackerman, 'The New Separation of Powers', 113 Harvard Law Review (2000) p. 633; Möllers, supra n. 26, at p. 41 and 49; A. Kavanaugh, The Collaborative Constitution (Cambridge University Press 2023); Waldron, supra n. 26, at p. 45, 52 and 62-65; A. Sajó and R. Uitz, The Constitution of Freedom: An Introduction to Legal Constitutionalism (Oxford University Press 2017).
Several states which had earlier happily accepted judicial self-governance bodies decided 65 See e.g.Italian High Council of the Judiciary presided over by the President of the Republic, Supreme Court Chief Justice who was ex officio the president of the National Council of the Slovak Republic until 2014, general prosecutor in Estonia, minister of justice in France.Vauchez, 'The Strange Non-Death of Statism: Tracing the Ever Protracted Rise of Judicial Self-Government in France', 19 German Law Journal (2018) p. 1613.We borrow the pushback-backlash-withdrawal analytical toolbox from M.R. Madsen et al., 'Backlash against International Courts: Explaining the Forms and Patterns of Resistance to International Courts', 14 International Journal of Law in Context(2018) p. 197, to address this reverse trend, but we apply it very loosely and adjust it to judicial councils.Hirschman's famous exitvoice-loyalty triad could be applied here too.

71
See e.g. the Polish response (UNHRC, A/HRC/38/38/Add2 Report of the Special Rapporteur on the independence of judges and lawyers on his mission to Poland: Comments by the State, 6 June 2018, https://documents.un.org/doc/undoc/gen/g18/170/42/pdf/g1817042.pdf?token= tX6kgqREEEW5FnNIXQ&fe=true, visited 8 February, p. 4) to the Report of the Special Rapporteur on the independence of judges and lawyers on his mission to Poland (UNHRC, A/HRC/38/38/Add1 Report of the Special Rapporteur on the independence of judges and lawyers on his mission to Poland, 5 April 2018, https://digitallibrary.un.org/record/1484929?ln=en, visited 8 February 2024).

Table 1 .
Four ideal types of judicial councils under the separation of powers theory The Methodology of the Social Sciences (Free Press 1949) p. 90.