171 46 3MB
English Pages [526] Year 2013
Acknowledgements This book originated in an international conference based on the academic cooperation of three law schools – Bar Ilan University Faculty of Law, the Buchmann Faculty of Law at Tel Aviv University, and the Radzyner School of Law at the Interdisciplinary Center (IDC). We are grateful for their support. We are particularly grateful to our students who helped us in the editing process – Avi Vaknin, Alon Dvir, Alon Nahear, Dudi Rabinowits and Neta Shapira. Roman Zinigrad acted as our deputy editor and we express our deep appreciation to his assistance throughout the whole process of editing. Gideon Sapir, Daphne Barak-Erez and Aharon Barak February 2013
1 Introduction: Israeli Constitutional Law at the Crossroads GIDEON SAPIR, DAPHNE BARAK-EREZ AND AHARON BARAK
I
SRAELI CONSTITUTIONAL LAW is a sphere of many contradictions and traditions. Growing from the tradition of British law which was absorbed by the legal system of Mandatory Palestine, Israeli constitutional law had followed the path of unwritten constitutional principles. At the same time, inspired by the new arena of postWorld War constitutionalism, as well as by Resolution 194 of the General Assembly of the United Nations recognizing the establishment of the new State and the wording of its Declaration of Independence, the newly established State planned to adopt a Constitution. However, at the time of writing, this vision has not been finalized due to inner controversies on the content of the Constitution. In the meantime, the vision has been converted to the enactment of a series of Basic Laws, which function as Israel’s de facto Constitution. The constitutional outcomes of this special history are the subject of many controversies, to be explored in this book. On 15 May 1948, close to the termination of the British Mandate in Palestine, the members of the People’s Council representing the Jewish community assembled and proclaimed the establishment of the State of Israel. The Declaration of Independence promised, inter alia, a Constitution for Israel: We declare that, with effect from the moment of the termination of the Mandate being tonight, the eve of Sabbath, the 6th Iyar, 5708 (15 May, 1948), until the establishment of the elected, regular authorities of the State in accordance with the Constitution which shall be adopted by the Elected Constituent Assembly not later than the 1st October 1948, the People’s Council shall act as a Provisional Council of State, and its executive organ, the People’s Administration, shall be the Provisional Government of the Jewish State, to be called ‘Israel’.1
The Declaration presented a course for the process of enacting a Constitution comprising three stages. In the first stage, the ‘Provisional Council of State’ was to act as a temporary legislative branch and in the second stage, elections were to be held for a ‘Constituent Assembly’ charged with drafting a Constitution. After accomplishing this task, the Constituent Assembly was supposed to disperse. In the third stage, elections were supposed to be held for a legislative authority according to the electoral system to be determined in the Constitution. With the election of a legislature, the Provisional Council of State would conclude its task and disperse as well. Declaration of the Establishment of the State of Israel, 1 LSI 7 (1948).
1
2 Gideon Sapir, Daphne Barak-Erez and Aharon Barak In reality, things had taken a different course. First, the Provisional Council of State decided it would cease to exist with the convention of the Constituent Assembly, and its powers would be transferred to the Assembly.2 Thus, the Constituent Assembly was assigned the role of the ordinary legislator, alongside its original role as a body charged with the drafting of a Constitution.3 Second, the Constituent Assembly, already known by its new name – the Knesset – did not succeed in getting to a consensual constitutional text. From May 1949 until June 1950, the Knesset was the scene of stormy debates.4 Disagreements persisted, and the inability to reach consensus on the contents, the form, or even the need for a Constitution, finally led to the adoption of the compromise formula, proposed by Knesset Member Yizhar Harari. This compromise took the shape of a decision to adopt a piecemeal Constitution by enacting a series of Basic Laws, stating the following: The First Knesset charges the Constitution, Law, and Justice Committee with the task of preparing a constitution for the country. The constitution will be built chapter by chapter, so that each one will in itself be a basic law. The chapters will be submitted to the Knesset as the Committee concludes its task and, together, all these chapters will become the constitution of the country.5
In fact, for many years, the ‘Harari decision’ was implemented only partly. Although the Knesset enacted a series of Basic Laws over the years, they addressed only the institutional aspects of Israeli constitutional law and focused on the ‘rules of the political game’, and not the arena of values and basic rights. In the absence of a formal constitutional Bill of Rights, the Israeli Supreme Court created alternative mechanisms for the protection of human rights, by reference to unwritten principles and values, as well as by applying the doctrines of administrative law to limit executive power.6 Judicial activism, however, had its limits. The Supreme Court perceived itself as generally precluded from reviewing Knesset legislation (in contrast to administrative actions), following the British tradition of parliamentary sovereignty. Therefore, the lack of a formal Bill of Rights was still felt and relevant. For decades, however, several attempts to anchor a Bill of Rights in the form of a Basic Law proved unsuccessful,7 and no breakthrough seemed in sight. A significant change followed the enactment of two new Basic Laws dealing with human rights – Basic Law: Freedom of Occupation and Basic Law: Human Dignity and Liberty (hereinafter: the new Basic Laws).8 As its name attests, Basic Law: Freedom of Occupation deals with one right only – freedom of occupation and trade. Its counterpart – Basic Law: Human Dignity and Liberty – applies to several rights – property, movement from and to Transition Law, 5709-1949, s 1. This decision evoked misgivings on both the right and the left of the political spectrum. See R Gavison, ‘The Controversy Over Israel’s Bill of Rights’ (1985) 15 Israeli Year Book of Human Rights 113, 117. 4 See J Segev, ‘Who Needs a Constitution? In Defense of the Non-decision Constitution-making Tactic in Israel’ (2007) 70 Alberta Law Review 409, s IIB; Gavison (n 3) 147–50. 5 DK 5 (1950) 1743 (in Hebrew). 6 See, B Bracha, The Protection of Human Rights in Israel (1982) 12 Israeli Yearbook of Human Rights 110; A Maoz, Defending Civil Liberties Without a Constitution: The Israeli Experience (1988) 16 Melbourne University Law Review 815. 7 Draft Bill Basic Law: Human and Civil Rights, 1973, HH 448; Draft Bill Basic Law: Bill of Rights, 1983, HH 111. 8 For an English translation of Israel’s Basic Laws see: knesset.gov.il/description/eng/eng_mimshal_yesod1. htm. 2 3
Introduction 3
Israel, liberty, dignity, and privacy (but does not mention specifically several other import ant human rights, such as equality, freedom of expression, and freedom of religion). Soon after the passing of the new Basic Laws, the Supreme Court stated that their enactment created a ‘constitutional revolution’, in the sense that they granted the Court the authority to review primary Knesset legislation.9 Since then, the Israeli constitutional arena has changed significantly. The Supreme Court has used the new Basic Laws as a platform for developing Israel’s constitutional law, by interpreting them broadly to include those rights not specifically mentioned in the constitutional text. It stated that many of these unenumerated rights were protected under the broad ‘umbrella’ of the right to human dignity.10 This judicial innovation was accompanied also by a rich scholarly discourse. Debates and controversies soon followed.11 Unfortunately, due to the language barrier, only a small portion of that judicial and academic activity is known to the international constitutional law community. This book attempts to make a modest contribution to overcoming the language barrier, and help expose the Israeli scene to the English speaking scholarly community. The book offers a comprehensive study of Israeli constitutional law in a systematic manner that moves from constitution-making to specific areas of contestation. The book features contributions by scholars of Israeli constitutional law, followed by comments by leading scholars of comparative constitutional law from Europe and the United States. In fact, it aspires not only to present Israeli constitutional law, but rather to present the controversies that shape it in a manner that sheds light on theoretical questions, as well as on the experience of other systems. Part 1 of the book presents the reasons and justifications offered for promoting the Israeli constitutional project and adopting a full formal Constitution. This is a question still considered open in Israel, a country that has not yet completed the process of enacting its Basic Laws. Gideon Sapir discusses the various functions a Constitution can fulfill and evaluates their relevance to the Israeli context. He argues that Israel needs a Constitution that will serve as a gag rule which bars constant controversies over certain issues. Alon Harel bases the need to complete the Israeli constitutional project on the importance of judicial review which grants a hearing to aggrieved individuals. Ariel L Bendor discusses the importance of having a constitutional regime by reference to its contribution to defining the basic principles of the legal system. Part 1 concludes with a comment by Sanford Levinson. Part 2 discusses the various institutional models for judicial review that may suit the needs of the Israeli constitutional arena. Currently, Israel follows the US model that recognizes the power of every court to review the constitutionality of legislation, but this has been a contested model in the Israeli public arena. Yoav Dotan argues that the scope of judicial review should be determined in correlation to the system for selecting judges, and therefore Israel should not follow the US model of full-fledged judicial review, as CA 6821/93 United Mizrahi Bank Ltd v Midgal Cooperative Village 49(4) PD 221 [1995] (in Hebrew). HCJ 6298/07 Ressler v Knesset (21 February 2012) (in Hebrew); HCJ 7052/03 Adalah v Minister of Interior 61(2) PD 202 [2006] (in Hebrew); HCJ 10203/03 Hamifkad Haleumi v Attorney General 62(4) PD 715 [2008] (in Hebrew). 11 See, eg D Barak-Erez, ‘From an Unwritten to a Written Constitution: The Israeli Challenge in American Perspective’ (1995) 26 Columbia Human Rights Law Review 309; G Sapir, ‘Constitutional Revolutions – Israel as a Case Study’ (2010) 5 International Journal of Law in Context 358; B Medina, ‘A Response to Richard Posner’s Criticism of Aharon Barak’s Judicial Activism’ (2007) 49 Harvard International Law Journal 1. 9
10
4 Gideon Sapir, Daphne Barak-Erez and Aharon Barak long as it follows a professional based process of selecting judges. Joshua Segev evaluates the prevalent view of the Israeli system and argues for a reform that will empower only the Supreme Court to practice judicial review of legislation. Ori Aronson criticizes this alternative, which seems to gain growing popularity. He argues that this model overlooks the redeeming potential of trial court adjudication for a deliberative, participatory, and pluralistic process of creating constitutional norms and understandings. Tsvi Kahana evaluates the Israeli experience with the notwithstanding mechanism which was incorporated in 1994 into Basic Law: Freedom of Occupation, as inspired by the Canadian Charter of Rights and Freedoms. Kahana argues that as long as Israel’s Basic Laws can be amended without a super majority, deviations from them should be made by constitutional amendments, rather than through a notwithstanding mechanism. Part 2 concludes with a comment by Víctor Ferreres Comella. Part 3 assesses the connections between Israeli constitutional law and global processes. Iddo Porat analyzes the persistent influence of foreign law on Israeli constitutional judicial decisions. Following a similar route, Moshe Cohen-Elliya discusses the influence of Western liberal thinking on Israeli constitutional law in the format titled by him as ‘transformative constitutionalism’. Margit Cohn offers a detailed analysis of the constitutional doctrine of proportionality in a comparative perspective. Part 3 concludes with a comment by Vicki C Jackson. Part 4 analyzes the process of constitutional balancing. Mordechai Kremnitzer offers a critical evaluation of the dangers ingrained in balancing for the protection of human rights. Yaacov Ben-Semesh offers an evaluation of the case study of balancing in the area of freedom of speech, with regard to speech allegedly offending the feelings of others. Part 4 concludes with a comment by Sujit Choudhry. Part 5 discusses the status of the so-called unenumerated rights in Israeli constitutional law, focusing on the role played by the benevolent interpretation offered by the Israeli Supreme Court to the concept of human dignity, which enables it to fill the void created by the failure of the current Basic Laws to specifically mention rights such as: equality, freedom of speech and freedom of religion. Tamar Hostovsky-Brandes offers a broad overview of current judicial interpretations of the right to human dignity. Sharon Weintal discusses the possibility of compensating for the supposedly missing rights in the Basic Laws by reference to non-written constitutional principles. Part 5 concludes with a comment by David Fontana. Part 6 focuses on the challenge of social rights in the context of Israeli constitutional law. The constitutional status of social rights and their enforceability is an open question in many systems, but even more so in Israel which lacks an express recognition of them. The result has been a limited protection of these rights as derivatives of the concept of human dignity, usually offering them only a ‘minimum’ level of protection. Both Aeyal Gross and Amir Paz-Fuchs discuss this model of mimimum-level protection of social rights in a critical manner. Neta Ziv offers another critical view, by focusing on the dynamic of conditioning social rights entitlements. Part 6 concludes with a comment by Mark Tushnet. Part 7 moves forward to assess the applicability of constitutional rights in private law. Aharon Barak discusses the Israeli approach to this question – defined as an ‘indirect application model’ – and compares it to other prevalent models. Michal Tamir discusses this question in a contextualized manner which focuses on contractual relations. Part 7 concludes with a comment by Stephen Gardbaum.
Introduction 5
Part 8 is dedicated to the issue of emergency constitutional powers and other national security related constitutional matters. This topic is of the highest importance in the Israeli context, taking into consideration the threats to Israel since its establishment. Daphne Barak-Erez analyzes the Israeli model of regulating the power to declare an emergency situtation in a comparative perspective. Barak Medina discusses the role of legislation in regulating national security threats. Part 8 concludes with a comment by Adam Tomkins. Part 9 evaluates various outcomes of Israel’s particular raison d’etre as a State defined in its Basic Laws as ‘Jewish and Democratic’. This constitutional formula has ramifications to both the nation state model and to the regulation of State and religion matters. Chaim Gans discusses the Israeli nation state model in the context of Zionist thinking, examining the various ideological streams within the Jewish national movement. Aviad Bakshi and Gideon Sapir evaluate the implications of Israel’s identity as the nation state of the Jewish people for its immigration policy, by analayzing the case study of family reunification applications of Palestinians. Gila Stopler explores the connection between national identity and the regulation of religion–State relations in Israel. She claims that the Israeli model has been only partially successful in shaping national identity and relates this limited success to the State’s lack of control over its own religious establishment. Part 9 concludes with a comment by Susanna Mancini and Michel Rosenfeld. * * * Read together, the various chapters and comments included in this book present Israeli constitutional law as a living sphere, which reflects the dilemmas the country is faced with, as well as the challenges of constitutional theory in general. As such, our hope is that the book will promote not only the future study and development of Israeli constitutional law, but also the understanding of the complexities of constitutional systems that are still coping with the challenge of nation-building and transitions.
2 Why a Constitution – in General and in Particular in the Israeli Context? GIDEON SAPIR
I. INTRODUCTION
A
CCORDING TO A common rationale, a Constitution is required to defend basic values. The constitutional model necessary for that purpose comprises three components: supremacy, entrenchment and judicial review. This chapter, however, considers two more constitutional rationales that are completely different from the one just described. According to one alternative view, the function of the Constitution is to create public dialogue over important issues, and according to the other a Constitution can serve as a silencing mechanism, or as a ‘gag rule’. This chapter is divided into three parts: in part II, I present the three rationales for adopting a constitutional strategy and three different constitutional models that derive from them, respectively; in part III, I examine which of the models is appropriate for Israel and in part IV, I examine which of the models has been chosen thus far in Israel. II. THREE RATIONALES AND THREE MODELS
A. Protecting Constitution According to a common rationale, a Constitution is required to defend basic values. This rationale is based on two main premises: that certain values are of supreme importance and that those values are under threat of attack. When sober, we know what is permitted and what is prohibited, but we also know that there may be situations – under conditions of inebriation – in which control over clear reasoning is lost. Hence we strive, while sober, to make it difficult for ourselves, when in a state of loss of control, to perform acts we will later rue.1 The constitutional model necessary for that purpose comprises three compon ents: granting supreme status to the document that anchors the values (supremacy), safe guarding it from ordinary legislative amendment (entrenchment), and granting authority 1 The technical term used in this context is precommitment, and a common metaphor used in this context is Ulysses and the Sirens. J Elster, Ulysses and the Sirens (Cambridge, Cambridge University Press, 1979) ch 2; S Freeman, ‘Constitutional Democracy and the Legitimacy of Judicial Review’ (1990) 9 Law and Philosophy 327; M Klarman, ‘Majoritarian Judicial Review: The Entrenchment Problem’ (1997) 85 Georgia Law Journal 491, 496.
10 Gideon Sapir to the courts to invalidate primary legislation that contradicts this document (judicial review). This rationale and its accompanying model are the best known of the three that will be discussed in this chapter, but, as is known, they suffer from a number of weaknesses that are the subject of an extensive body of literature. We shall note some of them in brief: even if we agree that there are values of supreme importance, in a pluralistic soci ety a consensus does not exist on the identity of these values. If a Constitution should be accepted by consensus and not forced on a segment of the public, it is not clear how it is possible to overcome the lack of agreement and to adopt a Constitution. Moreover, even if we were to succeed in solidifying an agreed upon list of supreme values, there would still be a significant dispute over how to apply them in concrete circumstances. This model authorises the court to interpret the Constitution and to enforce it, thereby grant ing the court the power to decide over many constitutional dilemmas, an approach that does not coincide with the democratic principle.2 Furthermore, owing to the tension between this model and the democratic principle, the court might be perceived by the public as a political player whose actions lack legitimacy, a situation that is liable to erode its standing. Finally, the great difficulty in changing the Constitution and the transferring of the moral discourse to the confines of the court is likely to weaken the public’s motivation to deal with issues of values, which would lead to the dilution of the democratic discourse.3 Various attempts have been made to overcome the failings of this constitutional model, or at least some of them. One proposed solution is to adopt a democratic appa ratus for electing justices that would grant the public control over the composition of the court. The selection of justices whose stance on disputed constitutional issues coincides with the public’s position could be used by the public as a means, albeit indirect and slow, yet effective, to participate in the process of fashioning the Constitution through interpretation. Thus, for example, Donald Kommers explains the system applied in Germany according to which the Parliament elects the judges of the Constitutional Court: What makes the Constitutional Court’s ‘activism’ less objectionable in terms of democratic theory, however, is that Parliament – not the executive – elects each justice by a two-thirds vote for a single nonrenewable term of twelve years, thereby averting the rise of an aging judicial oligarchy out of tune with major currents of modern life.4 2 For a classical expression of this argument, see AM Bickel, The Least Dangerous Branch: The Supreme Court at the Bar of Politics (New Haven, Yale University Press, 1962) 16: ‘The root difficulty is that judicial review is a counter-majoritarian force in our system’. For a contemporary expression of the argument, see J Waldron, Law and Disagreement (Oxford, Oxford University Press, 1999) 255–62. 3 The classical expositor of this argument is James Bradley Thayer. See JB Thayer, OW Holmes, F Frankfurter and PB Kurland (eds), John Marshall (Chicago, University of Chicago Press, 1967) 106–07: ‘the exercise of judicial review, even when unavoidable, is always attended with a serious evil, namely, that the cor rection of legislative mistakes comes from the outside, and the people thus lose the political experience, and the moral education and stimulus that comes from fighting the question out in the ordinary way, and correcting their own errors’. For a contemporary expression of this claim, see M Tushnet, ‘Policy Distortion and Democratic Debilitation: Comparative Illumination of the Countermajoritarian Difficulty’ (1995) 94 Michigan Law Review 245, 247. 4 See DP Kommers, ‘Comparative Constitutionalism: German Constitutionalism: A Prolegomenon’ (1991) 40 Emory Law Journal 837, 844. On this argument, within the American context see, TJ Peretti, In Defense of a Political Court (Princeton, Princeton University Press, 1999) 84–85, 131. See also L Hilbink, ‘Beyond Manicheanism: Assessing the New Constitutionalism’ (2006) 65 Maryland Law Review 15, 21–25; W Cohen and M Cappelletti, Comparative Constitutional Law: Cases and Materials (Indianapolis, Bobbs-Merrill, 1979) 76.
Why a Constitution? 11
Similar statements were made by Victor Comella: Judicial review of legislation may give rise to a ‘democratic objection’, inasmuch as the legisla tion in question is the product of a democratic legislature. This objection may be minimised if the members of the court are selected in ways that are relatively democratic.5
There are those who propose adding to the democratic election apparatus a term limit for judges. Comella explains this as follows: [A] country . . . may limit the term of judges of the constitutional court to reduce the risk of a serious gap between the constitutional jurisprudence of the court and the basic moral and political beliefs of the people and their elected representatives.6
It is customary in most Western countries to limit the tenure of judges to about 10 years.7 The United States is an exception to this trend: the term of service is not set in years, and there is no mandatory age of retirement.8 Yet, scholars in the United States are critical of the current apparatus and call for aligning it to that of other democracies.9 A standard argument raised against democratization of election apparatuses is that this would expose the appointment process to political wheeling and dealing. This claim should be rejected outright for two reasons. First, if we agree that deciding constitu tional issues is not mechanical but demands the application of strong discretion, and if it is presumed that the stances of a judge influence his discretion, then political trading at the election stage, during which every side tries to bring about the election of a judge whose worldview seems to that side fitting and proper, is precisely the aim which the democratic apparatus for electing justices wishes to obtain and not a shortcoming it wishes to avoid. Second, political tradeoffs around the appointment to an office that concentrates within its grasp such great political power is unavoidable.10 The question is 5 VF Comella, ‘The European Model of Constitutional Review of Legislation: Toward Decentralization?’ (2004) 2 International Journal of Constitutional Law 461, 468; J Bell, Judiciaries within Europe: A Comparative Review (Cambridge, Cambridge University Press, 2006) 25–26. 6 Comella, ibid. 7 In Germany the judges of the Federal Constitutional Court are appointed for a 12-year period, and they cannot be reappointed at the end of the period. See DP Kommers, The Constitutional Jurisprudence of the Federal Republic of Germany, 2nd edn (Durham, Duke University Press, 1997) 20–21; In Italy, and Spain, judges of the Constitutional Court are appointed for nine years and cannot be reappointed for a second con secutive term. For Italy, see Costituzione [Cost], art 135; TG Watkin, The Italian Legal Tradition (Farnham, Ashgate Publishing, 1997) 90; for Spain, see Constitución Española, BOE n 311, 29 December 1978, s 159(3); E Merino-Blanco, The Spanish Legal System (London, Sweet & Maxwell, 1996) 97. 8 This situation raises more than a few doubts as to the capability of the aged judges. See, eg DJ Garrow, ‘Mental Decrepitude on the US Supreme Court: The Historical Case for a 28th Amendment’ (2000) 67 University of Chicago Law Review 995: ‘Mental decrepitude and incapacity have troubled the United States Supreme Court from the 1790s to the 1990s. The history of the Court is replete with repeated instances of justices casting decisive votes or otherwise participating actively in the Court’s work when their colleagues and/ or families had serious doubts about their mental capacities’. It should be noted, though, that the average age of retirement of American Supreme Court Justices is 70, after an average term of 15 years. See the statistics appearing in HJ Abraham, Justices, Presidents and Senators: A History of the US Supreme Court Appointments from Washington to Clinton, revised edn (Lanham, Rowman & Littlefield Publishers, 1999) 379–81. 9 See SB Prakash, ‘Book Review: Americas Aristocracy: Taking the Constitution Away from the Courts. By Mark Tushnet’ (1999) 109 Yale Law Journal 541; J Resnik, ‘Judicial Selection and Democratic Theory: Demand, Supply and Life-Tenure’ (2005) 26 Cardozo Law Review 579; DR Stras and RW Scott, ‘Retaining Life Tenure: The Case for a Golden Parachute’ (2005) 83 Washington University Law Quarterly 1397. 10 See JC Yoo, ‘Choosing Justices: A Political Appointments Process and the Wages of Judicial Supremacy’ (2000) 98 Michigan Law Review 1436, 1437: ‘If . . . judicial power has expanded such that in one direction or another, the Court will be a pervasive influence on a wide range of issues that can only in a partial and periph eral way be considered legal rather than political, it is only inevitable that players in the political process will seek to advance their preferences via Supreme Court nominations’.
12 Gideon Sapir only whether it will take place under fair conditions and in the light of day or will be determined under inequitable conditions and in secrecy. Of these two possibilities, the first seems more attractive. If there is any drawback at all to the proposal to adopt a democratic judicial selection process and to limiting judges’ terms of service, then it resides in this proposal not prop erly avoiding the dilution of the democratic discourse. As noted, the source of the diffi culty rests in making the court the final arbiter over serious issues. This problem does not change even if judges are elected democratically. To be sure, the adoption of a demo cratic apparatus allows the public to influence the judicial product, but it does not return the moral discourse to the public arena. The public can, perhaps, try to influence the outcome, but it cannot take an active part in the decision-making process. I agree whole heartedly with the opinion of Jeremy Waldron that ‘impotent debating about what a few black-robed celebrities might decide in the future is hardly the essence of democratic citizenship’.11 B. Dialogical Constitution According to a second rationale, the function of the Constitution is to create public dia logue over important issues. This rationale is based on a number of premises: first, in every political unit, as homogeneous as it may be, there are disputes over many issues. Second, under these circumstances, the principle of justice that should be applied is the principle of democratic determination. Third, democracy requires that decisions over essential issues be taken by the public. Democracy not only grants the public a right to discuss and decide on these issues, but imposes on it a moral obligation to effectuate this right. Fourth, the structure of the political system provokes a fear that the decisions over essential issues will be made in passing. The role of the Constitution is to provide a response to this fear and to ensure that essential questions will receive proper attention. In a constitutional system based on the premises just described, the constitutional model will be delineated as follows: the Constitution will determine a series of abstract principles, leaving their precise manner of implementation open to discussion. The court will be given a certain amount of authority to interpret the Constitution and review pri mary legislation but the final word will be reserved for the legislature. According to this model, the function of the court is to stimulate the political system to discuss constitu tional issues seriously. The rationale of a Constitution as promoting dialogue is relatively new in constitu tional discourse, but in recent years it has been gaining momentum. As Stephen Gardbaum notes,12 constitutional models appropriate for this rationale were adopted in recent decades in a number of countries belonging to the British Commonwealth group, such as Canada, New Zealand, and Britain. Adopting the dialogic model involves two main dangers that have been well described by Mark Tushnet. The first peril is that the legislature will do as it wishes while ignoring the court’s positions and the second, that the legislature will accept the court’s positions See Waldron, Law and Disagreement (n 2) 291. S Gardbaum, ‘The New Commonwealth Model of Constitutionalism’ (2001) 49 American Journal of Comparative Law 707. 11 12
Why a Constitution? 13
without invoking independent reasoning. Both dangers are actually one set in two sce narios, namely, that the model will not succeed in creating true dialogue. Tushnet argues that the Canadian experience demonstrates well the realization of the second danger.13 He points out the dearth of instances in which the Canadian legislature – federal or provincial – made use of the known ‘Notwithstanding Clause’ as an indica tion that the Canadian legislature accepts the judiciary’s positions without reflection, in a way that prevents the formation of a true dialogue.14 Yet, just as Tushnet himself alludes, it is possible to propose at least two alternative explanations for the scant use of the ‘notwithstanding’ mechanism. First, the Canadian ‘notwithstanding’ mechanism is formulated not as an instrument for overriding the court’s interpretation of the Constitution, and for the proposal of a no less legitimate alternative interpretation, but as an instrument for overriding the Constitution.15 Under these circumstances the legis lature’s abhorrence of the use of the ‘notwithstanding’ mechanism is understandable, since it erroneously perceives – or at least assumes that the public will perceive – this use as an illegitimate attempt to detract from basic constitutional principles.16 A change in the formulation of the ‘notwithstanding’ apparatus, which would be accompanied by a campaign explaining the dialogic logic underlying it, might lead to a conceptual change, which would increase the legislature’s willingness to use the implementation that was put at its disposal.17 Another possible explanation for the scant use of the Canadian ‘notwithstanding’ mechanism is related to the way the apparatus and its use were etched into the awareness of the Canadian public. The first to use the mechanism, applying it most extensively, was Quebec. In June 1982, a short time after the enactment of the Charter, the Quebec National Assembly passed a law that presumed to immediately add to all existing Quebec legislation a ‘notwithstanding’ mechanism, which would apply retroactively through to the day of the passing of the Charter. This law did not pretend to defend specific enact ments against annulment but rather to challenge the very legitimacy of the Charter.18 The law was discussed in a number of courts and ultimately ratified in principle by the Canadian Supreme Court.19 Perhaps the use Quebec made of the ‘notwithstanding’ mech anism, as well as the confirmation the Canadian Supreme Court gave to this use, unfairly 13 M Tushnet, ‘Comparative Constitutionalism: State Action, Social Welfare Rights, and the Judicial Role: Some Comparative Observations’ (2002) 3 Chicago Journal of International Law 435, 450: ‘Canada’s experi ence with the notwithstanding clause suggests, although not conclusively, that the clause has failed to create a distinctive form of judicial review, and that Canada has a rather robust form of judicial review, the notwith standing clause notwithstanding’. 14 ibid: ‘The clause has rarely been invoked, for reasons that are complex’. For a review of the range of the use of this apparatus up to 2000, see B Billingsley, ‘Section 33: The Charter’s Sleeping Giant’ (2002) 21 Windsor Year Book of Access to Justice 331, 339–43; K Roach, ‘Constitutional, Remedial, and International Dialogues About Rights: The Canadian Experience’ (2005) 40 Texas International Law Journal 537, 543. 15 S 33 of the Canadian Charter states that: ‘33(1) Parliament or the legislature of a province may expressly declare in an Act of Parliament or of the legislature, as the case may be, that the Act or a provision thereof shall operate notwithstanding a provision included in section 2 or sections 7 to 15 of this Charter’ (emphasis added). 16 See J Waldron, ‘Some Models of Dialogue Between Courts and Legislatures’ (2004) 23 Supreme Court Law Review (2d) 7, 36–38. 17 Yet, I agree with the misgivings expressed by Waldron concerning the chance of convincing the supporters of a human rights regime to agree to his formulation of the ‘“Notwithstanding” clause that involves an honest acknowledgment that . . . a . . . legislature might have a view of rights that was, though controversial, no less reasonable than the view arrived at by the judiciary’ (ibid). 18 See CP Manfredi, Judicial Power and the Charter: Canada and the Paradox of Liberal Constitutionalism (Oklahoma, University of Oklahoma Press, 1993) 200–01. 19 Ford v Quebec (AG) [1988] 2 SCR 712 (Can).
14 Gideon Sapir blemished this apparatus and created an image of it as a tool undermining the essential public responsibility to guard basic fundamental values.20 In conclusion, the scant use of the Canadian Notwithstanding Clause does not necessarily attest that the mechanism is unable to serve as an instrument for the creation of dialogue. Proper formulation and judicial meticulousness over the proper use of the apparatus will enhance its chances to attain the goal for which it was created. C. Gagging Constitution In a democracy decisions are commonly made after open public discussion in which each side attempts to gain supporters for its position. Yet, even in a democratic system there may be reasons to want to silence certain disputes and to prevent dealing with them publicly and openly. This refers to topics that are particularly divisive, and on which public discussion is liable to arouse negative feelings, to deepen fissures and demand a high cost in terms of time and political inputs. In such instances, and in contrast to accepted psychological intuitions, gagging can play a positive role. An entrenched Constitution can be of help in the creation and maintenance of a particularly strong gag rule. The reason is rather simple: the supremacy and entrenchment of the Constitution make it difficult to amend. Under these circumstances, it is very likely that civil motiva tion to take part in the decision-making processes will decline, a fact that will limit the scope and lower the intensity of the public discourse over the issues being anchored in the Constitution. This last point clarifies one major difference between the first justification for utilizing constitutional strategy and the rationale proposed now. It has to do with the way each model perceives the silencing effect of the constitutional strategy. According to the first understanding, the silencing effect is perceived as a disadvantage that one should try to overcome, or to take into overall consideration.21 Yet, according to the understanding proposed now, the dilution of the public discourse is essentially the aim for which the constitutional strategy is chosen. Note well that the perception of the Constitution as a gag rule does not mean that the dilution of public discourse constitutes by its very nature an advantage, but that for certain issues and under certain circumstances, it serves a positive purpose. A second significant difference between protecting Constitution and gagging Constitution has to do with their criteria for selecting the issues to be included in the Constitution. Protecting Constitution attributes a positive intrinsic value to the content of the Constitution. The Constitution contains the ‘correct’ value determinations that need protection. In contrast, gagging Constitution does not assume the justness of the determinations anchored in the Constitution. The criterion for selecting the issues to be included in the Constitution according to this model is not justice or importance but rather utility. It is highly probable that all the rival groups will find flaws in the agree ment anchored in the gagging Constitution and will even criticise it harshly. Yet, despite this, they will agree that it is worthwhile to include the determination in the Constitution, 20 For a description of Quebec’s use of the Notwithstanding Clause as a blemishing factor, see PW Hogg and AA Bushell, ‘The Charter Dialogue Between Courts and Legislatures (Or Perhaps the Charter of Rights isn’t Such a Bad Thing After All)’ (1997) 35 Osgoode Hall Law Journal 75, 83. 21 See n 3 above.
Why a Constitution? 15
for fear that the political system will pay too heavy a price for leaving it outside the Constitution. The constitutional document that will result in the gagging Constitution model will not be one that we will hang on the walls of schools and oblige our children to memorise. It does not set high-minded principles. Just the opposite, it anchors com promises that no one actually loves, and despite that, we can agree that it is valuable for its silencing effect. For a Constitution to reach its goal as a gag rule, it must silence all the players in the arena, that is, to gag both the political system and the court. At first glance, it seems that this is an impossible goal, since ostensibly there are only two possibilities: that the consti tutional method will create a selective gag rule, or that it will not succeed to create a gag rule. The question as to which of the possibilities will materialise depends upon the consti tutional model that will be chosen. If the protecting Constitution is selected, the gag will be selective since the Constitution will be submitted to the interpretation of the court. Conversely, in choosing the dialogical Constitution, no gagging will be created at all. According to this model, the court, to be sure, does not enjoy preference over the political system, but the equality between them is not in that both are gagged but in that both are not gagged. The question, therefore, is whether a Constitution can serve as a gag rule silencing all, or whether it is destined to one of two failures: fanning the discussion, or selective gagging that leaves the court free to fashion the arrangements as it wishes. Elsewhere I have argued that it is possible to employ a Constitution to obtain full gag ging by means of including in the Constitution detailed arrangements for the topics whose gagging is sought. This specification will achieve gagging because whether the right to the last word is reserved to the court (as in protecting Constitution) or given to the legislature (as in dialogical Constitution), the holder of the right to the last word is not authorised to ignore or annul clear instructions rooted in the Constitution.22 The proposal to include in the Constitution detailed arrangements and not just to include abstract principals does not fit the common approach that presumes as obvious that the Constitution is intended to be vague by virtue of its definition. Yet, generaliza tion and vagueness are not a logical derivative of the concept ‘Constitution’. The ques tion as to whether to specify and clarify or to generalise and make vague this document of the highest normative validity depends on the aim that is to be achieved. In order to achieve the silencing effect, clarity and specification are needed. III. WHICH MODEL IS APPROPRIATE FOR ISRAEL?
A. The First Model Does Not Fit The protecting constitutional model is patently unfit for the State of Israel because the special nature of Israel’s society and political system hones and worsens each of its short comings. The Israeli society is characterised by deep rifts among the groups composing it. 22 It should be noted though, that even with the most detailed text it is not possible to completely prevent the court from employing creative methods of interpretation. The continental experience clarifies this point. See, eg JR Maxeiner, ‘Legal Certainty: A European Alternative to American Legal Indeterminacy?’ (2007) 15 Tulane Journal of International and Comparative Law 541, 570–71; R Zimmermann, ‘Statuta Sunt Stricte Interpretanda? Statutes and the Common Law: A Continental Perspective’ (1997) 56 CLJ 315, 320–21, 325–26.
16 Gideon Sapir In such circumstances, doubt grows stronger over the possibility of achieving consensus on the momentous questions of values; an agreement which is, of course, an imperative condition for the legitimacy of a protective Constitution. Placing the authority to enforce the Constitution in the hands of the court would allow it to fashion the Israeli value sys tem. The latter danger is particularly strong in the light of the judicial culture that has developed over the past few decades in Israel, which is characterised by a self-awareness of mission and duty to serve as the delineator for issues of values. The culture of political discourse and decision-making in Israel is woefully poor, and the ethos of governmental fairness is deficient. These facts usually serve in Israel pre cisely as an argument in favour of enhancing the court’s role in supervising government goals as well as fashioning them – with or without the aid of a Constitution – with the claim that the court is the last hope. It seems, however, that reality, after three decades of intensive judicial intervention, shows that alongside a certain benefit (that grew out of the court’s ever-increasing intervention) in eliminating phenomena of corruption and improper administration, this intervention also contributed to the weakening of the political system and to detracting from its ability to develop an independent moral spine. When the court took upon itself the role of gatekeeper, the legislature considered itself free of acting according to its wishes without being bothered by the question of whether its behaviour was upstanding.23 The final reason for the incompatibility of the first model with Israeli reality resides in the Israeli judicial selection mechanism. This mechanism ascribes to the incumbent Supreme Court justices a great deal of power in choosing those who will join their ranks.24 In the past, the composition of the Israeli Supreme Court reflected the values of the old elite. As long as the system for electing judges remains as it is, this elite is ensured that its values will continue to dominate the court. This fact increases the tension between the great power placed in the hands of the court and the principle of democratic determination, and it severely erodes the Israeli public’s faith in the court.
23 D Barak-Erez, ‘The Justiciability of Politics’ (1999) 8 Plilim 369 (in Hebrew). A similar danger exists in other countries as well. For such a fear as expressed by a New Zealand scholar, see G Huscroft, ‘Protecting Rights and Parliamentary Sovereignty: New Zealand’s Experience with a Charter-Inspired, Statutory Bill of Rights’ (2002) 21 Windsor Year Book of Access to Justice 111, 123: ‘The power of the courts to strike down legislation can . . . result in perverse incentives. Governments may exploit judicial willingness to make hard decisions by leaving rights to be vindicated in litigation rather than dealing with them in the legislative process. American experience suggests that the temptation is always there to leave the constitutional dirty work to the courts’. For the argument that that is what happens in reality, in the American context, see M Tushnet, Taking the Constitution Away From the Courts (Princeton, Princeton University Press, 1999) 54–71. 24 See, eg M Edelman, Courts, Politics, and Culture in Israel (Charlottesville, University of Virginia Press, 1994) 34: ‘By established practice, appointments to the Supreme Court require an affirmative vote of all three justices on the panel’. Moshe Ben-Zeev, who served as the Attorney General between the years 1963–68 wrote: ‘It is impossible to appoint a person to the post of a judge, and certainly not to the post a Supreme Court jus tice, if the appointment is uniformly opposed by the three Supreme Court judges participating in the Committee. I had hoped that this was an unwritten custom, but if this is not the case, it should be anchored in law’ (my translation, GS), M Ben-Zeev, ‘Politics in the Appointment of Judges’ (27 May 1981) The Lawyer 13 (in Hebrew). For additional sources see M Haller, ‘The Court That Packed Itself’ (1999) 8 Azure 64. Over the last few years fissures are beginning to appear in this conventional understanding, and the Supreme Court just ices are increasingly confronting Ministers of Justice who are unwilling to be submissive.
Why a Constitution? 17
B. The Second and Third Model Fit As the first model does not fit Israel, it is not recommended to adopt it. There is, however, definitely room to consider use of the third model, the second model, or a combination of the two. As stated, the decision to impose a gag over a certain issue derives from its being the subject of deep dispute, and from the conclusion that an open discussion on that topic is liable to be detrimental to national cohesiveness and to waste a great deal of precious public energy. In Israel, which is riven in terms of values and which grapples with main taining the elementary conditions of security and stability, there are a number of dis puted issues whose gagging by means of the Constitution would likely be beneficial. The most salient candidate for such a treatment is the controversy over matters of religion and State, which has split the Jewish public in Israel from the beginning of the Zionist movement. Another dispute that warrants consideration of gagging by means of the Constitution is the one between the Jewish majority and the Arab–Palestinian minor ity over the identity of the State. Conducting debate on these two issues within the ordin ary political process utilises unreasonable inputs from the political system and sunders the Israeli public, so it seems reasonable to anchor arrangements in these two spheres in a Constitution of the gag rule type. A possible argument against the proposal to employ a gagging Constitution in Israel is that it is not necessary. In order to gag a given topic, it is not imperative to adopt a con stitutional strategy. The gag rule can also be rooted in regular legislation, in secondary legislation, or even in a political agreement. The ‘status quo’ arrangement on matters of religion and State, which operated within the Israeli political system for a long period,25 could serve as a good example of a successful, non-constitutional gag rule. From the establishment of the State, and for over a generation, the realm of religion and State in Israel enjoyed relative quiet and stability. This is somewhat surprising considering the rampant differences of opinion among the various groups in Israel on these matters. The startling relative quiet in this sphere may be attributed to the notion that the political system chose to invoke a gagging approach for this topic. Traditionally included in the coalition agreements for establishing the government was a paragraph stating that the status quo over matters of religion and State would remain unchanged.26 As many writ ers note, that gagging agreement, which was re-anchored from one Knesset (Israeli Parliament) to the next and from government to government, derived from exactly those reasons cited above as the motive for adopting a gagging Constitution: understanding 25 On the ‘status quo’, see, eg G Sapir, ‘Religion and State in Israel: The Case for Reevaluation and Constitutional Entrenchment’ (1999) 22 Hastings International and Comparative Law Review 617. 26 See, eg The Coalition Agreement between the Alignment Parties (The Labour Party – Mapam – Arab Lists), The National Religious Party, The Independent Liberal Party, DK 69 (1974) 706: ‘(e) Status quo over religious issues (1) In continuation of previous coalition agreements, also in the period of the Eighth Knesset the status quo will be maintained concerning public transportation on the Sabbath and holidays. (2) As above, the status quo will be maintained over the law of marriage and divorce and all other religious matters’. The Coalition Agreement between the Parties, The Israel Labour Party, the Herut Bloc – Liberals, the National Religious Party, The Independent Liberal Party, DK 56 (1970) 272: ‘(e) Status quo over religious issues (1) In continuation of coalition agreements in the Third, Fourth, Fifth, and Sixth Knessets, there will be maintained in the term of the Seventh Knesset the status quo concerning public transportation on the Sabbath and holidays. (2) As above, the status quo will be maintained over the law of marriage and divorce and all other religious matters’ (my translation, GS).
18 Gideon Sapir that the topic is hotly disputed and that open discussion is liable to create a rift among different groups that will deter from national cohesiveness and draw a large part of the public energy that is needed for the solution of other vital tasks.27 If, as just demonstrated, the goal of gagging destructive disputes is realised even without using constitutional weaponry, the question arises as to why this should not be sufficient without involving the Constitution on the issue at hand. Two responses can be given to this question: the first is that there is indeed no need to use the Constitution for purposes of gagging, and it is required only in cases in which we are interested in a par ticularly forceful gag rule. The second answer is more interesting: precisely when the political system chooses to switch from a model of parliamentary democracy to a model of constitutional democracy, the efficacy of the usual gagging techniques is weakened and the need arises to use a constitutional gagging technique. Let us assume that the parties to the dispute agree to gag it. The gagging will be main tained as long as the sides continue to want it, or if the agreement included an efficient enforcement mechanism. That is the usual situation, but when the Constitution enters the picture, the fate of the gag rule is expropriated from the grip of the parties, and it becomes exposed to possible third-party intervention, since the court is empowered to interpret and implement the Constitution. The only way to ensure effective gagging in a constitutional democracy is to upgrade the gag rule and give it constitutional status. It is possible to illustrate this argument through the history of the Israeli gag rule in the area of religion–State relations. After a long period in which it was, more or less, maintained, the agreement concerning religion and State, and with it the gagging, is quite obviously collapsing. Arrangements maintained for a great many years with no significant change are being opened and abandoned with the waging of a fierce, harsh dispute. This change can be attributed to many factors,28 but it seems that the Supreme Court is playing a central role in this change. If in the past, the players in the political system could suffice with political agreements for removing certain issues from the pub lic agenda, once the court determined that it was within its purview to invoke judicial review, they became dependent upon the good will of the court. In these circumstances, the need arose to change the pattern of gagging and to reinforce it. The dialogical Constitution model is also particularly appropriate for Israel. As stated, one of the premises at the base of this model is that the regular political appara tus does not motivate the political players to discuss vital issues seriously enough. This assumption is clearly verified in the State of Israel, a country in which there is a danger ous combination between the intensity of the disagreements over essential issues and the 27 Eliezer Don-Yihya characterizes the ‘status quo’ arrangement as consociational, See E Don-Yihya, Religious Institutions in the Political System – The Religious Councils in Israel (Jerusalem, Jerusalem Center for Public Affairs, 1989) (in Hebrew). As Lijphart explains, at the basis of consociationality stands the aware ness of the political leadership for the potential for a rift and instability inherent in a dispute that splits society, an awareness that leads to the development of patterns whose aim is to enable coexistence, and foremost among them being to refrain from applying the principle of decision by majority. See A Lijphart, ‘Consociational Democracy’ (1969) 21 World Politics 207. My argument is that the ‘status quo’ arrangement is in effect an informal gag rule coinciding well with this description. Accordingly, not only is the content of the arrangement intended to ensure stability and cooperation but also willingness in principle to immunize it against discussion and change. 28 For a discussion, see A Cohen and B Susser, ‘Between Fragile Consensus to Breaking Consensus – Changes in the Relationship between Religion and State: Between Consociationalism and Resolution’ in M Mautner, A Sagi, and R Shamir (eds), Multiculturalism in a Democratic and Jewish State (Tel-Aviv, Ramot Publishing House, 1998) 675, 675–701 (in Hebrew).
Why a Constitution? 19
lack of discourse and deep thinking about those issues before deciding upon them. It is therefore important in Israel to create mechanisms that will guarantee a minimal degree of seriousness. Yet, these apparatuses must absolutely not bypass the political system and replace it, as the first model proposes, but rather guide the political system and obli gate it to develop a proper culture of discourse and decision-making. This is precisely what the model of a Constitution as promoter of dialogue offers. As was pointed out above, the great doubt hovering over the dialogic model is whether it will succeed in achieving the double purpose set for it: reinforcing seriousness over vital issues without taking away from the public the right and duty to decide upon these issues by itself. As noted, there is a fear that the dialogic model will not attain the goal of dialogue, either because the legislature will ignore the position of the court and refuse to seriously consider constitutional issues, or because it will unquestioningly accept the court’s stance, without actively participating in the decision-making process. There are some who argue that in light of the political culture prevailing today in Israel, the greater fear is the legislature ignoring the court.29 Conversely, Jeremy Waldron is convinced that the opposite fear is always stronger: Partly because systems of judicial supremacy are associated with a strong culture of self- righteousness on the part of the judiciary and its academic supporters, there tends to be much less in the way of genuine dialogue than in constitutional systems where there is legislative supremacy (of one kind or another). Legislative supremacy is very seldom accompanied by a sense, among legislators, that they have nothing to learn from any other branch of government; and in cases where the legislature remains sovereign, but the judiciary is given a role, the judges are more likely to be listened to by the legislature than the legislators are likely to be listened to in a system that gives the judges final say.30
It seems that Waldron’s fear is especially relevant in the Israeli context. Those who set the tone in the media and in legal academe in Israel regularly provide broad public back ing to the Supreme Court and take care to describe its rulings as deriving directly from the Constitution, without admitting that on many topics there is legitimate debate that cannot be decided upon easily. This approach delegitimises the legislators who try to stand as equals vis-à-vis the court. Under these circumstances, if the dialogic model is to be adopted in Israel, it would have to be fashioned in a way that would not only grant the legislature the theoretical possibility to overturn the court’s position, but would rein force the legitimacy of this step. Owing to lack of space, I will not expand on this issue. C. Can the Dialogical and Gagging Models be Combined? Thus far this chapter has argued for the option to use the dialogical and gagging models separately, but their combination is also possible, both intellectually and practically. As for their respective rationales, these two models are akin. To be sure, the second model seeks to achieve dialogue while the third wishes to gag it, but both of them begin at the same starting point, namely, the existence of ideological pluralism. Even the different 29 See AL Bendor and Z Segal, The Hat Maker – Discussions with Justice Aharon Barak (Or-Yehuda, Kinneret Zmora-Bitan Dvir, 2009) 138 (in Hebrew) (Barak says, ‘The British approach is considered by the English as good for them . . . It’s not England here. In Israel there would be a great deal of hoopla and nothing would change’) (my translation, GS). 30 Waldron, ‘Some Models of Dialogue’ (n 16).
20 Gideon Sapir mechanisms that they propose do not clash. Anyone convinced that public discussion is valuable can agree that for certain issues prolonging discussion, rather than being bene ficial, will be harmful. The same applies to the obverse. Those who support gagging certain issues do not do so out of an abhorrence of debates over essential issues but because they feel that with regard to those issues the danger inherent in discussion is greater that the possibilities it provides. The aim of the gagging Constitution is to remove the issues included in the Constitution from public discourse. To attain this goal, the Constitution must contain a detailed arrangement for the gagged topics. Conversely, creation of dialogue obliges the constitu tional anchoring solely of abstract principles. Nevertheless, it is possible to combine these two models, although such a blend will create a hybrid product with obvious dif ferences among its elements. The part whose aim is gagging will be detailed while the part whose goal is encouragement of dialogue will be succinct and abstract. The combin ation of the objective of the second model with the third is, therefore, intellectually logi cal and practically possible, but the seam between its different parts will be sewn together with rough stitching. IV. WHICH MODEL IS USED IN ISRAEL TODAY?
A. Not Dialogue Some people argue that the State of Israel has already adopted the dialogical model. This claim is supported by two pieces of data. The first is that in contrast to other Constitutions, in which there is great difficulty in changing the constitutional text, Israel’s Basic Laws are relatively easy to change. This fact enables the Knesset to over turn a decision of the court – if it does not see this determination as favourable – by an amendment to the Constitution. Yoav Dotan feels that in line with this concept both the Knesset and the court have to reconcile to the idea that a situ ation in which the Knesset overturns a determination by the court and amends the constitution is not a situation of severe ‘constitutional crisis,’ indicating a major confrontation between these two authorities, but rather a normal, possible situation in which each side fulfills its func tion in the constitutional dialogue.31
Aharon Barak, too, has reiterated this point several times in his academic writings,32 even though he himself has expressed support for raising the threshold of the entrench ment.33 Another datum cited as support for the argument that Israel has adopted the dialogical model relates to the ‘notwithstanding’ mechanism, which is contained in the Basic Law: Freedom of Occupation. To evaluate these claims, especially the second one, one must first describe the circumstances in which the ‘notwithstanding’ mechanism was adopted and the context in which it was employed in Israel. Since the establishment of Israel, frozen beef has been imported exclusively by the State. The reason given was that this was a necessity since Israel was in a state of emer 31 Y Dotan, ‘A Constitution for the State of Israel – The Constitutional Dialogue after the “Constitutional Revolution”’ (1997) 28 Mishpatim 149, 207 (in Hebrew). 32 A Barak, The Judge in a Democracy (Princeton, Princeton University Press, 2006) 236 ff. 33 ibid 239–40.
Why a Constitution? 21
gency and therefore had to ensure the regular supply of basic commodities and prevent speculation. According to this arrangement, the State imported only kosher meat, explaining that this simplified the task, since all citizens of the State can eat kosher meat. In 1992 the Government decided to privatise the import of meat. The Shas Party, which was a member of the coalition, opposed granting import licenses in an uncontrolled manner. Shas argued that giving a license for the import of non-kosher meat would dis rupt the status quo on this issue and lead to the flooding of the country with non-kosher meat whose price is lower than kosher meat. The Government discussed this issue a number of times and ultimately decided to arrange for the privatization of the meat import branch through legislation. Until the legislation is completed, the prevailing situ ation will remain as it is, meaning, the State will be the sole importer of meat. The Mitral Company – which sought to import meat to Israel – submitted a petition on this issue.34 The Supreme Court accepted the petition and ordered the State to give Mitral import licenses. The court determined that the refusal to grant import licenses violated Mitral’s freedom of occupation, and noted in dictum that this violation would not meet the proportionality test set in the Basic Law: Freedom of Occupation.35 The ruling in the Mitral v Prime Minister case made the coalition frantic. The Shas faction, which understood that the usual political instruments would not suffice in this instance, aimed its arrows at the Basic Law: Freedom of Occupation. The political crisis kept intensifying and the Basic Law was put in danger. At this stage Supreme Court Justice Aharon Barak entered the fray and proposed to the political system a compro mise solution – the inspiration for which he gained from the Canadian Charter – accord ing to which the Basic Law would be amended and a Notwithstanding Clause would be added to it.36 Barak’s intervention in the political crisis derived from his desire to protect the Basic Law: Freedom of Occupation and the entire endeavour of Basic Laws. As he saw it, it was preferable to weaken the Basic Law a little rather than endanger its very existence. The Government adopted Barak’s compromise proposal, and following it, the Knesset, too. After the amendment of the Basic Law: Freedom of Occupation, the Knesset enacted the Import of Frozen Meat Law, 5754-1994. The law – which met the requirements set forth in the ‘notwithstanding’ mechanism – determined that ‘no person shall import meat unless he has received a certificate of kashrut in relation to it’.37 From 1994 until today the Knesset has made no further use of the Notwithstanding Clause. Is it true that the constitutional model applied in Israel promotes dialogue between the legislature and the court? I think not. Amendment of the Constitution, even if it can be done relatively easily, does not constitute dialogue, since the very amendment gives vent to an ‘admission’ of deviation from the Constitution in its present format. The Notwithstanding Clause in the Basic Law: Freedom of Occupation is phrased not as an instrument for overriding the court’s interpretation of the meaning of the Constitution, HCJ 3872/93 Mitral v Prime Minister and Minister of Religious Affairs 47(5) PD 485 [1993] (in Hebrew). ibid para 24 of Or J’s opinion: ‘It seems that any legislation which is seeking to condition the import of meat upon its being “kosher meat” is imposing restrictions upon the freedom of occupation. These restrictions are in contradiction to the restriction criteria’ (my translation, GS). 36 See A Barak, ‘On Amendments to the Basic Law: Freedom of Occupation’ (1994) 2 Law and Government 545 (in Hebrew); Bendor and Segal (n 29) 57 (Barak says: ‘I am the one who advised adding this paragraph, the Notwithstanding Clause’). 37 S 2 of the law. Later the name of the law was changed, and today it is the Meat and Meat Products Law, 5754-1994. 34 35
22 Gideon Sapir and for the proposal of an alternate interpretation no less legitimate, but as an instru ment for overriding the Constitution, and even then, this is possible only for a short period of time. The circumstances for the introduction of the ‘notwithstanding’ mechan ism as well, and the fact that apart from once the Knesset has made no use of it, show that the apparatus is not perceived by the Israeli public as a legitimate tool for conduct ing a dialogue between the legislature and the courts. It seems, therefore, that the presentation of the relatively easy possibility to amend the Constitution or use of the Notwithstanding Clause in its present format as dialogue is erroneous and attests to a basic lack of understanding of this concept. Additional proof of such misunderstanding can be found in Aharon Barak’s claim that dialogue exists even when the legislature does not use the Notwithstanding Clause, and it does not even amend the Constitution, but simply amends a law invalidated by the court according to its directives.38 As Manfredi and Kelly note ‘Genuine dialogue only exists when legisla tures are recognised as legitimate interpreters of the constitutions and have an effective means to assert that interpretation’.39 B. Not Gag Rule In part II I argued that for a Constitution to serve as a successful gag rule it must gag all the players and not only some of them. For that purpose, the Constitution must anchor a detailed arrangement concerning the issues that are to be gagged. The Israeli Constitution does not meet this requirement. The two Basic Laws dealing with human rights are extremely laconic in formulation. This conciseness serves the Israeli Supreme Court as an instrument for developing the Basic Laws and for using them as a platform for the creation of a complete Constitution, while totally ignoring the original intention of their framers. The enactment of the two Basic Laws on human rights took place after 44 years of failure to carry out this mission. Elsewhere I have dealt with the question of how did the Knesset, one fine day, find itself establishing a charter of human rights in a Basic Law? What prepared the way for enacting what many thought to be impossible?40 One of the answers I presented there, which was provided by the Knesset Members who promoted the step, is that the success stemmed from adopting a compromising approach in which disputed rights, such as freedom of speech, were removed from the Basic Laws and a number of concessions were made to the Orthodox camp, which traditionally opposed anchoring human rights in the Constitution. Unfortunately, it is already completely clear today that the intentions of compromise have been set aside by the Supreme Court, through its rulings that granted the Basic Laws a totally different meaning than intended by their framers. In this reality, the Israeli Constitution, a large part of which is the product of the Supreme Court, not only does not gag disputes and rifts that threaten the stability of Israeli society but even inflames disagreements and deepens rifts. If in the past decisions Barak, The Judge (n 32) 157. See CP Manfredi and JB Kelly, ‘Six Degrees of Dialogue: A Response to Hogg and Bushell’ (1999) 37 Osgoode Hall Law Journal 513, 524. 40 G Sapir, ‘Constitutional Revolutions – Israel as a Case Study’ (2010) 5 International Journal of Law in Context 355. 38 39
Why a Constitution? 23
over disputed issues were made within the political system, some of them even as a con sensual compromise, today are taken by the court despite the opposition of the players in the political field. The fact that the composition of the Israeli Supreme Court does not represent the gamut of opinions in Israeli society, but only a certain sector of it, deepens the frustration of the sector that is not represented and whose positions are consistently rebuffed. C. Yes Protection of Basic Values by the Court Israeli reality is definitely exceptional in the democratic landscape. The Supreme Court writes the Constitution, interprets it, and thereby decides in disputes concerning values that were gagged in the past by the political system. This unusual reality, which clashes with basic concepts of democracy and fairness, encounters strong opposition and shatters the public’s faith in the Court. How can the Supreme Court and its supporters justify such a situation? The answer to this question, in my opinion, can be found in statements written by Menachem Mautner in 1993, shortly after the enactment of Basic Laws on human rights: In the years following the Six-Day War, there was a strengthening, simultaneously, of all three elements taking part in the struggle for the cultural image of Israel: extreme nationalism (a prod uct of Zionism), Judaism, and western liberalism. Moreover, in those years it not infrequently seemed that the powers of nationalism and Judaism had joined forces to shove aside western liberalism from Israeli life. The Supreme Court has always made an important, perhaps deci sive, contribution to fortifying Israel’s connection with the values of the liberal west. In the 1980s, the court came out against those who challenged Israel’s link to the values of the west and acted assertively and decisively for the continued existence of this link. Anyone who believes that in the coming years Israel’s attachment to the liberal west should continue and even grow stronger must hope that the battle waged by the Supreme Court will be crowned with success.41
The statements quoted clearly express the Supreme Court’s perception of reality and well explain the justification that it and its supporters among the Israeli public ascribe to its conduct. As they understand it, Israeli society is a polarised one – in which a war is being waged between liberal forces, under the leadership of the Court, and clerical, antiliberal forces. In such a situation, whoever desires progress and light must abandon the path of dialogue and compromise that gags and support the Court’s battle. From their perspective, the Israeli Constitution is not an instrument for the protection of a common value system but a tool for the protection of the desirable value system against the threat to which it is exposed. V. SUMMARY
This chapter explained that alongside the constitutional model of a Constitution anchor ing basic values and of a court charged with enforcing it, it is possible to propose two alternate models of a Constitution. One sees the Constitution as a catalyst for serious 41 M Mautner, ‘The Decline of Formalism and the Rise of Values in Israeli Law’ (1993) 14 Tel-Aviv University Law Review 504, 596 (in Hebrew). This article is translated to English in Mautner’s 2011 book, published in English (M Mautner, Law and the Culture of Israel (Oxford, Oxford University Press, 2011) ch 4).
24 Gideon Sapir discussion of disputed essential issues, and the other considers the Constitution as a means for gagging debates on crucial questions, after a compromise has been reached over them. The model of protecting Constitution has many shortcomings, and the unique cir cumstances in the State of Israel make these flaws particularly serious. In Israel, which is riven with different values, in which the culture of political discourse is weak, and in which judges play a significant role in the appointment of their colleagues, it is unjusti fied and illogical to adopt this model. The alternate models, however, are especially appropriate for the special circumstances prevailing in Israel. In light of the poverty of political discourse in Israel it is very important to create within it a mechanism that will encourage serious discourse over essential questions, as the dialogical model proposes. In a country torn regarding values and still struggling to ensure elementary conditions of security and stability, there are a number of disputed issues the gagging of which, after having reached a compromise, is likely to be beneficial, so there is importance in creating a gagging mechanism as proposed by the model of a gagging Constitution. Unfortunately, the model customary in Israel today is the first model. Some will argue that initial encouraging signs can be discerned about the recognition of the need to use precisely the other two models, and particularly that of a Constitution as promoting dialogue. Testimony to the recognition of the importance of dialogue may, perhaps, be found in Aharon Barak’s proposal to add a Notwithstanding Clause to all Basic Laws.42 A proposal of this type also appears as one of the alternatives in the draft Constitution published by the Knesset Constitution, Law and Justice Committee.43 At the same time, there are early inklings of signs of willingness to consider use of the gag rule model.44 Despite these encouraging signs, there is definitely room for concern. For the two alternate models proposed here to be accepted and achieve their goal, a change of heart is necessary. For the creation of a true dialogue on issues of values, one must internalise that there exists in Israel a legitimate dispute over questions of values, a dispute that remains despite an initial agreement on the underlying premises concerning certain val ues. One must understand that handing the authority to decide on these issues to the court involves a heavy moral price and that the place best suited for making these deci sions is the political system. Regretfully, many Israelis have not yet internalised these truths. In order to create an effective gag rule, there must be thorough appreciation of the insight that the Israeli collective has value and that its survival depends, among other things, on its ability to bind, through compromise, the various groups within it. This truth, too, is far from having become ensconced in the hearts of many citizens of Israel. In Israel the model of the Constitution as a defender of basic values through the court became a means for extra-political decision-making. It was not born into a vacuum, but reflected – and still does – the moods prevailing among the Israeli elite. So that decisionmaking will return to the political system, so that it will be made after serious consider ation of values, and so that the gagging compromise will once again be an accessible, legitimate option, the value of democratic dialogue and the value of compromise must make their way into people’s hearts. 42 See, eg Bendor and Segal (n 29) 141–42; see also Y Dotan, ‘Judicial Review and Accountability – A Comparative Analysis’ (2007) 10 Law and Government 489, 519 (in Hebrew). 43 See S Kogut and E Zandberg, Annotated Version of Proposals for a Constitution (Constitution, Law and Justice Committee, 2006) www.metzilah.org.il/webfiles/fck/file/hatzaut_lachuka.pdf (in Hebrew). 44 See, eg the proposal for a Constitution by the Israel Democracy Institute, which includes a type of gag clause regarding issues of religion and State: Constitution by Consensus (Jerusalem, The Israel Democracy Institute, 2007) 299–302 en.idi.org.il/media/1529178/ConstitutionByConsensus_Draft.pdf.
3 The Right to Judicial Review: The Israeli Case ALON HAREL
I. INTRODUCTION
T
HE COURTS’ GROWING involvement in matters of public/political signific ance is controversial. There are many voices which maintain that courts are too powerful and influential and that they make decisions that ought to be decided exclusively by majoritarian/democratic processes. One expression of this sentiment is the growing opposition to judicial review of statutes. Another expression of this senti ment is the opposition to what is often labelled ‘creative/purposive interpretation’ – interpretation which conflicts with legislators’ ‘real’ intentions or with the literal meaning of a statute. Both judicial review of statutes as well as ‘creative/purposive inter pretation’ are perceived to replicate the political and ideological whim of judges and therefore to be illegitimate. The attack on the courts’ growing powers is shared by theorists of opposing political persuasions including liberals, such as Jeremy Waldron, Mark Tushnet and Larry Kramer, and conservatives such as Bork and Scalia.1 In Israel this scepticism is also shared by conservative leaders as well as by some leftist theorists.2 Unlike Europe and the US, the attack on the alleged activism is part of a general attack on ‘elites’. In the last few years the hostility to elites targeted the judicial branch as well as the academic elites.3 1 For liberal critics, see eg J Waldron, Law and Disagreement (Oxford, Oxford University Press, 1999) (argu ing against judicial review on the grounds that it violates the right of political participation); M Tushnet, Taking the Constitution Away from the Courts (Princeton, Princeton University Press, 2000) (challenging the constitutional powers of the courts on various grounds in particular the lack of competence of the courts); M Tushnet, Weak Courts, Strong Rights: Judicial Review and Social Welfare Rights in Comparative Constitutional Law (Princeton, Princeton University Press, 2008) (establishing that courts are not effective in protecting wel fare rights); LD Kramer, The People Themselves: Popular Constitutionalism and Judicial Review (Oxford, Oxford University Press, 2005) (challenging the belief that courts were historically designed to have exclusive constitutional powers to interpret the Constitution). For conservative critics, see, eg A Scalia, A Matter of Interpretation; Federal Courts and the Law (Princeton, Princeton University Press, 1998) and R Bork, ‘The End of Democracy? Our Judicial Oligarchy’ (November 1996) 67 First Things 21, 21–24. 2 For conservative opponents, see, eg C Shine, ‘The High Court’s Voyage of Disengagement’ (2006) 5 Netanya Academic College Law Review 105 (attacking the alleged over-activism of the Israeli Supreme Court); on leftist voices opposing the Court, see, eg R Shamir, ‘The Politics of Reasonableness’ (1994) 5 Theory and Criticism 7 (in Hebrew) (demonstrating the ineffectiveness of the Court in protecting rights). 3 Among the most pernicious manifestation of this hostility to elites is the new ‘report’ of the extreme rightwing movement ‘Im Tirzu’ against what they perceive as the leftist inclinations of the universities: www.imti. org.il/Reports/AcademicSpeechGag.pdf (in Hebrew). For a popular exposition of the faults of one of the more
26 Alon Harel Let me speculate here and suggest that some of the reasons for the anti-elitist senti ments are grounded in the psychology of those who share these sentiments and not in the sins of their targets. For various reasons these anti-elitist sentiments are directed against courts.4 The resentment directed against courts resembles sometimes the hysterical reac tion of spoiled babies who are confronted for the first time with the word ‘No’! The courts in Israel are pushed to a position of a parent who ought to discipline children who have lost control. The underlying sentiments guiding the opposition to the courts and the fact that Israeli society has lost its ability to set limits on itself suggests that even the most compelling arguments favouring judicial review will not be heard. Trashing the court is conducive to political success. And, yet, despite the passions which silence the voice and blind the sight of reason I wish to examine the significance of judicial review and its proper scope in a liberal society. I started by observing that two questions occupy the participants in the debate about judicial activism. The first question concerns the very willingness of courts to intervene in legislative decisions, that is to strike down laws.5 Some voices argue that judicial review is never justified;6 others believe that it is sometimes justified;7 and last some believe it is always justified.8 The second question – the question of ‘creative interpretation’ – is much broader; it arises also in cases involving judicial review of secondary legislation, decisions of the executive and even in cases involving interpretation of contracts. This latter ques tion touches upon the depth, the stringency, and the scope of legal interpretation. hideous reports of this movement, see A Harel, ‘The Elitophobic Anti-academic Movement in Israel: Who Wants to Destroy the Elites and Why’ (7 November 2010) hebrewu-law.blogspot.com/2010/11/blog-post_07. html (in Hebrew). 4 The conjecture that the attack on courts is part of a broader phenomenon, namely what can be labelled elite-phobia, explains the frequent link made between elites and courts. Gadi Taub, eg argues that: ‘There is something illusory in the manner in which the new elite struggles to base its hegemony: while it supports righteconomic policy, and judicial activism meant to take political power from the hands of the democratic process – the Knesset -and place it in hands of the judges of the Supreme Court’. See G Taub, ‘Israeliness – That Isn’t Us’ (2008) 45 Eretz Acheret 30, 31 www.acheret.co.il/en/?cmd=articles.323&act=read&id=2042&print=1. Guy Bechor believes that: ‘A new Justice Minister is appointed and before he opens his mouth the old elites whose power depends on the Supreme Court, universities and the media become hysterical’. See G Bechor (7 February 2007) www.gplanet.co.il/prodetailsamewin.asp?pro_id=226 (in Hebrew). It is interesting to note that while Gadi Taub identifies the Court with the ‘new elite’, Guy Bechor identifies it with the ‘old elites’. This contrast indicates that there is no serious attempt to identify who the elites which allegedly govern the courts are. The alleged link between the elites and courts was also discussed in academic literature. See M Mautner, ‘The 1980’s Years of Anxiety’ (2002) 26 Tel-Aviv University Law Review 645 (in Hebrew). Foreign theorists also attributed the alleged rise in the activism of the Israeli Court to interests of elites. See, eg M Mandel, ‘Brief History of the New Constitutionalism, or How We Changed Everything so that Everything Would Remain the Same’ (1998) 32 Israel Law Review 250, 252. 5 This question occupied the Israeli public in the 1990s as a result of the enactment of the two Basic Laws: Basic Law: Human Dignity and Liberty and Basic Law: Freedom of Occupation. 6 This position has been the dominant position for many years and was labelled by theorists ‘the omnipotent Knesset’ position. An example is the view of Berinson J in HCJ 188/63 Batzul v Minister of Interior 19 (1) PD 337, 349 [1965] (in Hebrew). It is significantly less popular now but it is still supported by some Knesset Members. See, eg MK Shaul Yahalom who argues: ‘our opposition is that we do not want the Supreme Court to have any powers to strike down legislation of the Parliament or any court whatsoever’. See Models of Judicial Review (Jerusalem, Israeli Institute of Democracy, 2002) 62 (in Hebrew). 7 This view has recently become popular in the Knesset. See the Draft Bill Amending Basic Law: The Judiciary (No 4) (Judicial Review), 2008, HH M 26, s 1, which limits the subject matters that are subjected to judicial scrutiny. The proposal was justified by its proponents on the grounds of separation of powers, the risk of mistakes on the part of the courts, the risk of inevitable delays resulting from the judicial process and the risks of undermining the public trust in the legal system. 8 See the decision of Barak J in CA 6821/93 United Mizrahi Bank Ltd v Migdal Cooperative Village 49(4) PD 221 [1995] (in Hebrew) (in particular para 79 to the judgment of Barak J).
The Right to Judicial Review: The Israeli Case 27
In this chapter I focus my attention on the first question and examine the arguments concerning judicial review of legislation. My conclusion would inevitably irritate both proponents and opponents of judicial review. On the one hand (in part II) I will argue that most of the existing arguments favouring judicial review are flawed, dangerous (in that they may contribute to judicial elitism) and, therefore give rise to justified resent ment. The conventional arguments favouring judicial review is that giving power to courts contributes to the better and more effective protection of important values. Often (although not always) this argument is based on the claim that judges are better able to identify the scope of rights and their weight.9 Part II examines critically this traditional justification and establishes its falsity. On the other hand, despite this conclusion, I shall argue in part III that judicial review is ultimately justified on the grounds that individuals have a right to a hearing. The jus tification for judicial review is simply to guarantee that individuals who claim (rightly or wrongly) that their rights were violated are provided with a forum in which they could raise their arguments, guarantee that these arguments be investigated, and trigger a reconsideration of the decision in light of the deliberation.10 II. JUDICIAL REVIEW OF STATUTES: WHY THE INSTRUMENTAL ARGUMENTS FAIL
Judicial review is often subjected to the accusation that it is anti-democratic. Judges, after all, are not elected. A judicial decision to strike down a law is therefore a violation of the right to political participation.11 Some would say that the right to political partici pation ought to be balanced against other rights.12 The results of this balancing depend on the importance and significance of the right to political participation. Others such as Jeremy Waldron regard the right to political participation as undermining the legitimacy of judicial review.13 Judicial review is a violation of the basic right of citizens to partici pate in decisions concerning the scope of rights and their weight. Controversies over the scope of rights and their weight ought to be decided by citizens in a democratic process. On a less abstract level some political scientists believe that judicial review is a scheme 9 See, eg O Fiss, ‘Two Models of Adjudication’ in RA Goldwin and WA Schambra (eds), How Does the Constitution Secure Rights? (Washington, Aei Press, 1985) 36, 43; OM Fiss, ‘Forward: The Forms of Justice’ (1979) 93 Harvard Law Review 1, 12–13. See also MJ Perry, The Constitution, the Courts, and Human Rights: An Inquiry into the Legitimacy of Constitutional Policy Making by the Judiciary (New Haven, Yale University Press, 1984) 102 (examining the contribution of constitutional provisions to the protection of rights); LG Sager, Justice in Plainclothes: A Theory of American Constitutional Practice (New Haven, Yale University Press, 2004) 199 (pointing out ‘structural features of a constitutional judiciary that make it a promising environment for the contestation of rights’). In Israel similar arguments have been made. See, eg A Barak in Models of Judicial Review (n 6) 28–29. 10 For earlier attempts to develop this argument, see A Harel and Y Eylon, ‘The Right to Judicial Review’ (2006) 92 Virginia Law Review 991; A Harel and T Kahana, ‘The Easy Core Case for Judicial Review’ (2010) 2 Journal of Legal Analysis 227; A Harel and A Shinar, ‘Between Judicial Review and Judicial Supremacy: A Cautious Defense of Constrained Judicial Review’ (2010)10 International Journal of Constitutional Law 950–975. 11 See, eg Waldron (n 1) ch 11. This argument is also heard in Israel. See also the statements made by Dr Dan Avnun, MK Shaul Yahalom and the Minister Zippi Livni in Models of Judicial Review (n 6). 12 See C Brettschneider, Democratic Rights: The Substance of Self Government (Princeton, Princeton University Press, 2007) ch 7. For my critique: A Harel, ‘Judicial Review and the Value Theory of Democracy’ (2011) 47 Representation 63. 13 See Waldron (n 1) 232 ff.
28 Alon Harel designed by social elites to guarantee their political dominance even after they lost con trol over the elected political institutions.14 Both the claim that judicial review violates the right to political participation and the view that it is being used by elites struggling to maintain their power have been raised in Israel by the opponents of the Supreme Court.15 The standard reaction of proponents of judicial review to this argument is that judi cial review is designed to guarantee better or superior outcomes. This instrumental jus tification for judicial review is based on the contingent outcomes of judicial review.16 Most of the instrumental arguments are based on meritocratic reasoning according to which courts are simply more likely to decide correctly than legislatures. The reason for the alleged superior ability of the court to make decisions is simply the superior ability of judges to make decisions in their specific areas of expertise, such as human rights or the institutional features of the court, which make it a superior decision-maker in certain designated areas. The view that judicial review is essential for the protection of human rights, the protection of minorities, the protection of democracy or the establishment of stability are all based on premises concerning the special qualities of judges or on the special features of adjudicative institutions.17 What is shared by all instrumentalist theories is the sharp differentiation between two steps of the analysis. First the theorist identifies the goals/values of the constitutional provisions. Thus, some argue that the Constitution is designed to protect minorities; others believe that it is designed to protect democracy and the integrity of the majoritar ian system, promote fertile dialogue or remove from the public agenda issues which are detrimental to social solidarity or stability. At the second stage the constitutional theor ist identifies what the institutional scheme which is the most capable of realizing these goals/values is. The power to interpret the Constitution ought to be given to the institu tion which is most capable of realizing the goals/values that the Constitution is designed to realise.18 Thus, for instance, Hamilton believes that the aim of the Constitution is to defend rights and that judges make better decisions with respect to rights.19 The constitutional theorist Ely argues that the aim of the Constitution is to defend democracy.20 Bruce Ackerman believes that the aim of the Constitution is to protect constitutional politics against mundane daily politics.21
14 M Mautner, Law and Culture in Israel at the Threshold of the Twenty-First Century (Tel-Aviv, Am Oved, 2008) 17 (in Hebrew). 15 For an academic defence of this view, see D Avnun, ‘The Enlightened Public: Jewish and Democratic or Liberal and Democratic’ (1996) 3 Law and Government 417 (in Hebrew). 16 Classical advocates of the instrumental view are Adrian Vermeule and Ronald Dworkin. See A Vermeule, Judging Under Uncertainty: An Institutional Theory of Legal Interpretation (Cambridge, Harvard University Press, 2006) 5; R Dworkin, Freedom’s Law: The Moral Reading of the American Constitution (Cambridge, Harvard University Press, 1996) 34. 17 For a discussion of instrumentalist theories, see Harel and Kahana (n 10) 230–38. 18 For a detailed analysis of this view, see Harel and Kahana ibid. 19 See A Hamilton, ‘Federalist No 78’ in The Federalist Papers 465 (New York, NAL Penguin, 1961). 20 See JH Ely, Democracy and Distrust: A Theory of Judicial Review (Cambridge, Harvard University Press, 1980) 74 (‘pursuit of participational goals of broadened access to the processes and bounty of representative government’ ought to replace ‘the more traditional and academically popular insistence upon the provision of a series of particular substantive goods or values deemed fundamental’). The Israeli Supreme Court voiced a similar view. See HCJ 11280/02 Central Elections Committee to the Sixteenth Knesset v Tibi 57(4) PD 1, 61 [2003] (in Hebrew) (emphasizing the role of the court in defending democracy). 21 B Ackerman, We the People: Foundations (Cambridge, Harvard University Press, 1991) 6.
The Right to Judicial Review: The Israeli Case 29
After identifying the values promoted by the Constitution, constitutional theorists identify the court as the institution which is most likely to realise successfully the consti tutional values. Various reasons including claims about the professional background of judges or the institutional structure of courts and their insulation from daily politics are provided to establish the conjecture that judges are likely to interpret the Constitution in a way that realises better the constitutional values. This part investigates one dominant instrumentalist theory. Perhaps the most domin ant instrumentalist theory maintains that courts make better decisions on issues con cerning individual rights because they are more protective of rights than the legislature.22 The conclusion of my discussion is: (1) the question of who makes better decisions, namely decisions that are more conducive to the constitutional values, has no clear answer. As a matter of fact the institutional competence of the court hinges on contin gent factors which cannot be universalised; (2) even if courts are better decision-makers with respect to individual rights, it does not follow that courts ought to be authorised to have constitutional powers over these issues. Let us start with some basics. There is not a single theorist who denies that individuals have rights and that the State ought to protect these rights. Opponents of judicial review do not dispute the fact that individuals have rights and that the State ought not to violate these rights; they dispute the claim that the best institution to do so is the court.23 The importance of individual rights in the governance of a liberal polity is not in itself suffi cient to justify judicial review. In addition one ought to establish that judges are better or more successful in protecting them than legislatures. Proponents of judicial review propose two possible explanations for the superiority of courts with regard to this task. Some theorists maintain that the superiority of judges is based on their professionalism or their expertise. Protecting rights requires legal exper tise. Legal expertise is acquired in a long process of thinking about rights and applying them.24 Other theorists maintain that the special ability of judges to protect rights is explained not on the basis of judicial expertise but on the basis of the distinctive struc ture of judicial institutions and, in particular, the relative independence of judges and their insulation from sectarian and political constraints.25 The legislature represents the majority and therefore promotes the interests of the majority even when promoting these interests violates minorities’ rights.26 The view according to which constitutional Harel and Kahana (n 10) 232. See, eg Waldron (n 1) 232 ff. 24 See, eg C Black, A New Birth of Freedom: Human Rights, Named and Unnamed (New Haven, Yale University Press, 1997) 125 (‘Human-Rights claims are made in the name of the law, as the outcome of reason ing from commitment; judges are practiced in this kind of reasoning, and some of them are expert at it’). This view is also endorsed by the Israeli Court itself. Thus the Court said: ‘The military commander is an expert of the military significance of the separation wall. We are experts with respect to its humanitarian aspects. The military commander determines where in the mountain and in the valley the separation wall should be built. This is his expertise. We determine whether the effects of this wall satisfy the requirement of proportionality. This is our expertise’. See HCJ 2056/04 Beit Sourik Village Council v Government of Israel 58(5) PD 807, 846 [2004] (in Hebrew). In another case the Court said: ‘Judicial review does not examine the wisdom to maintain military activity. The examination done by the review is an examination of the legality of the military activity. We assume therefore that the military activity in Rafah is from a military perspective necessary . . . We do not replace the judgment of the military commander with respect to military considerations. This is his expertise. We examine the results from the perspective of humanitarian law. This is our expertise’. HCJ 4764/04 Physicians for Human Rights v IDF Commander in Gaza 58(5) PD 385, 393 [2004] (in Hebrew). 25 See, eg n 9. 26 See, eg LH Tribe, American Constitutional Law, 3rd edn (New York, Foundation Press, 2000) 896. 22 23
30 Alon Harel constraints are designed to protect rights is, as Jeremy Waldron noted, the most domi nant view among contemporary constitutional proponents of judicial review.27 But a systematic examination of this argument exposes difficulties in this conventional view. To justify judicial review it is not sufficient to determine that the legislature is not sufficiently protective of individual rights. In order to generate this conclusion one needs two additional premises. First one ought to establish that legislatures are less protective of rights than courts. Second judicial review is costly in other respects as courts may wrongly protect concerns which ought not to be protected as rights. In other words judicial review may be detrimental because of judicial mistakes concerning the scope and weight of rights. The advocate of the instrumentalist/meritocratic justification of judicial review ought to establish that judicial mistakes (eg unjustified intervention in legislative decisions) are less dangerous or costly than the concerns resulting from violat ing human rights in a system which does not authorise courts to review legislation.28 Regrettably both historical and theoretical considerations do not support either of these conjectures.29 Historical evidence does not support the view that courts protect human rights better than legislatures.30 The famous case of Dred Scott illustrates that courts may misunderstand what rights are and what values they are designed to pro tect.31 The case of Lochner illustrates that courts can defend too fanatically what is wrongly perceived as rights and thereby undermine the legitimate pursuit of important social and economic goals.32 Perhaps this historical evidence led some theorists to chal lenge the assumption that judges are better than other institutions in protecting rights: Before accepting [the authority of the court] . . . it is necessary to ask about judicial competence to evaluate moral arguments of this sort, and also to ask about facts and incentives. Perhaps the Court is not especially well equipped to evaluate those arguments; and if consequences matter, the moral arguments might not be decisive.33
Theoretical examination also raises doubts concerning the alleged superior compe tence of courts in protecting rights. First, as Andrei Marmor argued, ‘the questions which the court is asked to settle in the context of constitutional law . . . are simply not Waldron (n 1) 11. See, eg Harel and Kahana (n 10) 234–38. 29 NK Komesar, Imperfect Alternatives: Choosing Institutions in Law, Economics, and Public Policy (Chicago, University of Chicago Press, 1997) 256–61 (pointing out weaknesses of courts in making decisions); Vermeule (n 16) 243 (establishing that often courts are institutionally inferior to other institutions). 30 See, eg DM Rabban, Free Speech in its Forgotten Years (Cambridge, Cambridge University Press, 1997) 131 (establishing that the court has not always been protective of freedom of speech). Waldron (n 1) 288 (ques tioning the assumption that courts are typically more protective of rights); W Sadurski, ‘Judicial Review and the Protection of Constitutional Rights’ (2002) 22 OJLS 275–78 (examining and questioning the competence and the willingness of courts in protecting rights). 31 Dred Scott v Sandford 60 US 393 (1857). 32 Lochner v New York 198 US 45, 76 (1905) (striking labour law legislation designed to protect workers). For a discussion of Lochner and the lessons of Lochner, see C Wolfe, That Eminent Tribunal Judical Supremacy and the Constituition (Princeton, Princeton University Press, 2004) 154 (explaining that the Court erred in Lochner by overextending rights protection beyond the provisions of the Constitution); WM Wieck, Liberty Under Law: The Supreme Court in American Life (Baltimore, The Johns Hopkins University Press, 1988) 123– 25 (‘Lochner has become in modern times a sort of negative touchstone. Along with Dred Scott, it is our fore most reference case for describing the Court’s malfunctioning . . . we speak of “lochnerizing” when we wish to imply that judges substitute their policy preferences for those of the legislature’). 33 Vermeule (n 16) 242; Sadurski (n 3) 299 (questioning the assumption that courts are better in protecting rights than other institutions); M Tushnet, Red, White and Blue: A Critical Analysis of Constitutional Law (Cambridge, Harvard University Press, 1988) 120 (identifying the mistakes made by courts). 27 28
The Right to Judicial Review: The Israeli Case 31
legal questions . . . but moral and political’.34 Both historical and theoretical analysis suggests that there is no reason to believe that judges have better ability to identify the scope of rights or their weight. Furthermore, many theorists argued that it is not only that judges have no superior ability to identify the scope of rights or their weight, but also that judges may be inferior in these respects as they belong to social or economic elites. Judges, in contrast to elected representatives, replicate the ideology of dominant classes or social elites and their cultural, ethnic background, and even their gender (pri marily male) distort their judgments.35 Legal feminism exposed the fact that the Israeli legal system suffers from gender bias and it does not sufficiently protect women.36 Furthermore, it was argued that the legal system may be too remote and detached and that the detachment of the [judiciary] from the public may distort its perceptions. Its independence may generate arrogance and undue preference for the institutional interests of the judiciary . . . the public will may often reveal hidden truths that are not known to the wise men who serve as judges.37
The detachment of courts is particularly problematic in the higher echelons of courts.38 Finally, some of the harshest critics of courts argue that the inevitable focus of courts on the words of canonical texts (eg the written Constitution) sustains and reinforces a formalistic rigid and legalistic reasoning in comparison to the more flexible, less rigid reasoning characterizing the legislatures. Courts rely in their reasoning on the canonical formulations of the Constitution, such as ‘due process’ or ‘equal protection’ in the American, or the concept of dignity in the German and in the Israeli Constitutions. The commitment to legalistic semantics and judicial scholasticism comes at the expense of serious moral deliberation that characterises (some) legislatures.39 But, even if, as a factual matter, we could establish the claim that courts are better than legislatures, there are grave doubts as to whether it follows that judges ought to have a veto right over the legislature’s decisions. Jeremy Waldron argued at length against drawing this inference.40 According to Waldron even if courts identify better the scope of the rights and their relative weight, judicial review violates the right to political participation. In a democratic society there is a bitter controversy concerning the con tent, scope and weight of individual rights. The right to political participation is the right to participate in the determination of the content, the scope, and the weight of individual rights. Hence one ought not to take the powers from the elected representa tives to determine the content, scope, and weight of rights even if judges make more accurate decisions than the legislature. Waldron’s view has been subjected to numerous
A Marmor, ‘Judicial Review in Israel’ (1997) 4 Law And Government 133, 154 (in Hebrew). The view is most forcefully voiced by K Marx and F Engels, The German Ideology (New York, International Publishers, 1970) 47. For a contemporary survey of this tradition, see ‘Law and Ideology’ (Stanford Encyclopedia of Philosophy, first published 22 October 2001) www.plato.stanford.edu/entries/law-ideology. 36 R Bogush and R Don-Yihyeh, Gender and Justice: Discrimination of Women in the Courts (Jerusalem, Jerusalem Institute, 1999) (in Hebrew). 37 See, eg G Sapir, ‘The Constitutional Judicial Proceeding as a Political Proceeding’ (2003) 19 Bar-Ilan Legal Studies 461, 476 (in Hebrew). 38 See also J Waldron, ‘The Core Case for Judicial Review’ (2006) 115 Yale Law Journal 1346, 1379–80. 39 See ibid 1381. 40 ibid 1386. 34 35
32 Alon Harel criticisms.41 Yet his view challenges successfully the instrumentalist meritocratic theory of rights. Last, I also said that this argument triggers justified resentment. Judges who strike down statutes on the ground that their judgments are better or superior presuppose their own superiority. Even if judges are indeed superior with respect to certain issues, for example issues concerning individual rights, it is difficult to infer from this fact that they ought to be given powers to decide these questions. Even if one proves beyond reason able doubt that the Lubavitcher Rabbi (the leader of an influential Hassidic community) or his many representatives are the best experts on the Israeli–Palestinian conflict, the citizens of Israel could not give him the power to make decisions concerning the resolu tion of the conflict. Of course they could elect him to be a political leader. But they could not deprive themselves of the power to replace him in the next elections. Granting him such powers undermines the view that we are all capable of reasoning, and that we all can, and should, make use of our powers of reasoning. Similar arguments can be made with respect to the courts. Even if a committee of rabbis proves beyond doubt that their intelligence, holy inspiration, or even the institutional structure of their institutions, guarantees greater accuracy and wisdom it would be inappropriate to grant them the power to make political decisions. The argument is not that the granting of powers to the Lubavitcher Rabbi or to a distinguished group of rabbis (or granting judges powers to protect individual rights) is incompatible with the foundational premises of liberal ism. The argument is only that granting such powers to them simply because they are better at making them is incompatible with liberal values. The mere superiority in one’s competence/ability to make certain types of decisions is not in itself an argument which can justify the authority to make such decisions. Ironically the meritocratic/instrumentalist argument is shared by both proponents and opponents of judicial review. Both proponents and opponents share the conviction that the only normative basis that can justify judicial review is the superior ability of judges to make the right decisions in specific areas such as individual rights, the protec tion of minorities and the protection of democracy. But, as we have shown, it is not only that there is little basis for the conviction that courts are in fact better than legislatures, but it is also the case that even if it could be shown they are better no conclusions could follow from it. Despite the failure of the instrumentalist argument I believe that judicial review is a fundamental component of any rights-based society. But the justification of judicial review is not an instrumental one and it has nothing to do with the prospects of reaching better decisions. The real justification of judicial review is grounded in the right to a hearing. Judicial deliberation concerning rights is in itself a realization of a distinctive privilege of right-holders: the privilege to challenge collective decisions reached by the polity, demand an explanation for these decisions and compel the government to rethink and reconsider its decisions. I will defend this view in Part III.
41 See, eg the symposium on Waldron’s ‘Law and Disagreement’ which was published in (2006) 39 Israel Law Review 13.
The Right to Judicial Review: The Israeli Case 33
III. JUDICIAL REVIEW AND THE RIGHT TO A HEARING
A. Introduction This section is divided into two parts. First I defend the right to a hearing as a basic right which ought to be respected. Second I argue that, as a conceptual matter, judicial review is the only way of honouring the right to a hearing. Judicial review is not therefore an instrument or a means to realise the right to a hearing; it is in itself the (only conceptu ally possible) realization of the right to a hearing. B. The Right to a Hearing Judicial review is designed to facilitate the raising of grievances by defending the right to a hearing. The right to a hearing consists of three distinct components: the opportunity to raise a grievance against what is perceived by an individual as a violation of his right, the duty on the part of the entity which is perceived (by the right-holder justifiably or unjustifiably) to be the duty-holder to provide an explanation/justification and the duty on the part of that entity to reconsider the decision which gave rise to the grievance in the first place. Note that I shall argue that the right to a hearing is intrinsically valuable; it is not designed to improve the quality of decisions rendered by the State. When and why do individuals have a right to a hearing? The right to a hearing presup poses a moral dispute concerning the existence of a right. One can identify two types of moral disputes. The first one is a dispute concerning the relative weight of the right, namely a dispute as to whether the right is overridden by conflicting considerations. In such a case the individual raising the grievance maintains that the violation of his right is unjustified given the circumstances and the entity against which the challenge is raised maintains that the circumstances justify such an infringement. The shared premise of both the entity which raises the grievance, and the entity against which the grievance is raised, is that the former has a prima facie right and the right to a hearing is designed to facilitate the rightholder an opportunity to establish that the infringement of the right is unjustified. The second type of moral dispute is the case where there is a genuine dispute concerning the existence of a prima facie right. The individual argues that there is a right that is being violated by a decision to act or to refrain from acting in a certain way. The right to a hear ing is designed in such a case to establish the existence of such a right. In both cases I will argue that the right to a hearing does not depend on the justifiability of the grievance. Even if the grievance is unjustified, the person raising the grievance has a right to a hearing. The dispute concerning the relative weight of the right involves a dispute as to whether there is a sufficient justification for the infringement of a right. In such a case both the entity which raises the grievance, and the entity against which the grievance is being raised, agree that the former has a right that was infringed. But the entity which raises the grievance thinks that the infringement is unjustified (and is therefore a violation) while the entity against which the grievance was raised thinks it is a justified infringe ment. Thus if, on my way to an appointment, I stop to save a child and, as a result, I am late to a meeting, the right of the person with whom I made an appointment is being (justifiably) infringed.
34 Alon Harel A careful investigation suggests that the person who raises the grievance in such cases may raise two distinct types of grievances. The first type is based on the claim that the infringement is an unjustified infringement, that is, a violation. The second type is a pro cedural complaint. When a person infringes a right, the victim may resent not the infringement itself but the fact that irrespective of whether the infringement is justified, the decision to infringe ought to be made in cooperation with the victim of the infringe ment. The victim of the infringement may therefore protest and say for instance: ‘you have no right to infringe my right without discussing or consulting me first’. The use of the term ‘right’ here supports the intuition that the victim of any (justified or unjustified) infringement of a right is entitled that the entity which makes the decision to infringe will give him the opportunity to raise a grievance, will explain its decision and will reconsider it on the basis of the grievance. This is not because the right to a hearing guar antees or even increases the prospects of a better decision. It is possible even that the hearing increases the chances that eventually a wrong decision will be made. Its justifica tion does not hinge on instrumental considerations. The right to a hearing in cases of infringement of a right presupposes the existence of a prima facie right when the debate is whether the right is overridden by conflicting con siderations. The right to a hearing is an indication that the powers of the right-holder do not disappear even when the right is being overridden. The right-holder who is provided with such a right is a partner in the making of the decision whether the prima facie right was overridden. The fact that the infringement is justified does not annul the participa tory privileges of the right-holder. The status of the right-holder as a participant in the deliberation is protected even when the right is overridden and, consequently, infringing the right to a hearing is wrong even if the right which the right-holder demands is being overridden. The right to a hearing consists of three components: an opportunity to raise the rightholder’s grievance; the duty of the entity against which this grievance is raised to provide an explanation and the principled willingness of that entity to honour the right if it tran spires that the infringement is unjustified.42 To establish the significance of these components think of the following example. Assume that Gideon promises Jeremy to meet him for lunch. Prior to the meeting Gideon asked Jeremy to cancel the meeting because of a memorial that takes place at the same time. Assume also that Jeremy believes that memorials do not justify the cancellations of lunch appointments. After all it could be argued that attending to the needs/interests of living people is more important than honouring the dead. It seems that even if Jeremy is wrong and memorials justify the infringement of prom ises, Gideon owes Jeremy ‘a right to a hearing’. His duty is not to apologise but a duty to hear patiently the grievance of Jeremy, to justify his decision to Jeremy, and finally to reconsider his decision given the moral deliberation. This right is not grounded in the fact that the hearing is more likely to bring about a better or a more just decision. The real justification for the right to a hearing is that Jeremy being a right-holder (of a right that may eventually turn out to be a prima facie right that is overridden) has to be a full partner in the deliberation of Gideon whether to infringe the right even when the decision to infringe the right is justified.
See Harel and Eylon (n 10) 1002 and Harel and Kahana (n 10) 238–39.
42
The Right to Judicial Review: The Israeli Case 35
Let us examine more closely the components of the right to a hearing in such a case. The first component – the opportunity to raise the grievance – is self-explanatory. The second and the third components of the right require some attention. To understand the nature of Gideon’s duty assume that Gideon announces to Jeremy that in the past (after long deliberation) he came to the conclusion that he ought always to follow the rule that in a case of a conflict between a lunch and a memorial he ought to go to the memorial. When Jeremy requires an explanation he simply reiterates the arguments leading him to adopt such a rule in the first place without examining the relevance and the applicability of these arguments to the present case. Such behaviour violates the right to a hearing and, especially, the second component of this right – the right of Jeremy that Gideon provides an explanation for the infringement. This right requires a concrete examina tion of the reasons underlying the decision. This is not because the original decision to adopt the rule is wrong or unjustified. It is possible that the considerations which led eventually to the adoption of the rule are impeccable and, it is possible that the best way to make a decision under these circumstances is indeed to blindly follow the rule. The duty to provide a hearing is not an instrumental duty designed to guarantee the quality of the decision or to increase the probability that the decision is correct, just, or appropriate. Last consider the third component – the willingness to reconsider the initial decision giving rise to the grievance. Assume that Gideon is willing to allow Jeremy to raise his grievance, and is also willing to explain his decision, but he announces in advance (or, even worse, decides without announcing) that the decision is final and will not be changed. It is evident that this is a violation of the right to a hearing. A real hearing requires willingness to reconsider the decision; it requires willingness to change it if it transpires that the decision is wrong and in particular it requires a willingness to act on the basis of the deliberation. Let us examine the second case mentioned above – the case where the dispute is not over the weight of the right, namely whether the infringement of the right is justified or not, but over the existence of the prima facie right in the first place. Assume that Gideon promised Jeremy that in the absence of special unexpected reasons, he will pick him up from the airport. Assume that a few hours before the flight Gideon calls Jeremy and says that due to a sore throat he cannot honour his promise. Given the qualification ‘in the absence of special unexpected reasons’ Gideon believes that Jeremy has no right (even not a prima facie right) to be picked up at the airport. Unlike the previous case, the debate between Gideon and Jeremy is not over the ques tion of whether conflicting considerations override the duty of Gideon, but over the question whether the phrase ‘in the absence of special unexpected reasons’ applies to this case or not. Gideon believes that a sore throat is a ‘special unexpected reason’ and, therefore, he believes that Jeremy has no right (not even a prima facie right) to be picked up from the airport. Jeremy disputes this claim and believes that a sore throat is not a special reason and therefore he believes that he has a right to be picked up from the air port. It seems that irrespective of who is right in this debate, Jeremy has a right to a hear ing, and it is wrong to violate this right irrespective of who is right in the substantive debate concerning the duty. Furthermore, Gideon’s duty to honour the right to a hearing does not depend on whether the hearing is conducive to reaching the right or correct decision as to whether Jeremy has (or does not have) a right to be picked up from the airport.
36 Alon Harel The structure of the right to a hearing in that case is similar to the structure of that right in the case where the dispute is over the justifiability of the infringement of a right. The right to a hearing in such a case consists of three components: first Gideon has to provide an opportunity for Jeremy to raise his grievance; second he ought to be willing to explain to Jeremy why he made the decision. Furthermore, it would be inappropriate to simply rely on a general principle under which any health concern (irrespective of how trivial it is) constitutes a ‘special reason’ which exempts Gideon of his duty without a concrete examination of the applicability of this principle to the circumstances of the dispute. Finally, Gideon ought not merely hear the grievance and provide reasons for his decision, but also reconsider his decision in light of this deliberative process. This example supports intuitively the claim that the right to a hearing is relevant, not only to cases involving the justifiability of an infringement of a right, but also to cases where the debate is whether there is a prima facie right in the first place. But it seems more difficult to justify the right to a hearing in such a case. In the case where the dispute is over the justifiability of the infringement we relied on the fact that the person who raises the grievance has a prima facie right. How can we establish the right to a hearing when there is a dispute as to whether there exists a prima facie right? What is the theor etical foundation which justifies the right to a hearing in such a case? If there is a right to a hearing in such a case it is grounded in the special status of rightholders, that is, in the fact that the entity which raises the grievance is a right-holder (even if not necessarily of the particular right that is the object of the dispute). Under this view entities which possess rights are entitled to have an opportunity to establish the justifiability of their grievances. Depriving them of such a right – the right to trigger a deliberative process – is unjust because depriving them of such a right does not honour their status as right-holders. A right-holder ought to be participating in the deliberation concerning the scope and weight of her rights. Thus, precisely as a prima facie right (even if it is overridden in a particular case) leaves a ‘normative fingerprint’ in the form of a right to a hearing, so a dispute concerning the existence of a prima facie right leaves ‘a normative fingerprint’ in the form of a right to a hearing even if, eventually, the ‘right’ at stake is ultimately found to be an imaginary one. If we apply our discussion to the constitutional context we may say that there are two cases which justify judicial review of legislation. The first case is when a person has a prima facie right and this right is (justifiably or unjustifiably) infringed by the legislature. The debate in this case is over the weight of the right and namely over whether the prima facie right is overridden or not. The second case is the case in which there is a debate as to whether there is a prima facie right in the first place. In both cases the right to a hear ing is composed of three distinct components. First, the State has a duty to provide an opportunity for the (purported) right-holder to raise his grievance, to explain his view and to justify it. Second, the State ought to provide an explanation why the grievance is flawed, that is, why the State’s decision is not a violation of a right. Third, the State ought to reconsider the decision in good faith on the basis of the deliberative exchange. Judicial review is not designed to improve the quality of the decision. It is unclear whether indeed it serves this function at all. The fact that this justification is a noninstrumentalist one immunises it from the objections raised against the instrumentalist approach discussed in Part II. But this does not imply that there is no link between the right to a hearing and the quality of decision-making. The right to a hearing presupposes that grievances concerning the violation of rights are taken seriously and that the
The Right to Judicial Review: The Israeli Case 37
relevant institutions examine these grievances in good faith. Furthermore, in examining the judicial procedures and the structuring of the relevant institutions the State ought to take into account instrumental considerations. The mode and the institutional mechan isms which realise the right to a hearing should be structured in a way which guarantees (as much as possible) the protection of rights. Instrumental reasons are not irrelevant in the design of relevant institutions. C. The Right to a Hearing and the Adjudicative Process So far we have defended the position that individuals have a right to a hearing. It is time now to explore the relations between the right to a hearing and judicial review. In what way can the right to a hearing justify judicial review? Can we substitute legislative review or other forms of non-judicial review for judicial review? This possibility challenges the theoretical distinction drawn between instrumental and non-instrumental considera tions. This is because the attempt to replace the conventional instrumentalist arguments (which are based on the degree to which judicial review is effective in protecting rights or democracy) with non-instrumental arguments (the right to a hearing) is based on the view that the right to a hearing is not an external goal or contingent by-product of judi cial review. Hence, under this view, it is impossible (as a conceptual matter) to realise the right to a hearing without guaranteeing judicial review. If it is possible (at least con ceptually) to replace judicial review with legislative review without affecting the right to a hearing, one could question why defend judicial review, as the right to a hearing could be protected by other means. And if the proponents of the right to a hearing defend judicial review merely on the grounds that courts protect (or are more likely to protect) the right to a hearing better than other institutions, the right to a hearing argument is as instrumental as any of the arguments discussed in Part II (and, consequently, it is vulner able to the objections raised there). In order to establish our argument that the right to a hearing is not justified on instru mental grounds one ought to establish that courts or, more precisely, adjudication is the only process that can realise the right to a hearing. Judicial review is not a better means to realise the right to a hearing; it is in itself the only realization of the right to a hearing. To establish this claim, I will argue that the judicial process is in fact at essence a process of a hearing. Of course institutions other than the court can grant a hearing but, if they do it well, they adopt in reality the basic components of the judicial process; they reason like courts. In other words, if they do it well, they thereby become judicial institutions. The more the realization of the right to a hearing is better, the more the process resem bles a judicial process (irrespective of what institution performs it). Think of the procedures used in courts. It is uncontroversial that courts (in contrast to legislatures) examine individual grievances.43 Examining such grievances involves three components which are known to us from the discussion concerning the right to a hear ing. First, the judicial process gives voice to the citizens’ grievances.44 Second, it imposes 43 A Bickel, The Least Dangerous Branch: The Supreme Court at the Bar of Politics (New Haven, Yale University Press, 1962) 173; D Horowitz, ‘The Judiciary: Umpire or Empire?’ (1982) 6 Law and Human Behaviour 129, 131; RH Fallon, ‘Reflections on the Hart and Wechsler Paradigm’ (1994) 47 Vanderbilt Law Review 953, 958. 44 See Mullane v Hanover Central Bank & Trust Co 339 US 306, 313 (1950); Boddie v Connecticut 401 US 371, 377 (1971).
38 Alon Harel a duty on the State or other entities to provide a justification for the decision which eventually gave rise to the grievance.45 Third, it also requires reconsideration of the deci sion.46 These are precisely the basic three components of the right to a hearing. The strong link between courts and the right to a hearing can be illustrated by examin ing the status quo of a court which fails to protect the right to a hearing. Such a failure is fundamentally different than the failure of a court to decide correctly. A failure of the second type does not undermine the status of the court as an adjudicative body. Judges, like any other institution, may make mistakes and still perform an adjudicative role. In contrast, judges who do not honour the right to a hearing do not function as judges. They do not act judicially because, after all, the essence of the judicial procedure is to provide an opportunity for parties to raise arguments, listen to the arguments, weigh them and decide in a way that reflects the relative weight and significance of the arguments. A sceptic may question at this point and ask whether it is not enough for the sake of realizing the right to a hearing to discuss the pros and cons of the decision in the legisla tive body. After all, our legislatures are asked again and again to justify their decisions. In addition to its legislative activity the Knesset (Israeli Parliament) is also an exciting debating society and the debates conducted there provide an opportunity for individuals whose rights are affected to voice grievances. This objection is misleading. Even if the Knesset examined grievances in an effective way this would not be sufficient to protect the right to a hearing. The distinctive feature of courts is not merely the facilitation of a hearing but the facilitation of particularised hearings – the right given to real as well as imaginary right-holders to raise their particu lar grievances. Gideon will not honour the right to a hearing of Jeremy if he merely asserts that he follows the rule that in every conflict between lunch and a memorial he chose to go to the memorial. Gideon has to hear the particular grievance of Jeremy and make a decision informed by the deliberative exchange with Jeremy. This cannot be the task of a legislature. It is true however, that courts are not necessary for protecting the right to a hearing; it is the adjudicative process rather than courts that is necessary. The right to a hearing requires the existence of an institution which conducts a hearing. It seems as if there is no reason to believe that the court is the only institution that can conduct a hearing. There is nothing in principle which prevents the legislature (namely the body which is responsible for the decision giving rise to the constitutional grievance) to conduct a hear ing and to allow a citizen to raise its particular grievance against the law, to listen to it attentively, to reason and to reconsider the decision giving rise to the grievance. The title or identity of the body which conducts a hearing is unimportant. The pri mary point is procedures conducted by this body and not the question of what body performs the hearings. The body which performs the hearing (irrespective of whether it is a legislature or a new body which is neither the court nor the legislature) ought to conduct procedures which are identical to those characterizing courts. In other words, such a body will have to provide an opportunity for individuals to raise their grievances, to address these grievances and to reconsider the decision giving rise to the grievance. ‘Non-judicial bodies’ such as the executive which provide for a hearing will mimic there 45 See, eg DL Shapiro, ‘In Defense of Judicial Candor’ (1987) 100 Harvard Law Review 731, 737; SC Idleman, ‘A Prudential Theory of Judicial Candor’ (1995) 73 Texas Law Review 1307, 1309. 46 See H Wechsler, ‘Toward Neutral Principles of Constitutional Law’ (1959) 73 Harvard Law Review 1, 19; SJ Burton, Judging in Good Faith (Cambridge, Cambridge University Press, 1992) 36–37.
The Right to Judicial Review: The Israeli Case 39
fore the procedures of the courts (and thereby lose their ‘democratic’ or ‘representative’ characteristics). The more these bodies are effective in performing a hearing the more they resemble the courts. They resemble courts so much that one can simply say that the right to a hearing requires the establishment of courts. After all a bird that looks like a duck, walks like a duck, and behaves like a duck is nothing but . . . a duck. An institution that is in charge of hearing grievances, providing a reasoned explanation and reconsider ing the decisions giving rise to the grievance is nothing but . . . a court. Assume that you adopt the explanation based on the right to a hearing. Can the right to a hearing justify judicial review? It seems that the review necessary to facilitate the right to a hearing is a minimalist review or more precisely a case by case review. The right to a hearing dictates that the individuals whose rights are violated or who claim that their rights are violated could raise their grievance, get an explanation and benefit from a reconsideration of the decision that gave rise to the grievance. But why should this decision oblige future cases? Why not have (as the Roman system had47) a case by case review which protects only the individual who raises the grievance without extend ing it to other individuals? Why should we extend the protection to other individuals who did not engage in the deliberative process (by striking down the statute)? It seems to me that there are compelling reasons why courts’ decisions ought to have implications which extend beyond the grievance of the specific petitioner. Considerations of certainty, predictability, and coordination provide general reasons why judicial deci sions ought to have a precedential value.48 The right to a hearing is just the beginning of an argument which justifies a case by case review. But a case by case review is incompat ible with other values of the legal system. Some may argue that the right to a hearing is incompatible with the practices charac terizing the Israeli legal system and, in particular, with the weakening of the requirement of standing and the rise in the activity of NGOs. The right to a hearing is the right of the person who maintains that his right was violated and not the right of human rights orga nizations representing the victim. But, as anybody who follows the practice of human rights litigation in Israel knows, this paradigm is very remote from the actual practice of the courts.49 The ‘public petitioner’ – the petitioner whose main concern is not the par ticularised violation of the rights of an individual but whose concern is a principled or ideological concern, namely to protect rights per se – is now a dominant feature of the Israeli legal system. My answer to this challenge has two parts. First, I am not committed to defend in this chapter each one of the doctrinal aspects of the Israeli legal system. It is possible that the right to a hearing cannot justify the institution of the ‘public petitioner’. Finally, I will argue that a possible defence of the public petitioner in terms of the right to a hearing 47 HF Jolowicz and B Nicholas, Historical Introduction to the Study of Roman Law, 3rd edn (Cambridge, Cambridge University Press, 1972) 12, 14. 48 GJ Postema, ‘Some Roots of Our Notion of Precedent’ in L Goldstein (ed), Precedent in Law (Oxford, Oxford University Press, 1987) 9, 15 (describing the rationales underlying the following of precedents in terms of ‘certainty and predictability of decisions . . . and in terms of utilitarian benefits of coordination of social interaction and respect for established expectations’). 49 See, eg HCJ 651/03 Association for Civil Rights in Israel v Chairman of the Central Elections Committee for the Sixteenth Knesset 57(2) PD 62 [2003] (in Hebrew) where Procaccia J writes: ‘The court expanded in its past decisions the standing of the “public petitioner” (NGOs). It did so in cases in which the issue at stake has a public character . . . Extending the status is part of a broad view that regards courts not merely as deciding over a conflict between parties but as being in charge of protecting the rule of law, even when this role does not involve resolving a conflict between two parties’ (para 7).
40 Alon Harel can be based on pragmatic considerations, in particular, the difficulty of access of ordin ary individuals to the legal system. These difficulties require the establishment of alter native mechanisms which facilitate human rights litigation. The conclusion of this part is therefore that individuals have a right to a hearing against the State and that the institutional embodiment of this right is judicial review. The only way to protect the right to a hearing is to establish adjudicative institutions designed to attend to grievances. Adjudication and the right to a hearing are conceptu ally interrelated. The link between the right to a hearing and the judicial process is not an instrumental link, but a conceptual one. IV. AND WHAT IF I AM A FANATIC INSTRUMENTALIST?
Judicial review is a way of facilitating a hearing to the (real or imaginary) victims of legislation. Such a hearing requires genuine willingness to reconsider the decision giving rise to the grievance and change it if necessary. Furthermore I argued that only adjudica tive process can (as a conceptual matter) protect the right to a hearing. The adjudicative process is not an instrument to maximise the realization of the right to a hearing; it is as a conceptual matter the only way the right to a hearing could be realised. My view deviates from the conventional defence of judicial review. It is often assumed that the opponents of judicial review have a non-instrumentalist argument against judi cial review based on the right to political participation while the proponents of judicial review must rely on instrumentalist concerns based typically on the superior quality of the decision-making of the courts. The argument based on the right to a hearing is a non-instrumentalist argument favouring judicial review and, in this respect, it deviates from standard arguments in favour of judicial review. But, for the sake of devoted proponents of instrumentalism, let me add that the voic ing of grievances may contribute greatly to the quality of decision-making, not because of the great wisdom of judges or the superior institutional structure of courts, but for the reason that the voicing of grievances may contribute to the quality of legislation. It is often too easy for the legislature to forget those whose life and rights are affected by the legislation. Judicial review provides an opportunity for the legislature as well as for the public to meet in the halls of the court the real worldly consequences of incompetent and unjust legislation. There is not a single enlightened legislative body that will not wish to encounter these implications in person. Such an encounter between the legislature and the citizens in the court is essential for the quality of decision-making not of judges but of legislators. Instrumental considerations touching upon the quality of the legislative decisions, and not those of judicial decisions, establish another compelling justification for judicial review.
4 The Purpose of the Israeli Constitution ARIEL L BENDOR
I. INTRODUCTION
T
HIS CHAPTER DISCUSSES the role of Israel’s Constitution – today, the Basic Laws – and the relationship between this role and judicial review. Constitutional law deals with the fundamental principles of the legal system, and in particular with the basic arrangements relating to the principal governmental branches and human rights. It is difficult to imagine a legal system which does not have basic principles, and consequently it is difficult to imagine a legal system without constitutional law. By contrast, it is not essential for constitutional law to be entrenched in a written constitution. There are countries, such as Great Britain, which do not have a written constitution, and the constitutional law of which is anchored in other legal sources, including common law and parliamentary statutes. Israel, like all other countries, has constitutional law. However, most Israelis, if questioned about the existence of a constitution, would deny it, and, apparently, so would most Israeli jurists. Certainly it is true that Israel does not possess a document that bears the title ‘Constitution’. At the same time, in practice, the Israeli Basic Laws form a constitution, even without being collated into a unified written document.1 This, in effect, has been the position held by the Israeli Supreme Court since its decision in the United Mizrahi Bank case of November 1995.2 Whatever the correct theory regarding the legal status of the Basic Laws is, the approach accepted in practice, not only by the Supreme Court but also by the Knesset – the ‘House of Representatives of the State’3 – is that the ranking of the Basic Laws in the hierarchy of Israeli legal rules is superior, from a functional and not only from a symbolic point of view, to that of ordinary parliamentary statutes.4 Accordingly, inter alia, first, a statute cannot annul or amend a Basic Law.5 Secondly, a statute cannot infringe a provision of a Basic Law or deviate from the contents thereof, unless authorised by a See, eg AL Bendor, ‘Is It a Duck? On the Israeli Written Constitution’ (2005) 6 Yale Israel Journal 53. CA 6821/93 United Mizrahi Bank Ltd v Migdal Cooperative Village 49(4) PD 221 [1995] (in Hebrew). In that case, the Supreme Court determined for the first time that the Knesset possesses constituent authority, and that Basic Laws enacted by the Knesset under this authority enjoy supra-legislative status. English translation available at: www.elyon1.court.gov.il/files_eng/93/210/068/z01/93068210.z01.htm. 3 As defined in Basic Law: The Knesset, s 1: ‘The Knesset is the house of representatives of the State’. 4 See, eg A Rubinstein, ‘The Knesset and the Human Rights Basic Laws’ (2000) 5 Law and Government 339, 351 (in Hebrew). 5 United Mizrahi Bank (n 2) paras 59–61 of Court President Barak’s opinion. 1 2
42 Ariel L Bendor Basic Law.6 Thirdly, a Basic Law can impose a duty on the Knesset to enact statutes.7 Fourthly, a statement in a statute, to the effect that the statute is valid notwithstanding the provisions of a Basic Law, cannot vest the statute with such validity. The only way to annul a right that is defined in a Basic Law is by changing the Basic Law itself.8 Fifthly, a Basic Law cannot be interpreted in light of the provisions of a statute.9 Sixthly, even though in practice the procedure for enacting a Basic Law is identical to that for enacting statutes, the Knesset is only empowered to entrench Basic Laws and cannot entrench ordinary statutes.10 In view of that set forth above, there is something tricky about the question as to whether Israel needs a constitution. This is because the question assumes, or at least hints, that Israel does not have a constitution. Indeed, some leading Israeli scholars have criticised the approach which holds that the Basic Laws have the status of a constitution.11 Nonetheless, today, about 17 years after the United Mizrahi Bank decision, the likelihood that the common approach will change seems small, in view of the fact that the Knesset itself enacts its legislation on the premise that it has the powers of a constitutive assembly and that the normative status of the Basic Laws is higher than that of ordinary statutes.12 In part II I will present the common attitudes – primarily as reflected in the Israeli Supreme Court case law – concerning the roles of the Basic Laws. These attitudes emphasize judicial review. The Israeli Constitution is not primarily perceived as guiding the branches of government, especially the legislative branch; rather, to a great degree, it is viewed as a basis for the criticism of the other branches by the judicial branch. This concept does not only reflect the intermixture of substantial constitutional law with the Court’s judicial review policy. It is also expressed in the unique Israeli phenomenon referred to in Israel as ‘rolling procedures’, which is characterised mainly by the fact that, in many cases, the Court does not issue its ruling in accordance with the existing legal and factual infrastructure, but rather, attempts to guide the state toward making required changes by way of a procedure which resembles negotiations and is likely to continue for years. In part III I will discuss the weaknesses of this conventionally common attitude, which, to a great degree, undermines the nature of constitutional law, and perhaps even that of law in general, as a sphere of knowledge, rather than a political arena; and which,
6 HCJ 212/03 Herut v Chairman of the Central Elections Committee 57(1) PD 750 [2003] para 4 of Court President Barak’s opinion (in Hebrew). English translation available at: www.elyon1.court.gov.il/files_ eng/03/120/002/A04/03002120.a04.htm. 7 HCJ 8300/02 Nasser v Government of Israel (22 May 2012), Nevo Legal Database (by subscription) paras 54–63 of former Court President Beinisch’s opinion (in Hebrew). 8 HCJ 4676/94 Mitral v Knesset 50(5) PD 15, 25 [1996] (in Hebrew). 9 HCJ 1384/98 Avni v Prime Minister 52(5) PD 206, 210–11 [1998] (in Hebrew). 10 HCJ 4124/00 Yekutieli v Minister of Religious Affairs (14 June 2010), Nevo Legal Database (by subscription) para 25 of Court President Beinisch’s opinion (in Hebrew). English translation available at: www.elyon1. court.gov.il/files_eng/00/240/041/n43/00041240.n43.htm. 11 See, eg R Gavison, ‘Constitutions and Political Reconstruction? Israel’s Quest for a Constitution’ (2003) 18 International Sociology 53, 62–67. Gavison claims that although the enactment of the Basic Laws relating to human rights was declared a ‘Constitutional Revolution’ by some, Israel’s constitutional history shows that such a declaration is mistaken; in the 1980s and 1990s attempts to enact a constitution were defeated because of the lack of broad social legitimacy, and this alone was the background for enacting the Basic Laws. 12 Rubinstein (n 4) 353–57. See also S 2 of Bill Memorandum of Basic Law: The Legislation, 2012, published by the Ministry of Justice: www.tazkirim.gov.il/Tazkirim_Attachments/41283_x_AttachFile.doc (in Hebrew).
The Purpose of the Israeli Constitution 43
in any event, also undermines a central justification of judicial review itself – its essentiality in maintaining the rule of law, including the rule of the constitution. In part IV I will discuss the role of the constitution as determining the legal rules which should not be subject to decision by a random political majority. These rules include the principles of the system of government, human rights and especially minority rights, and in Israel also basic values resulting from the basic reason for its existence – as a nation-state which expresses the right of self-determination of the Jewish people. Finally, in part V I will suggest that the purposes of the constitution require that the authority which will ensure that all of the branches of government act according to the Basic Laws shall be a professional entity, which does not depend upon majority representation. That authority is the Court. A vast body of literature, all over the world, focuses on the broad normative and theoretical aspects covered by this chapter. The chapter will accordingly concentrate on the Israeli discourse, and even in that sphere will seek to present only initial comments, each one of which is worthy of separate, detailed discussion. II. THE COMMON ATTITUDE – CENTRALITY OF JUDICIAL REVIEW
The jurisprudence of the Supreme Court in Israel, as in other countries, emphasises the role of judicial review in constitutional law. This is expressed in two principal ways. First, the rulings in constitutional matters are often based on institutional considerations, which deviate from the interpretation of the Basic Laws. Secondly, protracted handling of some of the constitutional decisions, in a manner known in Israel as ‘rolling procedures’, with a view to finding a solution for the dispute by amending legislation or enabling the accumulation of information on the practical impact of the disputed law. I will present these two aspects in the order listed above. A. The Integration of Institutional Considerations into Judicial Review The Israeli Supreme Court’s decisions indicate that judicial review is not only an instrument for resolving disputes with regard to the question of whether a statute enacted by the Knesset is consistent with the provisions of the Basic Laws. Rather, the settling of the differences is justified on the basis of considerations, some of which are of an interinstitutional nature, which concern the relationship between the non-elected judicial branch and the elected legislative branch.13 This attitude, which attributes significant weight to the status of the judicial branch and is not based solely on the interpretation of substantive constitutional rules set forth in the Basic Laws, is not limited to economic or social policies. The attitude also applies in additional contexts, such as legislation which is intended to promote the objectives of state security.14 13 See, eg United Mizrahi Bank (n 2) para 69 of former Court President Shamgar’s opinion and para 79 of Court President Barak’s opinions; HCJ 6304/09 Lahav v Attorney General (2 September 2010), Nevo Legal Database (by subscription) paras 62–63 of Justice Procaccia’s opinion (in Hebrew). 14 See, eg HCJ 7052/03 Adalah v Minister of Interior 61(2) PD 202 [2006] para 132 of Court Vice President Cheshin’s opinion (in Hebrew). English translation available at: www.elyon1.court.gov.il/files_eng/03/520/070/ a47/03070520.a47.htm.
44 Ariel L Bendor The emphasis on the role of the judicial branch is not only in the context of reservations against that role. At times, the role of the Court is presented in positive contexts, of special contribution to the protection of human rights.15 At the same time, in many cases, the Court does not distinguish between the question of the interpretation of substantive constitutional law and its own considerations in exercising judicial review.16 Indeed, the Court demonstrates theoretical ambivalence with regard to its role. If the tool is to preserve the criteria for the constitutionality of the statute, which are set forth in the Basic Laws and to prevent any transgression beyond their boundaries, why does it sentence itself to ‘restraint’ and ‘moderation’? What is the relationship between the policy of judicial moderation and the role of the Court which requires it to hand down decisions according to the rules set forth in the Basic Laws? The answer to this question, as the Court sees it, may apparently be found in the following sentences by Court President Dorit Beinisch: [T]he Court should carry out the role given to it in our constitutional system and examine the constitutionality of the legislation enacted by the legislative branch. This examination should be made by striking a delicate balance between the principles of majority rule and the separation of powers, on the one hand, and the protection of human rights and the basic values underlying the system of government in Israel, on the other.17
The meaning of the above statement is that the Court takes interest not only in the balance between human rights and the purposes of the violation thereto, as is prima facie required by the Basic Law on human rights, which state that: ‘there shall be no violation of rights . . . except by a law befitting the values of the State of Israel, enacted for a proper purpose, and to an extent no greater than is required’.18 In addition, the Court also refers to the balance between the need to protect human rights and the remaining basic substantive values of Israeli society, on the one hand, and, on the other, the principles which concern the relationship between the non-elected judicial branch – that is, the principles of majority rule and the separation of powers. That is likely to imply that constitutional judicial review is operated according to principles which are not identical to those which apply to the Knesset as the legislative branch. Whereas the Knesset, when operating as the legislative branch, is required to uphold the Basic Laws, and only those laws, the Court, while taking into account the substantive constitutional law set forth in the Basic Laws, also takes into consideration 15 See, eg HCJ 8192/04 Movement for Quality Government v Prime Minister 59(3) PD 145 [2004] para 8 of Justice Rivlin’s opinion (in Hebrew). The question in front of the Court was whether a criminal interrogation of the Minister of the Interior, obligates not only his suspension, but also his dismissal. In his opinion, Justice Rivlin declared that even if it were determined that there is no need for the Court intervention in this case, ‘The judicial review which intends to protect human rights will be as comprehensive as ever’. 16 See, eg HCJ 1715/97 Investment Managers Bureau v Minister of Finance 51(4) PD 367 [1997] para 20 of Court President Barak’s opinion (in Hebrew). The case dealt with the constitutionality of the Regularization of Occupation in Investment Consulting Law, 5755-1995. Court President Barak declared that the Court must restrain its judicial review and not put its judgment in place of that of the legislature; however, restraint does not mean stagnation, and if the legislature violates a protected constitutional right the Court shall take a clear position. 17 HCJ 2605/05 Academic Center of Law and Business v Minister of Finance (19 November 2009), Nevo Legal Database (by subscription) para 14 of Court President Beinisch’s opinion (in Hebrew). 18 Basic Law: Human Dignity and Liberty, s 8. English translation available at: www.knesset.gov.il/laws/ special/eng/basic3_eng.htm. See also Basic Law: Freedom of Occupation, s 4. English translation available at: www.knesset.gov.il/laws/special/eng/basic4_eng.htm.
The Purpose of the Israeli Constitution 45
principles which are external to the substantive constitutional rules set forth in the Basic Laws. Nonetheless, in certain case law, there is a tendency to combine – at least rhetorically – considerations of substantive law with considerations of separation of powers and majority rule, which are external to them. Thus, for example, Court President Barak wrote: In cases where a range of means exists, it is necessary to recognise the maneuvering power and the sphere of discretion of the legislators . . . The determination of social policy belongs to the legislators, and its implementation belongs to the Government, to both of which room for legislative maneuvering has been given.19
But these statements, which prima facie reflect interpretation of the Basic Laws as providing the elected branches with room for legislative manoeuvring, are justified by addressing the relationship between the roles of the legislative branch and those of the judicial branch, while ascertaining the need for special care and restraint on the part of the Court. Emphasis on the institutional aspect of judicial review – at times, in a way which does not indicate awareness of the substantive legal basis of the judicial review – is also characteristic of the public and media discourse on the activity of the Supreme Court. One example of this are the various proposals for constitutional changes, which were submitted in recent years pursuant to decisions of the Israeli Supreme Court, and which focus on defining the authority of the Court, rather than the content of the Basic Laws which the Court interpreted. Thus, for example, it was proposed to amend the Basic Law: The Judiciary, so as to include a provision that ‘the Supreme Court . . . shall not issue orders in security matters which are directly related to human life’20 it was also proposed to amend The Courts Law (Consolidated Version), 5744-1984, so as to increase the number of Supreme Court Justices, with a view to enabling ‘the appointment of Justices with a variety of backgrounds and positions, in such a way as to faithfully reflect the various world-views which prevail among the Israeli public’.21 A large proportion of academic writing in the constitutional sphere is also focused on ‘judicial review’, in such a way as to play down the professional legal aspects of constitutional law and to focus on the role of the judiciary. Thus, for example, in his recent book Gideon Sapir discusses a number of alternative constitutional models, the choice between which is presently to be made in accordance with considerations related to judicial review;22 and Eyal Benvenisti discusses judicial review policy, almost without addressing the Basic Laws on which it is founded.23 In an older article, Ruth Gavison calls for refraining from anchoring value-laden decisions in binding legal rules in Israel Investment Managers Association (n 16) para 19 of Court President Barak’s opinion. Draft Bill Amending Basic Law: The Judiciary (Powers of the High Court of Justice), 2010, P/18/2018 (laid before the Knesset on 25 January 2010, and denied by the Ministerial Committee for Legislative Affairs on 30 May 2010). 21 Draft Bill Amending the Courts Law (Number of Justices in the Supreme Court), 2009, P/18/1760 (laid before the Knesset on 16 November 2009, and denied by the Ministerial Committee for Legislative Affairs on 3 January 2010). 22 G Sapir, The Constitutional Revolution – Past, Present and Future (Tel-Aviv, Haifa University Press, 2010) (in Hebrew). 23 E Benvenisti, ‘Judicial Review and Democratic Failures: Minimizing Asymmetric Information through Adjudication’ (2010) 32 Tel-Aviv University Law Review 277 (in Hebrew). 19 20
46 Ariel L Bendor general, and Basic Laws in particular, principally so as not to transfer the decision in these matters to the Court and thereby to violate its status.24 B. Rolling Procedures: Monitoring by the Court of Amendment or Application of Laws The special Israeli phenomenon of rolling procedures – which, in one study, was referred to as ‘law as dialogue’25 – reflects an understanding of the role of the Court as transcending the mere examination of the compliance of a law with the requirements set forth in the Basic Laws. It should be noted that rolling procedures do not take place exclusively with regard to petitions directed against the constitutionality of statutes, but also with regard to petitions directed against decisions by the executive branch – whereby some of these petitions are based on constitutional arguments and others come from the realm of administrative law. One type of rolling procedures is that in which the Supreme Court, in cases where there is a prima facie problem with the constitutionality of a statute and the state expresses its willingness to examine the possibility of amending it, enables the state to make its examination while guiding it as to the amendments required and monitoring its progress on an ongoing basis, at times while waiting for the issuance of other rulings which have an affinity to the subject of the petition. Following, there is an example which, although extreme from the standpoint of the length of time during which the procedures kept rolling, are not unique or exceptional in nature. In the Yekutieli case,26 a petition was filed in 2000 against a section of the annual Budget Statute – which had been included in the Statute each year for a number of decades – which provided an income assurance allowance to Yeshiva27 students who devote their lives to religious study, although such an allowance is not paid to other students, such as college and university students. The argument advanced by the petitioners was that paying the allowance to Yeshiva students alone violates the principle of equality. The petition was granted, and the Court ruled that the payment of the allowance was not legal. The judgment, however, was issued in 2010, about 10 years after the petition was filed.28 After providing a detailed explanation of the chain of events over the years, and declaring that the complexity of the issues led to the lengthy examination of the factual and legal framework, the Supreme Court judgment, written by Court President Beinisch, stated as follows: Sometimes issues reach the court, the decisions for which require preparations that are different from the ordinary, inter alia, by giving the opportunity to the various entities, including the executive authority and the legislative authority, to examine the extent of their involvement in the matter and render the judicial decision superfluous. In the petition before us, the first years were devoted to clarifying the factual framework and determining the conditions for eligibility for the income support benefits, assuming that the benefits are, indeed, paid. Afterwards, adju24 R Gavison, ‘A Jewish and Democratic State: Political Identity, Ideology and Law’ (1995) 19 Tel-Aviv University Law Review 631, 661 (in Hebrew). 25 D Scharia, The Pure Sound of the Piccolo: The Supreme Court of Israel, Dialogue and the Fight Against Terrorism (Srigim-Li’on, Nevo Publishing, 2012) 201–10 (in Hebrew). 26 n 10. 27 ‘Yeshiva’ is a Jewish educational institution for learning traditional religious Jewish texts, primarily the Talmud and the Torah. 28 Yekutieli (n 10).
The Purpose of the Israeli Constitution 47
dication of the petition was delayed until the decisions were rendered in various petitions in the matter of deferral of military service.29
This dynamic attitude of the Supreme Court – which discloses interest in organizational and social developments which concern the implementation of a statute, the constitutionality of which is under examination, and attributes significance to those developments – also transcends the understanding of the role of the judicial branch as the examiner of the compliance of laws with constitutional rules, which, in Israel, are anchored in the Basic Laws. III. WEAKNESSES OF THE COMMON ATTITUDE
Notwithstanding the importance of judicial review, a constitution and Basic Laws, like most other legal rules, are not directed at the courts. Rather, they are primarily directed at the elected branches of government and at the individuals – the citizens and residents of the state. The unique aspect of the Israeli Basic Laws as legal rules which are superior to other laws lies in the fact that they are also primarily directed at the legislature: they prevent the Knesset from enacting certain statutes, or require it to enact others. As in the United States and Canada, the power to engage in judicial review in Israel is derived from the very jurisdiction of the courts – which is generally a mandatory jurisdiction – to decide legal disputes. One can argue that all these assertions are formal and technical, and that in fact both the Basic Laws and the theory regarding their supremacy over other laws are intended to enable judicial review of parliamentary statutes. Thus, judicial review was not intended to give effect to the Basic Laws. Rather, the Basic Laws were intended to enable judicial review. Nonetheless, for the time being, the principal practical impact of the Basic Laws which embody the ‘Constitutional Revolution’30 – that is, the two Basic Laws which relate to human rights – has been on Knesset legislation and not on judgments which have exercised judicial review of the constitutionality of statutes. Since the outbreak of the Constitutional Revolution the Supreme Court has only invalidated 10 statutory provisions. In contrast, since the United Mizrahi Bank decision, the constitutionality of all legislative bills has been meticulously examined by the legal advisors to the Knesset and Government and by the Knesset Members themselves. Bills which raise doubt as to their compliance with constitutional requirements are normally cancelled, delayed or redrafted.31 Moreover, certainly as a result of the influence of the Basic Laws, the Knesset has enacted a series of new statutes, in relation to which an effort has been made to respect ibid para 9 of Court President Beinisch’s opinion. See, eg YM Edrey, ‘The Israeli Constitutional Revolution: Evolution, Models of Constitutions, and a Lesson from Mistakes and Achievements’ (2005) 53 American Journal of Comparative Law 77, 85. Edrey writes that ‘The phrase [Constitutional Revolution] was intended to indicate that those Basic Laws introduced into Israeli constitutional law, both the Israeli Bill of Rights and the Concept of Judicial Review, granted the courts the authority to declare “ordinary laws” unconstitutional and null and void’. 31 See, eg ‘The Knesset Annual Report of 2011 according to “The Freedom of Information Law, 1998”’ (Jerusalem, June, 2012) (in Hebrew) main.knesset.gov.il/Activity/Info/Documents/report_2011.pdf, which describes the work of the legal department of the Knesset and states that one of its responsibilities is to examine whether bills that are brought to the Knesset table are in compliance with constitutional principles. 29 30
48 Ariel L Bendor human rights.32 These laws replace previous laws which disproportionately violated human rights. In cases where the Court invalidated Knesset statutes by virtue of the Basic Laws, the Knesset amended the statute in accordance with the guidance given in the judgment. At the same time, it is preferable to focus on the substance of the constitutional constraints on the legislature and the executive, not only from a practical point of view but also in terms of the strategy for determining the content and the interpretation of the Basic Laws and, in the future, of the Israeli Constitution. Systematic discrepancies between the rules of judicial review and the principles set forth in the Basic Laws are likely to give rise to various problems. This applies a fortiori with regard to the replacement of the constitutional rules by the rules of judicial review, or the obscuring of the distinction between the constitutional rules and the policy of judicial review, as happens more than occasionally in Israel and other countries, and as was illustrated in the previous part of this chapter. Thus, the common approach is likely to lead to a lack of clarity with regard to the nature of the limitations and duties which are imposed upon the legislative and executive branches. The starting point should be that these branches are interested in operating in a manner reconcilable with the constitution. The Court shall exercise the judicial review power by adopting an interpretation of the Basic Laws which focuses on the nature of the rules which apply to the elected branches, and not on the nature of the rules which apply to the Court itself. Such interpretation will enable the other branches, provided that they desire to do so, to operate according to law. First, it is quite possible, that the correct interpretation of the Basic Laws will allow the elected branches a broader range of discretion in matters which concern economic or security-related policy than in other matters. Accordingly, it is quite possible that the range of discretion which is given to the legislature and the executive, in matters related to human rights, is more restricted than the range of discretion given to them in other matters. At the same time, the interpretation of the Basic Laws should not be derived from considerations which concern the judiciary or its relationship with the other branches. Even if in the opinion of the Court there is justification, in certain cases, for not basing its decisions exclusively on the interpretation of the Basic Laws and their application to the facts of the case before it; even if it is appropriate for the Court to take constraints of the judicial branch into account as well, it must make a clear separation between the various aspects, in order to enable the elected branches and their legal advisors to know which constitutional limitations and duties are incumbent upon those branches. Secondly, a separation between the interpretation of the constitution and judicial review policies is required in order to enable a critique of the Court’s decisions in a manner which distinguishes between constitutional interpretation and considerations which concern the policy of judicial review. Thus, it is difficult to justify judicial review which is broader than required by substantive constitutional law, as the invalidation of a statute which does not run counter to a Basic Law is devoid of any legal basis and cannot be reconciled with the rule of law. It is also difficult for considerations of separation of powers to constitute a sufficient basis for requiring the judiciary to refrain from constitutional interpretation and from deciding whether the other branches have deviated See, eg Criminal Procedure Law (Enforcement Powers – Detentions), 5756-1996.
32
The Purpose of the Israeli Constitution 49
from the constitutional rules which apply to them. The power to decide legal questions, which concern the interpretation of existing law, clearly belongs to the judicial branch according to the separation of powers. On the other hand, the conclusion as to the legit imacy of the Court’s decisions may be different if the minimalism and restraint adopted by the Court are not derived from considerations which concern the relationship of the powers to each other, but rather, result from an interpretation of various constitutional limitations as minimal limitations. At the same time, procedural and evidential constraints, which are not parts of substantive constitutional law, as well as the recognition of the advantage which the legislature and the executive may have over the judiciary in clarifying facts and adjudging policy in complex situations, may justify the Court’s recognition of presumptions in favour of the constitutionality of legislative or executive decisions. This may be correct, provided that it has been proved that the legislative and the executive branches actually based their decisions on a sufficiently in-depth procedure, founded on the intention of respecting the rules of substantive constitutional law. Thirdly, mixing considerations of constitutional interpretation with judicial review policies is likely to give rise to errors in the Court’s decisions. These errors, in turn, are likely to have an undesirable impact on the method according to which the principle of binding precedent behaves. Thus, judges who hold before their eyes inter-institutional considerations, which are likely to justify the restraint of judicial review in the context of a certain right, are likely to formulate their rulings in such a way as to imply that the weight of that right is materially less. Moreover, judges might not carry out their examinations with the requisite degree of stringency if there really is sufficient justification for the outcome of the decision, not to comply with substantive constitutional law, but rather, to be biased by considerations or constraints of the judicial branch itself. Fourthly, and ironically: it is precisely the stressing of considerations which concern the policy of judicial review, without distinguishing between these considerations and substantive constitutional law – not to mention the replacement of substantive constitutional law by the policy of judicial review – which can prejudice the public status of the judicial branch. The public belief that the judiciary operates according to professional legal considerations and not according to the judges’ own political agenda may be prejudiced by a declared practice of underestimating the weight of substantive constitutional law. The phenomenon of rolling procedures is likely to empower the perception that the Court does not focus on the interpretation and application of substantive law, but rather, that judicial procedure is a blood-relative of political proceedings. Admittedly, one cannot rule out the possibility that the Court, in the appropriate cases, will encourage the parties to settle the dispute between themselves with no need for detailed judicial intervention – whether the respondents agree to grant the relief sought in the petition, or the petitioners agree to strike it out. There are also cases in which there is reason for the Court to ask the parties to clarify the factual or legal aspects of this or the other matter and to delay the continuation of the hearing until the results of the clarification have been obtained. In the appropriate cases, there is also nothing wrong with the Court, even if it has found that a statute is not constitutional, not to grant the extreme relief of retroactive invalidation of the statute, but rather, for example, to rule that the invalidation will apply prospectively or, in the case of a temporary statute, to deny the petition while giving the Knesset the possibility of correcting the flaws. At the same time, the monitoring of governmental procedures by the Court over a period of years is not in line with the essential role of the judiciary, which is to settle
50 Ariel L Bendor differences according to existing law. It contributes to the feeling that the judicial process is part of the procedure of the legislative or the executive branches, and that it entails an aspect of negotiation, in which the Court finds itself involved. This, in turn, gives rise to the possibility of an impediment to the internalization of the understanding that judgments of the Court reflect existing law, and that the duty of acting in accord ance with the law is not subject to negotiation or discretion – not even judicial discretion. Furthermore: protracted procedures which involve monitoring by the Supreme Court are likely to detract from the clarity, or even the legality, of legal doctrine. This is because the Court may not substantiate its proposals on an in-depth analysis of the law. The extent to which the arrangement which was eventually achieved corresponds to existing law is not always clear. Dragging out Court procedures over a period of years, with no interim order, is likely to give rise to a protracted breach of the law, which, on more than one occasion, has turned out to be irreversible, even if, at the end of the procedure, the Court allows the petition. This also applies to cases in which it was obvious, from the very beginning of the procedure, that the conduct of the governmental authority in question was unlawful. Thus, in the Nasser case,33 in which the Supreme Court kept the procedure ongoing for 10 years with a view to bringing about the amendment of a statute (although even the Attorney General was of the opinion that the statute was unconstitutional) it would have been proper for the Attorney General to instruct the authorities not to act according to the statute in question, or to ask the Court to declare the statute in question invalid. In any event, once the matter had been brought before the Court, it should have ruled on the question of constitutionality as quickly as possible. The Court, generally speaking, should rule according to the law even when the majority of the public will not favour the outcome or the grounds of the ruling. This holds true particularly in constitutional matters, in which the assumption is that the law whose constitutionality is under attack, reflects the wishes and interests of the majority of the public. The principal problem, then, is the very dominance, in judicial policy, of considerations which do not reflect an interpretation of substantive constitutional law. IV. THE PURPOSE OF THE ISRAELI CONSTITUTION
A. Three Models for a Constitution According to Gideon Sapir In his seminal book34 Gideon Sapir presents three possible models for a constitution, variants of each of which are implemented in various countries throughout the world. These models are referred to by Sapir as the ‘protection of basic values by the constitution and the Court’ model, the ‘constitution as a dialogue’ model and the ‘constitution as a silencing agent’ model. Sapir did not invent these models. His writing is based on an in-depth study of the constitutional law of many countries and much literature on polit ical theory. Nonetheless, his analysis will provide me with a useful starting point for discussion. Nasser (n 7). Sapir (n 22). Sapir also discusses the three models in G Sapir, ‘Three Models of Constitutionalism’ (2007) 37 Mishpatim 349 (in Hebrew) and in his chapter in this volume. 33 34
The Purpose of the Israeli Constitution 51
The first model is the most common one. According to this model, the purpose of the constitution is to protect basic values against infringement. The constitutional fettering is required, notwithstanding the agreed nature of these values, in order to cope with ‘states of intoxication’, in which the public would lose control of the requisite discretion. The second model is based on the fact that the last word in constitutional questions belongs to the public, through its elected representatives. However, the judiciary is provided with tools which enable it to urge the legislature to discuss and settle these questions seriously. The third model is based on the anchoring in the constitution of detailed compromises, which do not necessarily reflect agreed basic values. According to this model the purpose of a constitution is to remove from the day-to-day agenda certain disputes, the discussion of which constitutes a threat to social stability. B. Israel and the Three Models Generally speaking, the prevailing model in Israel today is the first model – the protection of basic values by the Basic Laws and the Court. Admittedly, most of the Basic Laws are not rigid; a special majority of the Knesset Members is not required in order to amend or even repeal them. Even when a special majority is required, it is a majority of the Members of the Knesset, which the Government – the executive branch – can prima facie summon up. However, there is a political difficulty in amending Basic Laws, since, for many years, the coalition agreements have specified that no Basic Law can be enacted or amended without the consent of all of the member parties in the coalition.35 Furthermore, there is an understanding to the effect that the incorporation of the Basic Laws into a complete constitution will take place pursuant to a broad (even if not an absolute) consensus. The Basic Laws, even though they have not yet been incorporated into a constitution, include fundamental principles of governance and some of the major human rights, and are perceived by the Supreme Court as superior to ‘ordinary’ statutes. The Knesset and the Government recognise this perception. There is judicial review of the congruence between the rules set forth in statutes and the rules set forth in Basic Laws. In the event of a contradiction between a Basic Law and a statute, the Court may invalidate the statute. A similarity between Israeli law and the ‘constitution as a dialogue’ model36 is found in section 8 of Basic Law: Freedom of Occupation, which enables the Knesset to enact statutes which detract from freedom of occupation without complying with the conditions set forth in the Basic Law for justified violation of the right, if the statute was adopted by a majority of the Knesset Members – the same majority that is required in 35 One interesting example is that of a statute, which states that any withdrawal from territories to which the law and administration of Israel apply may take place, as a general rule, only if supported by the majority in a referendum. See, Law and Administration Law (Cancellation of Application of Law, Jurisdiction and Administration) (Amendment), 5771-2010, SH No 2263, p 58. An attempt to establish an identical arrangement in a Basic Law did not succeed for many years. 36 Another dialogue-type aspect, which Sapir does not mention, is the rolling procedures described above, which, in fact, constitute a dialogue between the Court and the elected branches. At the same time, by contrast to the ‘constitution as a dialogue’ model, the last word regarding the constitutionality of a statute, when the procedure finally stops rolling, belongs to the Court and not to the elected branches.
52 Ariel L Bendor order to amend or repeal the Basic Law itself – and if the statute expressly states that it is valid notwithstanding that set forth in the Basic Law. The validity of such a statute may not exceed four years – one ‘political generation’.37 C. Difficulties with Sapir’s Approach As to the first model, although there are various interpretation theories which are intended to guide the judges in interpreting the constitution,38 Sapir insists that the adoption of these theories depends upon the good will of the judges themselves. Moreover, the other solutions proposed are not effective, or else they ‘throw out the baby with the bathwater’ – miss the target of constitutional protection of basic values. Sapir accordingly prefers a combination of the ‘constitution as a dialogue’ model, in which the last word will be left to the elected branches, with the ‘constitution as a silencing agent’ model, which anchors in the constitution a number of detailed compromise arrangements regarding issues which divide the public.39 A definite empirical difficulty in Sapir’s approach towards the first model is reflected in his argument to the effect that, in a pluralistic society such as that which characterises all of the Western countries, it is not possible to reach a consensus in determining the basic values to be anchored in the constitution, and that, in any event, it is ‘not clear how it is possible to overcome the barrier of non-consensus and to adopt a constitution’.40 And yet, the first model is the most common model in the Western world.41 Moreover, even though the Israeli Basic Laws have not yet been incorporated into a constitution, this model, or at least the essence thereof, is in force in Israel as well. As Sapir himself admits, precisely the feasibility of adopting the other two models, which he supports, is not high.42 However, the fundamental difficulty with all the three models which Sapir presents does not lie in the feasibility of the models. The most problematic aspect has to do with two assumptions. The first assumption is that the basic values actually constitute an empty idea, and that, as Sapir himself puts it, ‘even if we agree that there are values of supreme importance’,43 it is not possible to reach a consensus as to the nature of those values. In any event, in the absence of a consensus as to which values are deserving of constitutional protection, and certainly as to the manner in which they are to be protected under given circumstances, judicial review is left as an empty vessel, which fills up with the political and ideological views held by the judges themselves (according to the first model) or by the elected representatives of the public (according to the other two models). The second assumption, which is also common to all three models, is that the values of the constitution in general, and their implementation under given circumstances in particular, must be consensual. Both of these assumptions are problematic. The first assumption underlies the criticism of the first model, according to which this model involves democratic difficulty in detail in empowering non-elected judges to make In Israel elections are held every four years. Sapir (n 22) 190–96. 39 ibid 254–61. 40 ibid 174. 41 ibid 172. 42 ibid 262–63. 43 ibid 174. 37 38
The Purpose of the Israeli Constitution 53
political decisions, including invalidating the decisions of the people’s elected representatives. Within the framework of this model, and in order to cope with his difficulty, John Hart Ely developed his famous theory whereby judicial review is only justified for the purpose of safeguarding the democratic character of the political process.44 Likewise, the minimalist approach, whereby the courts can only intervene in the decisions of the other governmental branches in extreme cases, focuses on the function of the Court and not on the determination of the constitutional constraints on the legislature and the executive. The focus on this assumption, to the effect that constitutional law is devoid of any substantial content, and that it principally serves to provide power for judicial review, also underlies the other two constitutional models of which Sapir writes. This is true of the ‘constitution as a dialogue’ model. Here, the purpose of judicial review is to ensure a dialogue between the various governmental branches and consequently provide a catalyst for profound debate on constitutional questions. This also applies to the ‘constitution as a silencing agent’ model, the primary purpose of which is to prevent the Court from interfering in areas where the silencing of acute public disputes is sought. This line of thought, however, encompasses a double failure. First, it is itself based on the values of democracy. But if there is no objective essence to basic values, reliance on democratic arguments also becomes meaningless. Secondly, the assumption that values are devoid of any objective content, and that the significance of the constitution is approval for the authorities authorised to settle constitutional differences to dictate their own values, undermines the rule of law in its entirety. In addition, it is not in line with reality, in which great influence, which transcends decisions made by judges is attributed to legal rules, including the rules of constitutional law. Indeed, legal rules in general, and constitutional rules in particular, do not always have an agreed meaning, and disputes are especially likely to arise with regard to the outcome of their application under specific given circumstances. At times, a decision which concerns interpretation or application entails a decision which has to do with values. No one can deny that such a decision is likely to be affected, deliberately or unintentionally, by the values or even the personal interests of those empowered to decide. This, however, does not mean that constitutional law, or law in general, is not a substantial professional sphere, or that the application of the constitution is always tantamount to the application of the individual wishes of the holders of judicial authority – with no affinity whatsoever to substantive legal criteria. Also problematic is the second assumption which is embodied in the three models discussed by Sapir and the criticism thereof, according to which the consensus of at least a large majority is an essential part of the nature of the constitution, and of the nature of decisions on the constitutionality of specific legal rules. In this context as well, the very fact of reliance on democratic values – which does not result from the consensual nature of those values – makes it obvious that at least those values can, and even should, be anchored in the constitution even without a consensus. In fact, in Israel it cannot be assumed that an entire constitution would be accepted without a broad consensus, nor should it be. At the same time, most of the existing Basic Laws were adopted with a clear majority, and in any event, as set forth above, there is currently no impediment to finding a majority in order to repeal them. 44 JH Ely, Democracy and Distrust: A Theory of Judicial Review (Cambridge, Harvard University Press, 1980).
54 Ariel L Bendor Whichever way, the absence of a consensus, in principle, does not apply to the provisions set forth in the Basic Laws, but rather, to the manner in which they are implemented. This, though, is the entire essence of a constitution by the first model – to protect the stability of the foundations of government and human rights against the will and the interests of the majority, and especially against the will and the interests of a fleeting majority. The absence of a consensus with regard to the implementation of the constitution – and, more precisely, situations in which the will of the majority is overcome by constitutional rules – does not constitute a constraint. They are the principal reason, if not the entire reason, for a constitution. The argument which holds that the purpose of the constitution and judicial review pursuant to it is to cope with ‘states of intoxication’ is an arrogant one. It is rooted in the assumption, the basis of which is unclear, that the elected representatives of the public do not know what is good for the public or for most of it, and that the judges know better (according to the first model), or that the judges should act as a catalyst – a sort of whip – for more serious and profound discussion than the elected representatives would hold (according to the second model). D. The Suggested Approach My approach is that the purpose of Israel’s constitution – and this is also the purpose of most of the existing Basic Laws – is to confer a lofty normative status, which is not easily vulnerable to harm by the majority, upon the basic values of Israel. These are its values as a Jewish and democratic state, values which are rooted in Israel’s Declaration of Independence.45 The fact that Israel is a national home for the Jewish people, its democratic regime and the ambition to achieve a stable system of government must constitute the very centre of the Constitution of Israel. The existing Basic Laws do not completely define those values. Nonetheless, the Basic Laws can – and were intended to – constitute a basis for an entire constitution. The rules which are currently set forth in Basic Laws, and, as we may reasonably expect, those which will be set forth in the future in a complete constitution, have substantive content which enables both the branches of government and the citizens and residents of the state to act ab initio in accordance with them, at least on the assumption that they receive legal guidance, and settlement of differences as to their application even after the fact. I shall illustrate this by means of a few cases in which the Supreme Court invalidated statutes by virtue of Basic Laws. In four cases, the Supreme Court has invalidated statutes which governed the Knesset elections, because these statutes were in contravention of the principal of equality in the elections, as set forth in section 4 of Basic Law: The Knesset, and were not enacted with the majority required by that section. In the Bergman decision46 a statute which conferred financing by the state (for the purposes of the elections) only upon parties which were represented in the Knesset at the time of enactment of the statute, and not to new parties which were proposing candidates for the first time, was invalidated as not egalitarian. In the Agudat Derekh Eretz decision47 a statute was invalidated because it dis Declaration of the Establishment of the State of Israel, 1 LSI 7 (1948). HCJ 98/69 Bergman v Minister of Finance 23(1) PD 693 [1969] (in Hebrew). 47 HCJ 246/81 Agudat Derekh Eretz v Broadcasting Authority 35(4) PD 1 [1981] (in Hebrew). 45 46
The Purpose of the Israeli Constitution 55
criminated against new lists and small lists of Knesset candidates, in favour of large lists, in allocating time for radio and TV election propaganda broadcasts. In the Rubinstein v Chairman of the Knesset decision48 a statute which retroactively applied new arrangements concerning sanctions imposed upon Knesset lists which exceeded the maximum permitted amount of expenses, so as to benefit lists which had exceeded the maximum permitted amount in elections which preceded the enactment of the statute over lists which had taken pains to act lawfully, was invalidated as not egalitarian. And in the Laor decision49 a statute which had not been adopted by a majority of Knesset Members, and which raised the amount of financing for the Knesset elections after the elections had already taken place, was invalidated as not egalitarian. In none of the above cases did the decision entail a substantive value component which involved the identification of basic values. Basic Law: The Knesset determined that Knesset elections must be egalitarian. This provision has a rather clear substantive content, the interpretation of which does not embody any special difference of opinion with regard to values. In light of the fact that ordinary statutes, which did not comply with the requirement of equality, had not been enacted with the majority required in the Basic Law, the statutes in question deserved to be invalidated. In 10 cases the Court has invalidated statutes pursuant to Basic Law: Human Dignity and Liberty or Basic Law: Freedom of Occupation. In the majority of those cases as well, the decision did not necessitate any significant consideration of values. Thus, in the Investment Managers Bureau decision50 the Supreme Court invalidated as unconstitutional a section of a statute which governed the profession of investment consultancy because the section in question imposed – as a prerequisite for continuing to practice the profession – a sweeping obligation to examinations, which had not previously existed, with regard to professionals who were already proficient in that field (unless they had worked in it for more than seven years), who had acquired experience in it and had come to rely on the fact that they could continue practicing their profession. The Court found that the imposition of such a sweeping duty is disproportionately prejudicial to freedom of occupation and is therefore in contravention of the Basic Law: Freedom of Occupation. In the Zemach decision51 the Supreme Court invalidated as unconstitutional a section of a statute which enabled soldiers suspected of having committed an offence to be detained for four days before being brought before a judge. The Court ruled that a detention period of four days exceeded the amount of time necessary to bring a detainee before a judge, and that, accordingly, the restriction of the right of freedom from detention, as set forth in section 5 of the Basic Law: Human Dignity and Liberty, was in contravention of section 9 of the Basic Law, which stated that [t]here shall be no restriction of rights under this Basic Law held by persons serving in the Israel Defence Forces . . . nor shall such rights be subject to conditions, except by virtue of a statute . . . and to an extent no greater than is required by the nature and character of the service.
The factual conclusion reached by the Supreme Court – that four days exceeded the amount of time necessary to bring a detainee, even if that detainee is a soldier, before a judge – did not necessitate any significant consideration of values. HCJ 141/82 Rubinstein v Chairman of the Knesset 37(3) PD 141 [1983] (in Hebrew). HCJ 142/89 Laor Movement v Speaker of the Knesset 44(3) PD 529 [1990] (in Hebrew). 50 Israel Investment Managers Association (n 16). 51 HCJ 6055/95 Zemach v Minister of Defence 53(5) PD 241 [1999] (in Hebrew). 48 49
56 Ariel L Bendor Indeed, there can be no doubt that cases are possible in which a constitutional decision requires judgment with characteristics which are related to values, to which the Basic Laws do not provide a response. An example of such a case is provided by the Academic Center of Law and Business decision.52 In that case, the Supreme Court invalidated in a majority ruling a statute which enabled the operation of a prison by a private concessionaire on a for-profit basis. The Court ruled that the granting of powers of incarceration to business persons violated the inmates’ constitutional rights of human dignity and liberty, in a manner which could not be justified, even if it were proved that the statute would give rise to an improvement in the well-being of inmates who were incarcerated in a private prison, relative to the well-being of inmates incarcerated in public prisons.53 This decision, the grounds for which included considerations and contemplations from the realm of political philosophy, is not self-evident and certainly does not necessarily result from that set forth in Basic Law: Human Dignity and Liberty. In fact, the minority judge, Justice Edmond Levy, believed that it was appropriate to allow the operation of the private prison for a number of years, in order to determine whether it would actually lead to an improvement in the well-being of its inmates.54 At the same time, the fact that the constitution, by its very nature, is characterised by a certain degree of vagueness does not and cannot render it meaningless or valueless, and does not and cannot justify denying the very existence of constitutional law as a realm of legal knowledge. In fact, the Knesset, in its role as a constitutional assembly, can consider whether it is indeed appropriate to determine different criteria for restrictions of different constitutional rights, as in the German Constitution. Nonetheless, the possibility and the necessity of improving the wording of the Basic Laws – and this will probably be done when the time comes for them to be incorporated into a complete constitution – does not and cannot rule out the very value of the constitution, and of constitutional law, in preserving the basic values of society. When the Basic Laws are interpreted – whether by the Court or by any other legal advisors or commentators – special attention must be paid to the fact that legal limitations restrain the elected branches of government, including the legislative branch. Accordingly, in cases of doubt, the question is whether it is justified for a certain matter not to be subject to regulation within the framework of the ordinary political process. Such a justification exists only when the imposition of constitutional limitations is required in order to preserve the existence of Israel as a Jewish and democratic state, or to preserve a stable system of government. This is true of the interpretation of the Basic Laws and should also apply to the wording of a complete constitution. There is also a doubt as to whether the constitution should properly include ‘silencing rules’, as Sapir suggests. It is highly problematic to establish such rules within a rigid constitution, insofar as they do not reflect the fundamental basic values of society. This is because, notwithstanding the advantages pointed out by Sapir, it is difficult to find any democratic justification for preventing the elected branches of government from freely deliberating and deciding on matters which are not essentially constitutional by nature. Indeed, excessive rigidity, including that of constitutional laws, is hazardous. It is likely to immortalise errors which may occur in formulating the constitution, to lead to the eternal imposition of archaic values, and even to enable arbitrariness on the part of Academic Center of Law and Business (n 17). ibid paras 33–54 of Court President Beinisch’s opinion. 54 ibid para 11 of Justice Levy’s opinion. 52 53
The Purpose of the Israeli Constitution 57
the entity in charge of interpreting, implementing and enforcing the constitution – whether that entity is the Court or any other authority. Accordingly, favourable consideration should be given to the possibility that, at least in the first years after the establishment of the constitution, excessively burdensome requirements should not be imposed on its amendment, and that it should include an ‘override clause’, in the spirit of the existing section 8 of Basic Law: Freedom of Occupation, which will enable the branch which will be authorised to amend the constitution, by means of the same procedures and with at least the same majority as that which will be necessary for the approval of an amendment, also to allow a temporary, local deviation from the provisions of the constitution.55 V. INSTEAD OF A SUMMARY – A RETURN TO JUDICIAL REVIEW
In this chapter, I have criticised the replacement, which is frequent in judicial and academic discourse – in Israel and many other countries – of substantive constitutional law by judicial review. Indeed, Thomas More’s Utopia56 is the only place which, although it has legal rules, has no judges or other persons competent to settle disputes concerning rights and duties. Israel is not Utopia. With no judges, there can be no law. In the absence of constitutional judging, it is doubtful whether substantive constitutional law will be upheld. To whom should the task of judging be assigned? Constitutional law, perhaps more than other branches of law, is directly concerned with ideology. The normative superiority of the constitutional rules over other legal rules, including statutes which, as a general rule, cannot deviate from the constitutional rules as interpreted and implemented by the competent authority, gives rise to the suspicion that those in authority – deliberately or unintentionally – will mix their own beliefs, and even their own personal or status-related interests, into their rulings, with no simple possibility of correcting their errors. This is the source of the proposal for the democratization of constitutional law – whether by giving the power to decide on constitutional questions to the legislative branch itself – or by democratic election of the judges and restriction of their terms in office. Nonetheless, the disadvantages of such a system outweigh the advantages, because the essence of constitutional law is the imposition of restrictions on the exercise of the majority’s desires and interests. Giving jurisdiction in constitutional matters to the representatives of the majority, precisely because they represent the majority, will frustrate the reason for the existence of constitutional law. Indeed, judicial review exercised by professional judges who are not elected democratically, but rather, selected on the basis of their skills, qualifications and professional experience, has its own disadvantages. At the same time, the fear that there will be cases in which such judges administer justice according to their private values cannot compare to the certainty that constitutional law, in its entirety, will be frustrated if jurisdiction in constitutional matters is given to the elected representatives of the majority. The way of 55 To date, the authority to deviate from the provisions of Basic Law: Freedom of Occupation has been used only once since 1994. See Meat and Meat Products Law, 5754-1994. Furthermore, in Canada – which, as set forth above, has a similar provision – the provision in question has been used since 1985 only a few times. 56 T More, Utopia (New York, Norton Critical Edition, 1992).
58 Ariel L Bendor coping with the certain amount of fear aroused by professional constitutional judging is by establishing an override clause, similar to the clause which currently appears in Israel’s Basic Law: Freedom of Occupation, which will enable the same majority as that required for an amendment of the constitution to deliberately deviate from it for a single political generation – the ordinary term in office of the Knesset. True, this solution is not perfect. Nonetheless, it is the least of all evils. Only professional judicial review is capable of promoting – though perhaps not of accomplishing in their entirety – the purpose of the Israeli Constitution.
5 Consensus, Dissensus, and Constitutionalism SANFORD LEVINSON*
I. THE CIRCUMSTANCES OF CONSTITUTIONAL FORMATION
O
N 23 DECEMBER 2012, a New York Times story reported on the significant discontent in Egypt regarding the approval in a national referendum of a new constitution for that country.1 One objection involved the relatively low turnout of approximately 30 per cent of the electorate – though a significant majority of those who voted supported the document – and the diminished legitimacy thought to attach to the failure of a substantial majority even to participate. Far more important, for purposes of this comment, is a statement by Hamdeen Sabahi, described as ‘a leftist and former presidential candidate’. Describing the Constitution as illegitimate and calling on people to organise in order to prevent its implementation, Mr Sabahi said, ‘This is a Constitution that lacks the most important prerequisite for a constitution: consensus. This means we can’t build our future based on this text at all’. Why do I begin my own commentary on three essays on Israeli constitutionalism with a reference to what is occurring in Israel’s neighbour? The answer lies not in any direct comparison between the socio-political realities of Egypt and Israel, even as that might be an interesting exercise in some respects, but, rather, in the weight given by Mr Sabahi to the importance of ‘consensus’ and the suggestion that any constitution drafted in such a way as to limit attainment of a consensus is necessarily doomed to fail. Indeed, he seems to suggest that such a constitution deserves to fail. There is little doubt that the Egyptian Constitution was drafted by persons beholden to the Muslim Brotherhood, with relatively little attention being paid to those who do not share its convictions, whether secularists, Christians, or more moderate Muslims. No doubt there are many things to object to in the new document; it is important, though, that Mr Sabahi’s critique, at least as described in the New York Times article is directed less at its specific deficiencies than at the fact that it manifests a lack of consensus. He does not address the obvious point that a more secular liberal constitution (which I would greatly prefer) * I am very grateful for the opportunity to have participated in the conference at which these papers were initially delivered and for the hospitality provided by the organizers of the conference. 1 See D Kirkpatrick and M El Sheikh, ‘Egypt Opposition Gears Up After Constitution Passes’ New York Times, 23 December 2012 www.nytimes.com/2012/12/24/world/middleeast/as-egypt-constitution-passes-newfights-lie-ahead.html?_r=0. For the Constitution see Constitution of the Arab Republic of Egypt, 26 December 2012.
60 Sanford Levinson would almost certainly have not received the support of the Brotherhood and, therefore, been just as lacking in whatever attributes define ‘consensus’ as the one they drafted. One can well wonder how likely it is, as a matter of empirical fact, that constitutions usually meet these attributes, at least if one rejects the circular proposition that ‘consensus’ exists whenever disparate groups agree, for whatever reasons, to agreement to any particular bargain, whether a constitution, peace treaty, or more mundane proposal. After all, many such agreements reflect only the belief that acceptance of the deal, with whatever anguish, is better than rejecting it. Beyond that, it may be very hard to discern any real ‘consensus’ about the desirability of specific terms. Constitutions frequently reflect the often dramatic circumstances – defeat in a catastrophic war, the end of a civil war, the dissolution of empire, or, as in Egypt, the overthrow of a long-established authoritarian ruler – that trigger a new play of political forces determined to create a new order that conforms to their own visions of a desirable polity. The phenomenon of ‘imposed’ constitutions is scarcely unknown; that term could easily be used to describe the two major constitutions that followed defeat in World War II of Germany and Japan, both of which have, by most ordinary terms, been quite successful, especially if one realises that most constitutions have a life span of less than 20 years.2 Less happily, that term can also be applied, as Noah Feldman has argued, to the Iraqi Constitution.3 As I argue in my recent book Framed: America’s 51 Constitutions and the Crisis of Governance,4 constitution-drafting is never truly a ‘Rawlsian process’, where authors self-consciously sit behind a ‘veil of ignorance’ and purposely become ignorant of the identity of actual winners and losers in any given decision about institutional structures, assignments of rights, or the like, in order to draft a maximally ‘fair’ constitution. Nor is a constitution the result of a seminar-like exercise in ‘deliberative democracy’ where individuals respectfully engage one another as they express different points of view and find themselves changing their minds in response to what they now see are ‘better’ arguments and ultimately converge in consensual agreement. Instead, politically engaged persons, very much embedded within one or another group within the wider society, contend in a process of self-interested bargaining over the design of the system that they will have to live with if the drafting process is successful. Indeed, if one looks at the constitutional history of the country that I know best, the United States, one sees little evidence of genuine ‘consensus’ in 1787 (and such ‘consensus’ as existed was no doubt helped along by the fact that the losers in the great struggle over American secession from the British Empire had fled to Canada, England, the Caribbean, or even India and, therefore, did not have to be placated). There was, of course, significant agreement, at least on the part of the elites who convened in Philadelphia, that America’s ‘first Constitution’, the Articles of Confederation, had created, in the memorable term of Alexander Hamilton, the ‘imbecility’ of American government at the time, which lacked basic authority to tax, among many other notable deficiencies.5 But there was scarcely broad agreement on the specifics of the system that 2 See T Ginsburg, Z Elkins, and J Melton, The Endurance of National Constitutions (Cambridge, Cambridge University Press, 2009). 3 N Feldman, ‘Imposed Constitutionalism’ (2005) 37 Connecticut Law Review 857. My own comments on Feldman’s essay, with which I am in substantial agreement, can be found in S Levinson, ‘“Imposed Constitutionalism”: Some Reflections’ (2005) 37 Connecticut Law Review 921. 4 S Levinson, Framed: America’s 51 Constitutions and the Crisis of Governance (Oxford, Oxford University Press, 2012). 5 A Hamilton, ‘Federalist No 78’ in The Federalist Papers 86 (Cosimo Inc, 2006).
Consensus, Dissensus, and Constitutionalism 61
should replace it. This was seen in the closeness of the vote in several states – the most dramatic being the 30–27 majority by which the Constitution was ratified in the New York State Convention – and the fact that two states, North Carolina and Rhode Island, had not yet ratified the Constitution when George Washington took office on 30 April 1789. We will never know, of course, whether the Constitution would have survived had it been subjected, as in Egypt and most modern countries, to popular ratification instead of the decidedly ‘filtered’ representative process of special conventions, as required by Article VII of the Constitution; given the widespread perception in both Virginia and New York that the delegates, at least at the time of election, were likely to be opposed to ratification, one might surmise that a popular vote would have doomed the enterprise. This, of course, was the fate of the initial draft of the treaty that many viewed as a de facto constitution for the European Union, when popular referenda in France and the Netherlands rejected the proposal. For whatever reason, its drafters required unanimous consent of all of the Members of the European Union instead of mimicking what in some ways was perhaps the most important single decision in Philadelphia, the stipulation in Article VII that the new Constitution would become effective with the ratification by only 9 of the 13 states. This provision unceremoniously rejected Article XIII of the Articles of Confederation, which had required unanimous assent of all state legislatures for any amendment and, therefore, one might argue, required a genuine ‘consensus’ before the Articles could be changed. This ‘anti-consensus’ Article VII not only eliminated the veto that Rhode Island would almost certainly have exercised had unanimity been required, but also rendered irrelevant the fact that North Carolina and Rhode Island were missing from the United States when Washington took the helm. II. COMPROMISES AS ANTECEDENTS TO CONSTITUTIONALISM
One must also attend to the fact that such agreement as existed regarding the text of the Constitution that emerged from Philadelphia was purchased by several compromises, two of which were absolutely crucial. The first was acquiescence by the larger states to equal voting power of states in the United States Senate. Delaware, with roughly oneseventeenth the population of Virginia, had, like Virginia, two senators. This was bitterly opposed by Virginia’s James Madison, among others. Though he is often called ‘the father of the Constitution’, he was decidedly unhappy with several features of his ostensible child. Madison would go on to be one of the two principal authors of The Federalist,6 the most important systematic defence of the Constitution, written precisely to encourage uncertain delegates to support it. Yet, when writing of equal voting power, in ‘Federalist 62’,7 Madison simply described it as a ‘lesser evil’, the greater evil being no Constitution at all because of the opposition of small states. It was simply a price that had to be paid. Not for one moment had Madison changed his mind about the basic injustice of equal voting power in the Senate. Partisans of the Constitution – and of the presumptive virtues of ‘compromise’ in politics – sometimes refer to the decision about equal representation as the Great Compromise. No one uses capital letters, however, to refer to the second all-important compromise that The Federalist Papers (Cosimo Inc, 2006). J Madison, ‘Federalist No 62’ in The Federalist Papers 400 (Cosimo Inc, 2006).
6 7
62 Sanford Levinson purchased the Constitution, which involved various accommodations with chattel slavery; most important by far was the so-called three-fifths compromise. Although no one could seriously claim that slaves participated in the system or that their masters were genuinely committed to protecting the slaves’ interests – an argument that was often made with regard to other excluded participants like women or children, thought to be ‘virtually represented’ by their fathers or husbands – the states nonetheless got to include slaves as part of the basis for determining their representation in the House of Representatives, even if each slave only counted as three-fifths of a whole person. (It should be clear, incidentally, that the slave-owners would have been delighted to count each slave as the equivalent of two whole persons, so long as they had no vote. It was the anti-slavery north that insisted on limiting the ‘slavery bonus’ to three-fifths of a state’s slave population.) In any event, this bonus enhanced the power of slave-holding states not only in the United States House of Representatives, but also in the Electoral College mechanism that ultimately dictates who shall become President of the United States. Each state, after all, receives a number of electoral votes equal to the sum of its representatives plus the two senators. Were it not for the bonus, for example, John Adams almost certainly would have received a second term in 1800 instead of being defeated by the Virginian (and slave-owning) Thomas Jefferson. Moreover, among the powers presidents receive is to nominate persons to serve on the federal judiciary. Because of these basic structures established by the Philadelphia Convention, the Supreme Court was resolutely pro-slavery until the Civil War. Israeli philosopher Avishai Margalit has written an important book tellingly titled On Compromise and Rotten Compromises,8 which describes the accommodations with slavery as an example of the latter; it is defensible, if at all, only if the alternative were truly some version of the heavens falling. Although Margalit, a notable member of the ‘peace camp’ within Israel, is a proponent of compromise, he does draw the line at ‘rotten compromises’ that inflict inhumane treatment upon the losers. Whether the failure of the United States to achieve unity behind a constitution, instead of fragmenting, for example, into two or three separate countries, would constitute the heavens falling is certainly open to doubt. Moreover, for what it is worth, the compromises ultimately did not work: The United States was rent by a savage war between 1861–65 that killed approximately 750,000 people, well more than two per cent of the total population at the time. As American constitutional historian Mark Graber has demonstrated, the war in many ways was caused not only by the fundamental cultural divide between ‘slave’ and ‘free’ states, but also by the constitutional structures that were established in 1787 ostensibly to control the divide.9 III. CONSTITUTIONS AND ‘PEOPLE OF FUNDAMENTALLY DIFFERING VIEWS’
Abraham Lincoln, quoting the Gospel of St Matthew, famously proclaimed that a ‘house divided against itself cannot stand’. And, ironically or not, it was his election only several years later that triggered secession and war. But what lesson do we learn from this capsule history? Is it that successful constitutions do in fact require consensus or only that there may be limits to the amount, and kinds, of disagreement that can be handled A Margalit, On Compromise and Rotten Compromises (Princeton, Princeton University Press, 2009). See, M Graber, Dred Scott and the Problem of Constitutional Evil (Cambridge, Cambridge University Press, 2006). 8 9
Consensus, Dissensus, and Constitutionalism 63
by constitutional mechanisms? Consider the famous statement of Justice Oliver Wendell Holmes in his canonical dissent in Lochner v New York: ‘a constitution is not intended to embody a particular economic theory, whether of paternalism and the organic relation of the citizen to the State or of laissez faire. It is made for people of fundamentally differing views’ (emphasis added).10 One might well generalise beyond the issue of economics. Holmes could presumably have written the same comment had the topic been, for example, religion or practically any other important aspect of statecraft. Thus anyone sympathetic to Holmes’s understanding of constitutional reality must be sceptical about proclamations, like those of Mr Sabahi, that a necessary condition of an acceptable constitution is that it instantiates a genuine consensus existing within the polity that it purports to structure. To be sure, one can be grateful if that is the case, but one might also well believe that it will in fact rarely exist. Because many, perhaps even most, constitutions are drafted within decidedly divided political orders, the aim of constitutional drafters must therefore be to achieve a modus vivendi whereby conflicting groups, who may with good reason barely trust one another with regard to some fundamental issues, can live in some degree of genuine civil peace. To be sure, the drafters may also aspire to creating a structure that will, over time, generate genuine loyalty to the new political order and something that one can describe as a consensus about its central values and purposes. But their first duty is to achieve some degree of civil peace. It is readily understandable, therefore, that one finds an outstanding collection of essays entitled Constitutional Design for Divided Societies11 or, to bring the discussion even closer to home – and to the three essays under review – Israeli political scientist Hanna Lerner’s extremely illuminating book Making Constitutions in Deeply Divided Societies,12 which focused on India, Ireland, and Israel. Needless to say, there is a vital difference between the first two and Israel inasmuch as India and Ireland both do have canonical written constitutions, whereas Israel is one of almost literally a handful of countries in the modern world lacking such a document, and it is the only country established since World War II in that category. Thus, whatever explains Israel’s outlier status, it cannot be merely the fact that there is insufficient ‘consensus’ within the country. One might well expect somewhat different kinds of constitutions in more rather than less homogeneous politics, but this is quite different from believing that pluralism condemns the constitutionalist project at the outset or even that it dictates particular behaviour by judges who may have a duty to enforce a constitution. IV. CAN JUDGES AND COURTS NECESSARILY SERVE AS GUARDIANS OF A CONSTITUTIONAL ORDER?
Perhaps the key question both for the constitution-drafter and outside analysts is deciding what one regards as the central function(s) of a constitution and the circumstances under which one can achieve at least some element of the goods attached to possessing a written constitution. So, turning to the essays under review, I want to suggest that it may be a fundamental error to assume that the most important function of a constitution is Lochner v New York 198 US 45, 75–76 (1905). S Choudhry (ed), Constitutional Design for Divided Societies (Oxford, Oxford University Press, 2008). 12 H Lerner, Making Constitutions in Deeply Divided Societies (Cambridge, Cambridge University Press, 2011). 10 11
64 Sanford Levinson to announce (and to protect) some set of substantive values that ostensibly define the polity. Especially in ‘divided societies’ the task is to construct a set of political inst itutions – sometimes called ‘frameworks of governance’ – that are empowered to make necessary decisions capable of meeting the pressing needs of the moment and, one hopes, thereafter. Quite often, as was the case in the United States in 1787 and in Israel throughout its six-decade history, these exigencies will involve various forms of ‘national secur ity’. One cannot, for example, read The Federalist without realizing that the proponents of the new Constitution feared a variety of threats to the fragile new nation, some emanating from European powers, some from American Indian tribes, many of whom had, with good reason, supported the British in their attempt to prevent American secession from His Majesty’s empire. It is only a slight exaggeration to suggest that behind many of the arguments for adoption of the new Constitution was the Hobbesian command that government exists primarily to provide security, with all else being secondary. But, of course, especially since World War II and the justified fear of claims to unlimited power by the state, notions of constitutionalism have emphasised ever more the protection of various rights; these are often proclaimed, in universalistic language, to be ‘human rights’ and not, for example, simply the set of rights protections adopted, perhaps for idiosyncratic reasons, by a particular polity. To focus on ‘national security’ as the basic aim of government is, almost necessarily, to privilege governmental power against the rights of its citizens (and, even more, those who are outside, for whatever reason, the official political community). For good reason, this monotonic conception of a constitution is rejected by almost all modern commentators; instead, as already suggested, emphasis is placed instead on the role of constitutions in protecting ‘basic values’ against governmental overreach.13 These values may be immanent within a given society or ‘universal’, thought to define the predicates of just rule in all times and places. But, from this perspective, constitutions are devices for instantiating values, whether immanent within a given society or ‘universal’, establishing the predicates of just rule in all times and places, and all institutions and the particular decisions they make are to be judged by reference to the degree that they effectively protect the values to which the polity is deemed committed. Almost inevitably, as revealed in all three of the papers by Gideon Sapir,14 Ariel Bendor,15 and Alon Harel,16 discussions of value protection lead quickly to discussions of the specific role of courts in protecting those values against incursions by the legislature/ parliament when passing statutes or by the executive branch when implementing them. After all, one of the most stunning results of World War II, in effect, was the rejection by European countries of the norm, going back to the French Revolution, against ‘judicial review’ of the legitimacy of legislation. Instead, partly under the influence of Hans Kelsen, partly, no doubt, because of the model provided by the American system of constitutionalism, the countries of Europe – and almost all of the rest of the world as well – have accepted the legitimacy of both national and transnational courts that have often proved more than willing to exercise such powers. One can, for example, scarcely speak any longer of ‘parliamentary sovereignty’ as the central feature of British government 13 See, eg the first of three rationales for constitutions discussed by Gideon Sapir in his interesting chapter in this volume. 14 ibid. 15 A Bendor’s chapter in this volume. 16 A Harel’s chapter in this volume.
Consensus, Dissensus, and Constitutionalism 65
given the embeddedness of Great Britain within the various European treaties and, perhaps most importantly, the European Convention on Human Rights,17 not to mention more recent domestic legislation empowering British courts to enforce European law directly. There is, of course, genuine debate about the specific roles of the judiciary, but some degree of ‘rule by judges’ seems to be widely accepted. Critics of judicial review, such as Jeremy Waldron, are distinct outliers. It may be relevant that Waldron is originally from New Zealand, perhaps the one remaining country in the world that lacks not only a canonical constitution but has also, in its ‘constitution-like’ New Zealand Bill of Rights Act of 1990, explicitly rejected the notion of judicial enforcement against Acts of Parliament that arguably contravene the Act. Still, it is obvious that the role of the judiciary continues to be a topic of sharp, often acrimonious, controversy within Israel. Perhaps this can simply be explained by the notorious proclivities for intellectual conflict often attributed to Jews in general. It is, however, difficult to ignore the relevance of the remarkable career and influence of former Israeli Supreme Court President Aharon Barak regarding the Court over which he presided. Though he was building on practices and precedents established by his predecessors,18 it was Barak who announced, wisely or not, that the passage of certain Basic Laws in the early 1990s, and their subsequent implementation by the Court, affected nothing less than a ‘constitutional revolution’, the safeguarding of which was the special responsibility of the Court. Part of the way it exercised this responsibility was to establish among the world’s most liberal ‘standing’ laws. Whether or not, as is said to be the case in India, a suit can be initiated simply by a disgruntled citizen’s sending in a postcard claiming that government is violating legal norms, it is clear that the Israeli High Court is stunningly more accepting of jurisdiction and willing to assert justiciability than are, say, national courts within the United States, which often seem far more eager to find ways to prevent access to courts than to enforce legal duties against the state. Barak may well have been the most widely known – and, by many, admired – judge in the entire world because of the mixture of his judicial opinions, his extra-judicial writings defending his strong view of the judiciary and of purposive interpretation, and his attendance at gatherings of judges throughout the world, such as the annual conference sponsored by the Yale Law School, where he has taught for many years. But as we know from another Gospel passage, this one from Mark, ‘prophets’ are often ‘without honour’ in their own countries, and an enduring memory of the initial conference at which these papers were presented was watching Barak, who loyally attended all sessions, listen to what was often scathing criticism of his work and legacy. An important question is whether Barak merits criticism in terms of his legal philosophy, a necessarily abstract question, or, instead, whether one can simply believe that he was sometimes unwise, pragmatically speaking, in choosing to exercise judicial power on certain occasions. My own mentor, Robert McCloskey, who published in 1960 a famous short book, The American Supreme Court,19 emphasised not only that one cannot understand American political history without paying some attention to the roles played by judicial decisions 17 Convention for the Protection of Human Rights and Fundamental Freedoms (European Convention on Human Rights, as amended) (ECHR). 18 See, eg P Lahav, Judgment in Jerusalem: Chief Justice Simon Agranat and the Zionist Century (California, University of California Press, 1997). 19 R McCloskey, The American Supreme Court 5th edn (Chicago, University of Chicago Press, 1960, 2010).
66 Sanford Levinson and, therefore, the judges who made them, but also that the Court was effective only if it remained reasonably close to the centre of American politics. It could gently prod the country toward its preferred vision or just as gently push back against what it deemed overreaching, but if it deviated too far from mainstream opinion or, more to the point, the views of political elites exercising power, then the Court itself would be threatened in one way or another. Almost necessarily, an emphasis on the judiciary – especially, as in most places throughout the world, where the judiciary is substantially ‘independent’ from ordinary political mechanisms of accountability20 – highlights the question of whether sufficient support exists within a specific society for given ‘basic values’ that are thought to underlie a constitution and that in turn provide the basis for a judicial decision. It should be clear, incidentally, that ‘sufficient support’ is quite different from ‘consensus’, inasmuch as it certainly allows the possibility that significant parts of the population, perhaps on occasion even a majority, will be sceptical of these values or, as likely, of the specific interpretations offered by the judiciary. In any event, all three of the essays seem to agree that Israel is so ‘pluralistic’ with regard to ‘basic values’ that the role played by the contemporary Israeli Supreme Court is problematic. Nor is it enough to embrace a notion of a constitution as encouraging ‘dialogue’ about ‘basic values’, because one must still suggest ways by which a given dialogue will come to an end, given the necessity to render a decision in real time. Who, practically speaking, and not only in terms of legal theory, will have the last word in any given dialogue? Sapir notes the dialogical potential of the ‘Notwithstanding Clause’ within the Canadian Charter of Rights and Freedoms, but he notes as well its practical failure. The reason is that popular culture views a decision by Parliament taking issue with a judicial decision not as the assertion of a legitimate, albeit conflicting, point of view, but rather as the ‘overriding’ of the constitution by politicians whose views are not to be accorded the same respect as judges. Sapir is scarcely a champion of the judicial vision articulated by Justice Barak, which he ultimately views as paying insufficient attention to the pluralism of Israeli society, which means that ‘Even if we agree that there are values of supreme importance, in a pluralistic society a consensus does not exist on the identity of those values’. Bendor, who champions ‘professional’ over ‘political’ judgment, nonetheless is extremely wary of judges who, according to him, rule on the basis of ideology rather than the strict confines of the law. Thus he also endorses the ability of the Knesset (Israeli Parliament) to ‘override’ judicial decisions, though, interestingly enough, only for the life of the Knesset itself. Harel, interestingly, even while defending judicial review as intrinsically necessary to protect the basic right of an individual to be heard when registering a complaint about treatment by the government, resolutely refuses to privilege the substantive determinations of the court. That is, he does not believe that the judges have any particular expertise in discerning the basic values of the polity, assuming they exist at all, or that their 20 Interestingly enough, the principal exception is the United States, though at the level of state, rather than national, government. Most judges in the United States are directly elected or subject to so-called ‘retention elections’ several years after their initial appointments. As a result, some analysts of state judiciaries reject the practical importance of the so-called ‘counter-majoritarian difficulty’ in favour of a quite different ‘majoritarian difficulty’ generated by over-acquiescence to majoritarian decisions by state legislatures or executives. For further discussion of this point, see Levinson (n 4) chs 12 and 13.
Consensus, Dissensus, and Constitutionalism 67
interpretations of such values as may exist will necessarily be better than those of the Knesset or administrators. Consequently, he must explain why a judicial hearing is necessary instead of legislative ones when deciding whether or not to pass a statute in the first place or, presumably, administrative ones when deciding on specific implementation schemes. I confess I am not convinced by his argument, though largely on the pragmatic grounds that it is just too expensive and time-consuming to expect that the judiciary will in fact be open to hear each and every complaint that disgruntled individuals might have about how the government is treating them. After all, almost all legislation is necessarily both ‘under-’ and ‘over-inclusive’, which means by definition that it is not ‘perfectly tailored’ to address only those particular instances meriting regulation. To adopt the terms of American constitutional law, if the regulation in question involves a ‘fundamental’ right or interest, one might well expect more than a ‘minimal’ fit between means and ends and the inevitable spillovers that this standard allows; otherwise, one must realise that legislation is often (usually?) going to be somewhat sloppy and imprecise. One can argue whether or not dogs should be allowed in restaurants, but it seems altogether implausible, in a society that does prohibit dogs (save for ‘guide dogs’), to expect a judge to take scarce time and energy to resolve a complaint that a particular dog is sufficiently gentle and quiet to pose no problem for other diners. Even with regard to ‘fundamental values’, though, we return to the problem of selecting them out from among the far wider class of ‘values’ instantiated in any social order and determining whether judges possess any special expertise of judges either in defining or implementing them. V. THE IMPORTANCE OF INSTITUTIONS AS WELL AS RIGHTS (OR JUDGES ENFORCING RIGHTS)
My own view is that such discussions, however valuable, are extremely unlikely ever to be resolved. Even if one agrees that it is a mistake to place excessive confidence in the ability of judges to provide wise counsel, it is equally problematic to accept an understanding of ‘democracy’ that allows legislatures or administrative agencies to do just whatever they want, especially with regard to the rights and interests of beleaguered minority groups that can scarcely look to ordinary politics as the venue to vindicate their claims to be treated with dignity and justice. I also increasingly believe that the almost exclusive focus on the role of the judiciary in interpreting inevitably ambiguous aspects of constitutions or ‘Basic Laws’ has led us to ignore the importance of a constitution’s role in structuring basic political institutions. In my new book Framed,21 I sharply distinguish between what I term, referring to the United States Constitution, ‘the Constitution of Settlement’ and the ‘Constitution of Conversation’. The latter is familiar to all lawyers; it consists of the ambiguous or ‘indeterminate’ parts of the text – think, for example, of the injunction that no state deny any person of ‘equal protection of the laws’ – that, almost by definition, provides no clear answers and, instead, invites seemingly endless conversation about the meaning or interpretation of the terms in question. These conversations are both ‘first order’ – what do the terms mean? – and ‘second order’ – which institution, if any, should get the ‘last word’ in assigning meaning? Above n 4.
21
68 Sanford Levinson But all constitutions very much include ‘Constitutions of Settlement’, which can be defined in large part by clauses using language that, practically speaking, does not permit genuine debates over meaning even if one can certainly debate the wisdom of a particular constitutional text. These clauses answer absolutely essential questions such as who is legally authorised to exercise powers, how such power-holders are selected, how many are there, and for how long they serve their terms. To take an American example, there is a single president, who serves a fixed four-year term, with no eligibility for reelection beyond a second term. There is, for example, no possibility of a ‘vote of noconfidence’ to remove a (non-criminal) president from office, nor can an unusually gifted chief executive be given a third term even if, for example, the United States is involved in international conflicts whose resolution might be easier under the leadership of the incumbent. There are, of course, many other examples that could be given of the Constitution of Settlement. My major point is that legal academics, whether in the United States or elsewhere, rarely discuss these aspects of constitutions, largely, I suspect, because they do not involve the skill-set that has become constitutive for most law professors. That is, our forte is increasingly the development of ‘theories of legal interpretation’ and, concomitantly, discussion of the particular role of judges and courts in giving meaning to the ‘Constitution of Conversation’. We are often very good at doing this, but the price we have paid, I am afraid, is failing to pay adequate attention to the issues instantiated in Constitutions of Settlement. Whatever the importance is of determining how to interpret controversial aspects of the Basic Laws, one might well think, with regard to the future of Israel, that the organization of the Knesset is at least as important. To offer only one, but vital, example, the Israeli system of proportional representation, with its extraordinarily low threshold for representation in the Knesset, has notoriously contributed to the fragmentation of Israeli politics and the ability of very small parties, in effect, to extort special treatment in return for providing key votes to maintain a given coalition in power. As a matter of raw fact, it is this feature of Israeli constitutionalism, rather than any ‘constitutional’ or ‘quasi-constitutional’ language in statutes, that assures that ultra-Orthodox Jewish political parties exercise remarkable influence over Israeli politics. One of the central lessons taught by James Madison is that constitutional provisions ostensibly promising the protection of certain rights are often mere ‘parchment barriers’, in contrast to hard-and-fast institutional features (such as the three-fifths compromise or equal voting power in the Senate) that will determine, as a practical matter, who can exercise decisive control in reacting to divisive political questions. I am, of course, well aware that many distinguished Israelis, including law professors, have concerned themselves with redesigning the ‘Basic Laws’ of Israeli political institutions, but I fear that, overall, these proposals get far less attention than do the latest attacks on ‘judicial overreaching’ or the merits of adopting an ‘override proposal’ within the Israeli legal system. Perhaps institutional design is thought to be less interesting than conversations about ‘basic rights’, even though any given institutional design will necessarily instantiate various normative views about governance. Or, as is certainly the case in the United States, perhaps it is simply that institutional reform appears to be a hopeless project, so that offering or discussing specific proposals is thought to be a simply quixotic enterprise. In the United States, this can be readily justified by reference to what is probably the worst single feature of the United States Constitution, the ‘Amending Clause’ of Article V, which in effect makes the United States Constitution the most dif-
Consensus, Dissensus, and Constitutionalism 69
ficult to amend of all constitutions in the world today. But, of course, Israelis can offer no such excuse, inasmuch as I assume the Knesset retains great power to redesign basic institutions if only it has the political will to do so. Indeed, even with regard to the issue of the judiciary and its role, it is obvious that one important power that the Knesset has exercised, entirely independent of any ‘override’ powers, is simply to increase the membership of the Court and therefore allow a certain degree of ‘court-packing’ through the appointment of judges deemed more compatible to the views of those doing the appointment. And this also necessarily involves diminishing the remarkable past influence enjoyed by the sitting members of the Court, particularly Justice Barak, in selecting newcomers to the Court. To be sure, Israel was in no way unique in privileging a genuine degree of institutional autonomy by the Court not only when issuing its decisions, but also in choosing its members. The Indian Supreme Court is said to exercise similar autonomy with regard to selection of judges, and this may be true in Denmark as well. Moreover, it is worth noting that Justice Barak was in effect forced off the bench by the simple fact that Israel, like almost all countries around the world – the United States is the most glaring exception – has rejected what might be called ‘full-life tenure’ of appointment. Most countries place stringent term limits of somewhere between 10 and 15 years; others, like Israel, require retirement at 70. I personally think that either is preferable to the American national practice, which allowed, for example, Justice John Paul Stevens to serve for 34 years, retiring only after he turned 90. An interesting question facing any constitutional designer is whether answers to these questions should themselves be ‘constitutionalised’ or whether it should be left up to the ordinary political institutions to change their minds. The United States Constitution, for example, is widely interpreted as requiring life tenure, but it in no way indicates exactly how many justices there shall be in the Supreme Court. Over American history, the number has ranged from 6 to 10, though for the past 150 years or so it has been 9, and one suspects that most Americans believe this is part of our ‘unwritten constitution’. In any event, for all of the genuine stimulation and illumination provided by the conference on Israeli constitutionalism, including these three papers, I was disappointed by the relative lack of explicitly institutional discussion, including, of course, the ways that given institutional designs inevitably reflect various normative views. One of the unfortunate similarities between contemporary American and Israeli politics is the widespread view that the political systems of both countries are dysfunctional. Thomas L Friedman described the American system (and would, no doubt, describe the Israeli system as well) as ‘pathological’.22 Whatever one thinks of the United States or Israeli Supreme Courts, or of their current or past judges, I strongly suspect that they have relatively little to do with the pathologies inflicting both countries. In both countries, the dilemma is whether the operating political systems can respond adequately to the challenges facing them. Perhaps this is not a conventional ‘legal’ question. But it should be of immense interest to anyone interested in the capacity of a constitutional order to function effectively and to generate the genuine loyalty (and affection) of its citizenry, especially if one cannot rely on an underlying ‘consensus’ that might otherwise bind a country together.
See, eg the discussion in Levinson (n 4) 8–9.
22
6 Majestic Constitutionalism? The Notwithstanding Mechanism in Israel TSVI KAHANA
I. INTRODUCTION: THE CHICKEN AND THE KITCHEN
I
N 1916, SHOLEM Aleichem, a famous Yiddish writer, published a book called Motl, the Cantor’s Son. In that book, he adopted the perspective of a young boy in order to describe the individual and collective story of the massive Jewish migration from Eastern Europe to America in the nineteenth and early twentieth centuries. A major theme in the family’s absorption into America is being exposed to English and learning to speak it. The character who has the hardest time adjusting to the new world and to New York City is Motl’s mother, who misses her dead husband, her friends in the village, and the vanishing Eastern European Jewish existence. As time goes on, however, she begins to adjust. Motl tells us: ‘Half of what she says is in American by now. She just gets everything backwards. Instead of cooking a tshikn in the kitshn, she cooks the kitshn in the tshikn’.1 In this chapter, I argue that Motl’s mother is not alone; the State of Israel is travelling with her. As it has been learning to speak the language of America – the language of constitutional rights – Israel has mixed the chicken and the kitchen. This began in 1994, when it experienced a constitutional crisis concerning one dish – a frozen meat dish – and instead of dealing with that dish alone, it imported into its constitutional structure a whole kitchen – a mechanism to allow for deviations from constitutional rights. Specifically, the Government wanted to ensure that the legislation banning the importation of non-kosher frozen meat would remain in force notwithstanding its potential violation of the right to freedom of occupation, guaranteed in Basic Law: Freedom of Occupation (BLO).2 However, instead of amending the BLO to create an exception for frozen meat importation, Israel created a mechanism which allows the Knesset (Israeli Parliament) to declare that any act of their choosing shall apply notwithstanding the BLO. I refer to this mechanism, created by section 8 of the BLO, the ‘Notwithstanding Mechanism’ (NM), and I refer to its use as a ‘constitutional deviation’.3 1 S Aleichem, The Letters of Menakhem-Mendl and Sheyne-Sheyndl and Motl, The Contor’s Son (H Halkin tr, New Haven, Yale University Press, 2002) 310. 2 Basic Law: Freedom of Occupation, 5752-1992, SH No 1387, p 114, repealed and replaced by Basic Law: Freedom of Occupation, 5754-1994, SH No 1454, p 90. 3 The headline of s 8 reads ‘The Validity of a Deviating Statute’ (my translation, TK).
74 Tsvi Kahana Since 1994, there has been some ‘kitchen’ talk in Israel; that is, talk about the idea of including an NM in a prospective Israeli Constitution. My main goal in this chapter is to advocate for focusing on the chickens as they come. In other words, I argue that as long as Israel’s Constitution – the Basic Laws – can be amended without a super-majority, deviations from them should be effected via constitutional amendment, not through an NM. In addition to this substantive argument, I advance a methodological point concerning comparitivism. I suggest that in the matter of an Israeli NM, the exercises of constitutional comparison and borrowing has had constraining, rather than enriching, effects. The adoption of the NM in Israel was inspired by the existence of an NM in Canada.4 Since 1994, the debate about the NM in Israel has been focused on whether Israel should follow the Canadian route in a future Israeli constitutional Bill of Rights. Little consideration has been given to the route of a constitutional amendment, an option not available in Canada. In Canada, the NM is based on what I called ‘the empowered legislature’ model. The Charter empowers legislatures to have the final word on constitutional issues, in order to avoid the familiar problems associated with judicial hegemony brought about by traditional Constitutions. But Israel does not have a traditional Constitution to start with. It has non-entrenched Basic Laws, most of which can be amended by an ordinary majority vote of the Knesset. In other words, the Knesset is already empowered to respond to or to pre-empt problematic judicial decisions.5 I demonstrate that the NM was introduced in Israel not in order to empower the Knesset, but in order to avoid addressing in the Basic Laws the matter of kashrut (the requirement that food be kosher), which seemed an ‘inappropriate’ subject matter for the Constitution to address. I therefore believe that the Israeli NM is based on ‘the majestic Constitution’ model. Under this model, even when an institution is empowered to amend the Constitution, it should not do so with regard to matters which are too specific or which do not befit the grandeur of the Constitution. The NM allows the Knesset to amend the BLO without reference to specific, non-majestic, matters.6 In part II of this chapter, I introduce the familiar story of the Canadian NM and the less familiar story of the Israeli NM. In part III, I argue that even if Constitutions should speak in general terms as a general matter, when a specific issue is of significance to the polity, the polity should not shy away from including this issue in the Constitution. In part IV, I ask whether, in addition to the notion of the majestic Constitution, there may be other reasons to deviate from constitutional rights through an NM rather than a constitutional amendment. I conclude that, quite to the contrary, there are additional reasons to prefer constitutional amendment as the method of deviation.
Canadian Charter of Rights and Freedoms, s 33. The fact that the Knesset may do this under its constituent power, and not under its legislative power, is important in theory but irrelevant to the question of the division of power between the Knesset and the Supreme Court. 6 The expression ‘majestic Constitution’ is inspired by Robert Jackson J of the US Supreme Court who talks of ‘the majestic generalities of Fourteenth Amendment’ in Fay v New York 332 US 261, 282 (1947). In this chapter, I assume that a provision is majestic when it is general and is not majestic when it is not general. There may be theories about other ‘majestic’ requirements of the Constitution beyond generality, but they are beyond my interest here. 4 5
The Notwithstanding Mechanism in Israel 75
II. THE NOTWITHSTANDING MECHANISM: CANADA AND ISRAEL
A. The Empowered Legislature Model Section 33 of the Charter gives Canadian legislatures the power to make laws notwithstanding most Charter rights. This provision was added to the Charter in order to break a deadlock in difficult federal–provincial negotiations surrounding the Patriation of the Constitution in Canada in 1982. While the Federal Government and two provinces were interested in a Bill of Rights that would provide for judicial supremacy, eight provinces objected to such a Bill altogether, fearing it would curtail provincial power too greatly.7 The compromise was a Bill of Rights with an NM.8 As the years went by a growing number of scholars, in Canada and worldwide, began to view the Charter not as a second-best solution, but as quite optimal. Their argument is that a constitutional Bill of Rights with an NM was preferable to either no Bill of Rights at all or an American-style, judicial supremacy-based Bill of Rights. An NM-bearing Bill of Rights would allow for a constitutional partnership between courts and legislatures, which would benefit the polity in two ways. First, it would allow for more constitutional dialogue – popular and institutional – than either having a Bill of Rights or having a traditional Bill of Rights, due to the fact that it would allow both courts and legislatures to have their separate views on constitutional matters aired in the public sphere.9 Second, it would allow the polity to benefit from the unique institutional capacities of both the courts and the legislatures, courts being the experts on rights issues and individual claims, and legislatures being the experts on policy issues.10 In Israel, the story is a bit more complicated. To begin with, we must distinguish between discussions about the desirability of including a hypothetical NM in a Constitution, if and when Israel chooses to adopt one, and discussions about the NM that was adopted in Israel in 1994. The former are very similar to the Canadian discussions; unfortunately, there simply appear to have been very little discussions of the latter variety. Proposals for the inclusion of an NM in a future Israeli Constitution (or Constitution equivalent, namely, entrenched Basic Laws) were made before and after the ‘Constitutional Revolution’ of 199211 and increasingly in the last decade. The discussion surrounding these proposals followed the contours of the Canadian discussion, with reference being made to both the compromise argument and the partnership argument. In the Israeli context, the compromise argument was that the adoption of an NM in Israel could serve to break the deadlock in constitution-making that has been present since the birth of the State. The deadlock is partially caused by disagreement between 7 See R Romanow, J Whyte and H Leeson, Canada . . . Notwithstanding: The Making of the Constitution, 1976–82 (Toronto, Carswell/Methuen, 1984). 8 See H Leeson, ‘Section 33, The Notwithstanding Clause: A Paper Tiger’ (2000) 6 IRPP Choices: Courts and Legislatures 3, 6–14. 9 See, eg S Gardbaum, ‘The New Commonwealth Constitutionalism’ (2001) 49 American Journal of Comparative Law 707, 747; M Tushnet, Weak Courts, Strong Rights: Judicial Review and Social Welfare Rights in Comparative Constitutional Law (Princeton, Princeton University Press, 2008). 10 See, eg CP Manfredi, Judicial Power and the Charter: Canada and the Paradox of Liberal Constitutionalism, 2nd edn (Oxford, Oxford University Press, 2001) 194–95. 11 See n 20 below.
76 Tsvi Kahana supporters of American-style constitutionalism with judicial supremacy and those who object to any type of constitutional judicial review. Assuming that both camps wish to put the discussion behind them and agree on a Constitution, an NM-bearing Bill of Rights is a compromise that each party would be able to live with.12 The partnership argument also closely followed the Canadian arguments, with an Israeli addition: in Israel, society is divided and there is no consensus either on the importance of constitutional rights or on their content. Therefore, it is even more appropriate in Israel to allow for institutional partnership and popular dialogue concerning constitutional values.13 Opposition for the NM was also cast in comparative terms. An obvious risk of leaving the last word on constitutional rights to the legislature is the spectre of tyranny of the majority (or, to use the popular euphemism, legislative blind spots). In Canada, so far, the NM has not been used to inflict abhorrent rights violations.14 However, some Israelis worry that ‘It is not Canada here’15 and that given the Knesset’s record on treatment of minorities, it should not be trusted with the last word on constitutional rights. The discussion about an NM for Israel assumes that Israel will eventually have a constitutional Bill of Rights, which would be accomplished either by adopting such a Bill or by entrenching the existing Basic Laws. While a discussion of a future Israeli Bill of Rights is important and interesting, a proper analysis of constitutional design must pay heed not only to ideal design, but also to current, idiosyncratic, non-ideal constitutional apparatuses, such as Israel’s. In this chapter, I focus on the particular NM that was adopted in Israel. As I shall explain, this mechanism is based not on the empowered legislative model reflected in the Canadian experience, but on a majestic Constitution model. B. Israel: The Majestic Constitution Model i. The ‘Constitutional Revolution’ of 1992 Like many developments in the world of Constitutions, the introduction of the NM to Israel was based not on thorough discussion and analysis, but on political expediencies and creative draftsmanship. Until 1992, Israel did not have a constitutional document for rights protection. While it did have several Basic Laws in place, these deal not with rights but with governmental structure16 and are not entrenched – they can be amended or repealed by legislative majorities.17 In fact, most of the Basic Laws do not even require an absolute majority in 12 See, eg F Raday, ‘The Notwithstanding Clause as a Substitute for a Complete Lack of Judicial Review’ (2005) HEMDAT – Council for Freedom of Science, Religion and Culture in Israel huka.gov.il/wiki/material/ data/H14-12-2005_10-35-30_piskat.doc (in Hebrew). 13 See, eg G Sapir, Constitutional Revolution in Israel – Past, Present and Future (University of Haifa Press & Yearbooks, 2010) 212–19. 14 See T Kahana, ‘What Makes For a Good Use of the Notwithstanding Mechanism’ (2004) 23 Supreme Court Law Review (2d) 191, 210. 15 See E Gross, ‘It’s Not Canada Here’, Haaretz, 28 February 2007 www.haaretz.com/print-edition/opinion/ it-s-not-canada-here-1.214222. 16 S 4 of Basic Law: The Knesset is the single exception to this rule, since it prescribes an equality provision in the specific field of elections. 17 The only two exceptions are the provisions providing that the legislature must be re-elected every four years and that the provisions of Basic Law: The Knesset cannot be changed by emergency regulations (ss 8, 44 and 45).
The Notwithstanding Mechanism in Israel 77
the 120-seat Knesset for amendment or repeal. Since the Knesset has no quorum requirement, a 2: 1 vote in the plenum is sufficient.18 In March 1992, the Knesset created the two new Basic Laws, the BLO and Basic Law: Human Dignity and Liberty (BLH), in what was later termed the ‘Constitutional Revolution’.19 The BLO protects the right to freedom of occupation; the BLH enumerates several rights including the right to ‘human dignity’, which has later been interpreted by the Israeli Supreme Court as encompassing several additional political and social rights.20 Both of the new Basic Laws adopt the Canadian structure of rights protection: they include limitation clauses that permit the rights to be infringed under certain conditions.21 Unlike the Canadian Charter however, neither Basic Law is entrenched. The BLO requires only an absolute majority of Knesset Members (ie 61 Members) for its amendment, while the BLH does not even require that, being subject to amendment or repeal by an ordinary vote in the legislature (ie a majority of the deputies present at the time of voting, which may be fewer than 61). Even with the introduction of the new Basic Laws, then, majority rule still reigns in Israel in the field of rights. Originally, neither Basic Law included an NM. However, as I explain in the next section, in 1994, a constitutional problem concerning the importation of non-kosher meat into Israel prompted the addition of an NM to the BLO. ii. The Addition of the NM to the BLO in 1994 Prior to 1994 the importation of frozen meat into Israel was undertaken almost solely by the State. The rationale for this restriction was that private entities should not be entrusted with the performance of essential services. Since Israel had to maintain a constant state of readiness for war, its designation of essential services was unusually broad and came to encompass many activities, such as the importation of frozen meat.22 While this near-monopoly by the State was not based on religious grounds, it yielded a situation where most of the imported meat was kosher.23 In 1992, the Government decided to privatise the meat market and declared its intention to change the governing regulations accordingly.24 Some religious communities in Israel, including the religious political parties, opposed this change, which they feared would result in non-kosher meat flooding into the country.25 The Government of Prime Minister Yitzhak Rabin, eager to retain 18 Until 2003, the Basic Laws were ruled by the Supreme Court to not have constitutional supremacy. That is, any future Act that was inconsistent with them would be deemed to have implicitly amended them, and hence would always be constitutional (provided that it met the absolute majority requirement if the relevant Basic Law required that). In 2003 the Supreme Court ruled that Basic Laws do enjoy constitutional supremacy. See HC 212/03 Herut v Chairman of the Central Elections Committee 57(1) PD 750 [2003] (in Hebrew). 19 See A Barak, ‘The Constitutional Revolution: Protected Human Rights’ (1992–93) 1 Law and Government 9 (in Hebrew). 20 Among these are, most notably, freedom of expression (CA 105/92 Reem Engineers Contractors v NazarethIllit 47(5) PD 189, 201 [1993] (in Hebrew)) and equality (HCJ 5394/92 Hupert v Yad Vashem 48(3) PD 353, 362 [1994] (in Hebrew). 21 S 8 of the BLH and s 4 of the BLO. 22 See HCJ 3872/93 Mitral v Prime Minister and Minister of Religious Affairs 47(5) PD 485, 492 [1993] (in Hebrew). 23 The Government did allow private entities to import non-kosher meat under a quota regime, but most of the imported meat was kosher. 24 ibid. 25 ibid 495.
78 Tsvi Kahana the support of religious political parties while pursuing the peace process (leading to and following the Oslo Accord), abandoned its plan to privatise the industry. However, following a petition by Mitral Inc, a meat importer, the Supreme Court ordered the Government to proceed with the privatization plan and to enact the relevant regulation within four months.26 In response, the Government decided to postpone the privatization indefinitely and explained that it planned to settle the issue at a later point in time through legislation (the original plan had been to implement the plan by regulation).27 This move provoked another petition by Mitral, which prompted the Supreme Court to nullify the Government’s decision to postpone privatization of the meat importation on the grounds that it was ‘unreasonable’ since it was based on the desire to impose kashrut on those who did not observe it.28 Concerning the Government’s plan to deal with the issue by way of legislation, the Court added in obiter that even primary legislation imposing kashrut on those who did not observe it would be unconstitutional if enacted. The Court reasoned that limits on importation are an infringement on the freedom of occupation, and limiting this freedom for the sake of religious coercion did not constitute a ‘proper purpose’, as required by the limitation clause of the BLO. This is the point where the Canadian and Israeli stories diverge. In Canada, no legislature alone can amend the Charter.29 If a court renders a statute unconstitutional due to a Charter violation, legislatures would have no recourse but for the NM, which empowers legislatures to act upon their constitutional disagreements with the courts. In contrast, in Israel the Knesset is already empowered to do this: even if the Court strikes down a piece of legislation because it is inconsistent with a Basic Law, the Knesset may amend the Basic Law. Despite this difference, the Government did not choose to amend the Basic Law to implement its plan; rather, it chose to add an NM to the Basic Law in order to secure the validity of legislation that would result in limiting the importation of non-kosher meat to Israel. On 9 March 1994, the Knesset passed this new version of the BLO including section 8, which created the NM.30 Six days later, it passed the Meat and Meat Products Law (hereinafter: Meat Law),31 a law that prohibited the importation of non-kosher meat into Israel. As required by the BLO, the Meat Law stated that it was valid notwithstanding it. Why was the NM chosen as the best way to deal with the meat importation issue, rather than a specific constitutional amendment? There is no explicit answer to this question in the debates of the Knesset’s plenum prior to the passing of the new version of the BLO. A hint appears in the explanatory notes to the Bill, which states that sometimes there is a need to address ‘a unique matter which does not necessarily belong in a Basic Law’ (emphasis added).32 The Minister of Justice used the same terminology again while HCJ 2015/91, 1775/93 (unpublished). Cited in Mitral (n 22) 493. Mitral (n 22) 497. 28 ibid 500, 504, 506–09, 515. 29 The general amending formula requires resolution by the Federal Parliament as well as seven provinces in which at least 50% of the population reside. See the Constitution Act 1982, Sch B to the Canada Act 1982 (UK), 1982, Pt V (Can). 30 BLO. The need for a new version of the BLO was unrelated to the matter of the NM. It was mostly due to the fact that both the BLO and the BLH were tasked as private bills and did not undergo the usual process and analysis in the Ministry of Justice and other relevant ministries. 31 Meat and Meat Products Law, 5752-1994 (in Hebrew). 32 Draft Bill Amending Basic Law: Freedom of Occupation, 1993, HH 128. 26 27
The Notwithstanding Mechanism in Israel 79
introducing the Bill.33 In committee, government lawyers reiterated this view, explaining that the addition of the NM to the BLO was ‘the only way’ to deal with the non-kosher meat issue.34 These statements by officials are the first indication that the Israeli NM is based on the majestic Constitution model, namely on a desire to ensure that the emerging Israeli Constitution had a certain grandeur and dignity. The second indication can be found in an important difference between the Canadian and Israeli NMs. Throughout the legislative process of the BLO amendment that created the NM, it was the Canadian precedent that endowed the idea of an NM with credibility. Reference to Canada was made by government lawyers, by the Minister of Justice, by other Knesset Members who spoke on the Bill, and by legal experts appearing before the Knesset’s Constitution, Law, and Justice Committee.35 Given that, one would expect the Israeli and the Canadian NMs to be similar. They are indeed similar in two important respects: each of them allows for: (1) a temporary deviation from a constitutional right; (2) via express declaration of the legislature. However, as the chart below demonstrates, the two mechanisms differ on perhaps the most crucial feature of an NM – the type of majority required for its invocation. Amending the constitutional document
Invoking the NM
Enacting ordinary legislation
Canada
Very difficult (Supermajority as per amending formulas)
Easy (Ordinary majority)
Easy (Ordinary majority)
Israel
Somewhat difficult (Absolute majority)†
Somewhat difficult (Absolute majority)
Easy (Ordinary majority)
Table: Invoking the NM versus amending the Constitutional Document in Canada and in Israel †
This pertains only to the BLO. As mentioned above, the BLH does not require even that for its change, which can be done by an ordinary majority of the Knesset Members.
In Canada, it is very difficult to amend the Constitution. However, invoking the NM is easy and such invocations are akin to ordinary legislation. By contrast, in Israel invoking the NM is somewhat difficult; indeed, it is just as difficult as it is to amend the Constitution. If the point of the Israeli NM was to empower the Knesset to deviate from constitutional rights, why make invoking it as difficult as making a constitutional amendment? Invoking the NM under such a regime takes on more of the hue and gravitas of a constitutional amendment than of a legislative deviation. This notion sits better with the majestic Constitution model than with the empowered legislature model. DK 133 (1994) 4329 (in Hebrew). Protocol No 161 of the Constitution, Law and Justice Committee, 13th Knesset, 3rd Sess, 39–49 (28 February 1994) (in Hebrew). This statement was made by the Vice Attorney General (Legislation), Shlomo Guberman, who was in charge of drafting the new version of the BLO. In conversation with the author (on 21 December 1994) Guberman reiterated this view and added that ‘It is not nice to write in a Basic Law: “The leg is kosher and the tail is not kosher”’. 35 See, eg the speech of the Minister of Justice, n 33 above. 33 34
80 Tsvi Kahana iii. The 1998 Amendments to the BLO It may be objected that since constitutional amendments do not normally expire, whereas the NM limits the duration of any invocation of it, the Israeli NM, like its Canadian inspiration, is about empowering legislatures and not about creating specific constitutional amendments. The response to this objection is that the temporary nature of the Meat Law turned out to be a sham. Under section 8 of the BLO, NM invocations were to expire four years after their commencement, so the Meat Law, enacted in 1994, was set to expire in 1998. However, before it expired, section 8 was amended to exempt from expiry any Acts that had been passed in the first year after the commencement of the BLO. Since the Meat Law was passed a few hours after the BLO, the exemption applied to it, and the Meat Law ceased to be temporary. Looking back at this story now, we can confidently say that what happened was not that the Knesset temporarily deviated from the BLO in order to temporarily ensure that the Meat Law was constitutional. Rather, what happened is that an amendment to the BLO took place, which made the Meat Law constitutional permanently. It happened in two stages: initially, the Meat Law was enacted as a temporary deviation from the BLO; then, the temporary became permanent. The BLO was thus effectively amended to allow the ban on the importation of non-kosher meat to Israel. Most strikingly, all this was done without using the words ‘kosher’ or ‘meat’ in the text of the BLO at any point. Consider the wording of section 8(b) of the BLO, added in 1998 to make the Meat Law permanent: (a) A provision of a law that violates freedom of occupation shall be of effect, even though not in accordance with Section 4, if it has been included in a law passed by a majority of the Members of the Knesset, which expressly states that it shall be of effect, notwithstanding the provisions of this Basic Law; such law shall expire four years from its commencement unless a shorter duration has been stated therein; (b) The stipulation concerning expiry, as per sub-Section (a), shall not apply to a Law passed within one year from the commencement of this Law.
The wording of sub-section (b) is, for a lack of a better word, comical. It reads like a general provision referring to ‘a Law’ passed within the first year; yet, whoever drafted that provision knew that there was only one Law passed within that year – the Meat Law. The legal consequences of sub-section (b) would have been exactly identical if instead of one year, it referred to one week, one month, or indeed any time period longer than six days, given that only one piece of legislation has invoked the NM. Furthermore, sub-section (b) was enacted in 1998, but since the new version of the BLO was enacted in 1994, the sub-section referred to the period of March 1994 to March 1995. Unless we find a way to go back in time, it was (and is) obviously impossible that another Act would be captured by the exception. But if the only point of the amendment was to make the Meat Law permanent, why, again, was it not explicitly mentioned by name? There is no answer to this question in the Knesset debates or in the debates of the Constitution, Law, and Justice Committee. A more honest drafting would simply state that the expiry stipulation did not apply to the Meat Law, but the drafters of the Basic Law apparently wished to avoid referring to the Meat Law or to kashrut in the Constitution, probably thinking that such reference would be unrefined and improper in the constitutional text. If you read the text of section 8(b) on its face, a nice and neat
The Notwithstanding Mechanism in Israel 81
language of principle is apparent: there is a general rule, and a general exception, phrased in general terms. Only if you study it more carefully will you discover that this general provision has the very specific purpose of invalidating the expiry of only the Meat Law. The 1994 choice to introduce a whole NM instead of simply addressing the matter of kosher meat in the text of the Basic Law may be understandable given that theoretically, other matters may necessitate the NM as well. However, the 1998 amendment’s use of the euphemism ‘a Law passed within one year from the commencement of this Law’ as opposed to simply mentioning the Meat Law could have no other explanation but a desire to avoid mentioning the words ‘kosher’ or ‘meat’ or ‘Meat Law’ in the Basic Law. The emphasis, then, was not on the temporary nature of the deviation – which did not persist. The emphasis was on avoiding calling a spade a spade and creating a constitutional amendment concerning meat importation, without subtracting from the grand language of the Constitution. This is the majestic Constitution model. Is there something to the notion of the majestic Constitution beyond niceties and ceremonial language? I turn to this question in the next section. III. THE MAJESTIC CONSTITUTION
What would have been wrong with mentioning ‘kosher meat’ in a constitutional document? The problem with a specific amendment could not be that kashrut issues were of little significance, given that the very justification given by the supporters of the Meat Law for its enactment was its importance to a Jewish State. To be sure, the issue of kashrut in the life of Israel is no less important than the issue of the minimum age of the US President, which is inscribed in the US Constitution and in the lives of the American people.36 Why not include a matter of such importance to the Israeli polity in the nation’s constitutional documents? In this part, I consider two possible reasons for keeping the Constitution majestic and general: to allow the Constitution to survive over time and generations, and to gain from the discursive and educational effects of constitutional cleanliness. Ultimately, I find them both unconvincing. A. Stability Constitutions are made to be stable, and thus should reflect a broad consensus within the polity. The more specific a Constitution is, the less likely it reflects broad consensus. If a Constitution does not maintain consensus, it will not maintain stability as various societal forces will attempt to erode it. Therefore, prudent constitution-making requires generalities. The response to this argument is that while it is true that all other things being equal, a general provision is more likely to attract consensus than a specific provision, other factors affect consensus as well. Two such factors are the intensity of the sentiment felt by a certain segment of the polity about a certain specific issue, and the way in which this intensity affects the polity generally. When a certain segment of the polity feels strongly about a specific constitutional matter and the rest of the polity is interested in appeasing that segment, that matter should be included in the Constitution. See the US Constitution, art II, §1.
36
82 Tsvi Kahana The Meat Law story exemplifies this notion. Let us consider this story from the perspective of consensus, and contrast what happened in the first stage of the story with what happened in the second stage. In the first stage, in 1994, there was consensus both about the importance of freedom of occupation generally and about the need to ensure that most meat imported into Israel was kosher. However, the lawmakers presumably assumed that by a later point in time, say in 2004, only the consensus about the general and hence majestic issue (freedom of occupation) would remain whereas the consensus about the specific issue (importance of kashrut) might dissolve. Given such a possibility, had the matter of kosher meat been included in the Basic Law, by 2004 one of its provisions could have become obsolete. The better choice was therefore to create an NM and leave the topic of kosher meat out of the Constitution, so that in 2004 all of its provisions would still reflect public consensus. In the second stage, however, it turned out that nothing had changed. There was still consensus about the importance of the specific detail of kashrut, and consequently, in 1998, as we saw above, the Knesset made the Meat Law permanent. But this is not surprising since in 1994, there was actually no reason to assume that the portion of Israelis who cared about kashrut as a symbol of Jewish identity would decrease, that the intensity of their sentiment would decrease, or that their relative power in society would decrease. In other words, there was no reason to assume that the consensus about the matter of kashrut would change after 1994.37 B. Discourse and Education Arguably, there is nothing wrong with ceremonial, subtle, and clean language. In fact, is it not better that the Israeli political actors – or legal political actors – are at least ashamed of admitting that they are manipulating the Constitution for the sake of specific issues? Is it not sometimes okay to ‘fake it until you make it’? Initially, the Israeli Constitution may only speak in a clean language and hide the stains: the Constitution would only mention the majestic idea of freedom of occupation, keeping the Meat Law within the parallel (and concealed) constitutional order created by the NM. However, eventually, after generations of Israelis discuss the Constitution, immerse in its ceremonial language, and internalise its ideas, everyone will understand that freedom of occupation is more important than the matter of meat importation, and the need for the latter will disappear altogether. This didactic and educational process will not occur if the meat issue is included in the text of the Constitution and the practice of deviating from the Constitution will continue. Moreover, Constitutions are internationally visible. A country eager to be a member of the liberal democratic family wants its Constitution to look ‘normal’ and maximally respectful of rights. If such a country seeks to make a policy that is potentially antithetical to rights, it would be wiser not to declare that policy explicitly in the Constitution. 37 As a matter of fact, if any consensus has changed at all since then, it is the consensus regarding the constitutional right to freedom of occupation. In the summer of 2011, many Israelis took to the streets to express their dismay with the Government’s economic policy and called for an increased Government role in social justice. Freedom of occupation has been traditionally viewed as an impediment to regulation and, as such, it may not currently reflect significant consensus in Israel. For an empirical argument that constitutions actually should be specific in order to be stable see Tom Ginsburg, ‘Constitutional Specificity, Unwritten Understandings and Constitutional Agreement’ in A Sajo and R Uitz (eds) Constitutional Topography: Values and Constitutions (The Hague, Eleven International Publishing, 2010) 69.
The Notwithstanding Mechanism in Israel 83
I offer two responses to this argument. First, while there are obvious advantages to pretending, the strategy also has drawbacks. The message parlayed by the constitutional array created by the BLO and the Meat Law is this: Indeed, we deviate from your constitutional rights. However, you cannot see it in the Constitution, which is ‘nice’, general, majestic. You have to delve into the nitty-gritty of legislation. You have to look not only at the Basic Law itself, but at its annex, its back yard, its hidden allies.
Put differently, the master, the BLO, is dignified and respectful; the servant, legislation created by the NM, does the dirty work. Such a message threatens to foster a cynical attitude about the Constitution, and in the long run may work to alienate and disconnect the polity from the idea of constitutional rights, and to contribute to an overall distrust in and apathy about politics and political participation. Second, many of the Knesset Members who supported the Meat Law were proud of it, and were adamant that the law did not violate the BLO. They believed that they were not overriding the BLO itself, but instead overriding the Supreme Court’s interpretation of the Basic Law. For them, rather than imposing the eating of kosher food on those who do not observe kashrut, the Meat Law worked to declare and celebrate the Jewish identity of the State of Israel by declaring the importance of kashrut.38 One may not share this perspective, but if a legislator did indeed believe that the court misinterpreted the Basic Law and that a correction to the mistake was needed, why would that legislator view the correction as a deviation from the Basic Law? On the contrary, such a legislator would instead insist that this correction should find its place in the text of the Basic Law, as the Basic Law’s true meaning. The educational message that this legislator would want to send is not that the Constitution is ‘majestic’, but to the contrary that it proudly protects the Jewish identity of the State through promoting kashrut. For this legislator, the didactic and educational consideration favours specific amendments, not majestic generalities. IV. THE NOTWITHSTANDING MECHANISM AND CONSTITUTIONAL AMENDMENT
A. Introduction In the previous part I argued that the majestic Constitution ideal is not a reason to prefer specific constitutional amendment to the employment of the NM. In this section I argue that there are good reasons to prefer constitutional amendment as the appropriate Knesset response to judicial decisions that the Knesset does not agree with. Consider this scenario, the first part of which has actually taken place. In February 2012, the Supreme Court ruled that exempting ultra-Orthodox Jews from military service is unconstitutional.39 It held that this exemption infringed upon the BLH-protected equality rights of Israelis who must serve in the army, and that this infringement could 38 See, eg the speech of MK Itzhak Levi in Protocol No 161 of the Constitution, Law and Justice Committee, 13th Knesset, 3rd Sess, 17 (28 February 1994) (in Hebrew): ‘We are allowed to disagree with the Court . . . we believe that [kashrut] is a proper purpose’ (my translation, TK). 39 HCJ 6298/07 Ressler v Knesset (21 February 2012), Nevo Legal Database (by subscription) (in Hebrew).
84 Tsvi Kahana not be saved under the limitation clause. While it is acceptable that some religious individuals may be exempted from service in order to allow them to fulfill their religious duty to study the Torah, exempting all of them is disproportionate and therefore unconstitutional. The Court gave the Knesset until 1 August 2012 to come up with an alter native arrangement that would respect the religious needs of ultra-Orthodox Jews while not disproportionately infringing the equality rights of others. As of October 2012, lawmakers were not able to arrive at an agreement concerning an alternative policy. Now we arrive at the hypothetical part. Suppose that while the Knesset cannot arrive at an alternative formula, its Members believe that the arrangement struck down by the court is actually a sound arrangement and there is a consensus among them to override the Court decision. Two options are available: the first is to amend the BLH to explicitly state that it does not apply to the exemption of religious individuals from the military service. The second is to deal with this crisis in the exact same way that the Knesset dealt with the meat importation crises: to amend the BLH in order to add an NM to it, and then to re-enact the exemption via the new NM. Which route should the Knesset take in this situation? I proceed on the assumption that there is a general commitment within the Israeli polity to rights protection. The specific definition of rights and the appropriate limits on them may be a subject of fierce disagreement, but if a system does not have a basic commitment to human rights as a general ideal, it ceases to be a democracy and is therefore beyond my analysis. One of the disagreements in Israel about the matter of rights regards the appropriate division of labour between courts and legislatures. On one end of the spectrum are those who object to any type of constitutional entrenchment of rights. For them, the democratic process in Israel can be trusted to produce an appropriate balance between the rights of individuals and minorities and those of the majority.40 On the other end of the spectrum are those who support judicial supremacy. According to their view, in order to effectively protect rights in Israel, a final decision concerning these rights should reside with judges and be beyond the reach of the majority.41 Along this spectrum are various forms of what Mark Tushnet has called ‘weak-form judicial review’42 that advocate the incorporation, in varying measure, of both the legislature and the courts in protecting the supremacy of rights vis-à-vis legislation.43 If one believes that legislatures alone ought to be responsible for rights protection in Israel, then the question whether to amend the BLH or use an NM in order to effect constitutional deviations is a question of choosing the lesser of two evils. For them, legislatures should address the merits of rights and should not worry about constitutional texts or the Supreme Court. Given that, they would probably prefer to allow legislatures to deviate from rights in legislation, since such an option brings the system closer to a regime of parliamentary sovereignty. However, if one subscribes to one of the other approaches, namely supporting either judicial supremacy or constitutional supremacy enforced by legislatures, by courts, or some partnership thereof, then the question becomes slightly more complex. In the paragraphs that follow I argue that according to these approaches there are two reasons to prefer constitutional amendments as the mode 40 See, eg R Gavizon, The Constitutional Revolution – Reality or a Self-Fulfilling Prophecy? (Jerusalem, The Israel Democracy Institute, 1998) (in Hebrew). 41 See, eg Raday (n 12). 42 See Tushnet (n 9). 43 See, eg Sapir (n 13) 265.
The Notwithstanding Mechanism in Israel 85
of constitutional deviation: the nature of constitutional deviation (section B), and the desire to reduce the number of deviations (section C). I also explain why we need not be concerned with the risk of increasing the number of constitutional amendments (section D), and conclude by arguing that constitutional deviations should not necessarily be temporary (section E). B. The Nature of Constitutional Deviation Constitutional deviations are more properly carried out by the constituent assembly than by the legislative assembly. Therefore, it is more appropriate for them to be effected in a Basic Law, thus expressing the will of the Knesset as a constituent assembly, than by a statute, invoking the NM, which expresses the will of the Knesset as a legislative assembly. My starting point is again a commitment to rights. This commitment does not depend upon any institutional setting. Different types of constitutional systems execute this commitment through different types of institutions. However, the actual right – including the right to only be deprived of a right based on certain types of considerations – must remain intact if we are to be seriously committed to rights. The deviation that is possible through the NM cannot therefore be a deviation from the right itself. Rather, it can only be a deviation from the right’s being a constitutional right – that is, a deviation from the specific way in which a particular constitutional structure secures the right. Part of what the constitutional structure in Israel provides for is judicial protection of rights. That protection, and only that protection, may be taken away via the NM.44 To clarify, my argument is not that rights are absolute. There may be some circumstances under which the State will not be obliged to respect rights.45 However, those reasons and circumstances are not dependent on the views of the constituent assembly or the legislature, but rather on principles of political morality. There is a range of legit imate options within political morality for the protection of rights, and the Constitution’s formula may be one of them. A legislature may deviate from the specific way in which the Constitution protects rights, but the policy created by the constitutional deviation must still be within the limits of what is allowed by political morality. It follows that when the deviating institution decides to deviate from the right in order to either prevent judicial review or override it, the deviator must do so on the basis of one of three reasons: 1. The deviator believes that the prescription of the text of the Constitution is not appropriate with regards to a certain statute. It wishes to replace that prescription with another prescription, which would still be respectful of rights but based on a different formula than the Constitution’s. Since the court interprets the Constitution, its interpretation ceases to be relevant. 44 In previous work, Alon Harel and I argued that judicial review is based on a right to a hearing and thus should not be deviated from. See A Harel and T Kahana, ‘The Easy Core Case for Judicial Review’ (2010) 2 Journal of Legal Analysis 227, 238–52. I leave for another day the question of whether it is legitimate for the State to deviate from this right in some circumstances. 45 According to some theorists of rights, the State must always protect rights. When there is a good reason to violate a right it means that the interest we thought was a right is not actually a right. These theorists would word the above proposition as follows: ‘there may be some circumstances under which the State will not be obliged to respect important individual interests, often protected in a constitutional Bill of Rights’.
86 Tsvi Kahana 2. The deviator keeps the prescription of the Constitution intact, but is so certain about its interpretation of the Constitution that it is comfortable substituting it for the court’s interpretation (actual interpretation in case of a responsive deviation, predicted interpretation in the case of a pre-emptive deviation). 3. The deviator keeps the prescription of the Constitution intact, but believes that the question of whether the relevant statute is constitutional is not justiciable; that is, it is not a question the court can answer. Option three is actually not independent, as it only tells us that the matter cannot be resolved (by judicial review). It does not tell us how the issue is to be resolved. To complete option three, we have to ask what happens due to the non-justiciability. The deviator then has two options. First, it could decide that it wishes to answer the question based on its own, non-judicial interpretation of the Constitution. In that case, we are back at option two. Alternatively, the deviator could decide to replace the regular constitutional prescription with a different prescription, in which case we are back at option one. Options one and two are therefore the only ones we need to address. If the deviation is carried out based on option one, then the decision to deviate resembles a constitutional amendment. Before the deviation, the constitutional prescription was applicable in every case; the deviation makes the constitutional text inapplicable to a certain case. Under option one, the deviation occurs on the same normative level of the constitutional document, and therefore should be done by the institution that created that document – the constituent assembly. If the deviation is based on option two, in that the deviating institution rather than the court interprets the Constitution, then the deviation could be effected by the legislature. While the court is more experienced in interpretation, constitutional matters are often less technical and more ambiguous than other matters of legal interpretation, and an informed Knesset Member may be just as capable of interpreting the Constitution as a Supreme Court judge. However, constitutional interpretation by legislatures raises an array of theoretical and practical problems, the most obvious of which is that legislators have a greater incentive to cater to the political preferences of their constituents than to engage in an intellectual process of interpretation. In other words, while deviation by the legislature is conceivable under option two, deviation by the constituent assembly is still preferable here. Of course, it is possible for the constituent assembly to delegate the power to deviate from the Constitution to the legislature. It could do so, for example, when it is not readily available, as is the case in Canada. There, the constituent assembly comprises various arrangements of various legislatures, according to the different formulas. In such a system, if we were to require constitutional deviations to originate from the constituent assembly they would never occur. But the fact that Canada opted for this less-than-ideal possibility, having no other choice, does not mean that Israel should also opt for this second-best option. As long as Israel does not have an amending formula that makes it more difficult for the constituent assembly to amend Basic Laws by requiring supermajority for their amendment, there is no reason why Israel should have an NM.
The Notwithstanding Mechanism in Israel 87
C. The Desire to Keep the Number of Deviations Down Requiring a constitutional amendment when the Knesset wishes to deviate from a constitutional right may decrease the number of deviations. The reason for this is the additional visibility that a deviation from a right has, if included in the constitutional text. While the Israeli public probably does not distinguish between amending the BLH and invoking the NM, it is likely that opposition parties, the media, foreign observers, and civil society groups will make rhetorical use of the fact that the deviation was done in the Basic Law itself. It is of course hard to quantify how much of the political cost will be created by the policy itself, how much will be created because of the overriding (or preempting) of a Supreme Court decision, and how much will be created because of the symbolic act of amending the BLH. However, it seems to me that at least some of the cost is due to the last reason.46 D. Too Many Constitutional Amendments? It might be objected that amending the BLH when a deviation is necessary would, over time, unnecessarily increase the number of BLH amendments. If the Knesset chooses to make the amendment temporary, then each time the Knesset renews the amendment, it would become yet another amendment. A multitude of amendments, or frequent amendments, would – so the argument goes – trivialise the Basic Laws. That trivialization, in turn, would have two implications. First, the political cost of amending a Basic Law would decrease. An erosion of the cost would in turn bring about more amendments, and so on. The second implication of the trivialization of the Basic Laws concerns the public. If the public views the Basic Laws as just as impermanent and amendable as ordinary legislation, it will feel less attached to the Basic Laws and thus in the long run develop apathy towards the idea of Israel’s (written or unwritten) Constitution. There are three responses to this argument. First the number of situations, such as the one arising in my hypothetical scenario, will not be large. Second, this argument assumes that the number of cases in which the Knesset will respond to Supreme Court decisions by either an amendment or the use of the NM remains constant, and the only relevant question is which method of response should be used. However, as I have argued, requiring a constitutional amendment in order to create a deviation would make deviations more costly. Thus, there would be at least some cases in which the Knesset would not deviate if that required a constitutional amendment. The case of the exemption from military service for the ultra-Orthodox discussed above47 is a case in point. As of October 2012, no politician in Israel has proposed to amend the BLH to allow for this exemption. Arguably, had an NM mechanism existed in the BLH, a proposal would have been made to use it. Third, the argument concerning public attachment assumes that the public 46 This argument is valid if someone believes, as I do, that in Israel the risk of overuse of the NM outweighs the risk of underuse of the mechanism, that is, the risk of an overly confident Knesset outweighs the risk of an overly activist Supreme Court. If one is more concerned about the latter, then one would want to ensure that constitutional deviations are not too difficult to execute, and therefore may prefer that they are done via an NM and not via constitutional amendment. 47 See n 39 above and accompanying text.
88 Tsvi Kahana actually distinguishes between Basic Laws and ordinary legislation so that its attachment to and respect for the Basic Law would decrease in the face of an excessive number of amendments. However, this assumption is unsubstantiated. My own impression is that when the public is interested in a constitutional issue, it is because of the subject matter or because of the clashes between the various branches of government. It is not the normative level of instrument involved that interests the public. Indeed, it is not clear that the Israeli public is attached to the Basic Laws to the same degree that either the American public is attached to the US Constitution, or the Canadian public to the Charter.48 This point requires clarification vis-à-vis the earlier point concerning the cost of amending a Basic Law. If the public does not distinguish between Basic Laws and ordin ary legislation, why is it more costly for the Knesset to amend Basic Laws? There are two answers to this question. First, the reason why it may be more politically costly to amend the Basic Law than to use the NM is not based only on polity sentiment. While legislators’ main incentive is to be re-elected, other factors may affect the cost of their decisions. Among those are media coverage, international responses, the view of scholars in the fields, the view of fellow legislators, and so on. Second, there is a difference between the concrete notion of objecting to the amendment of a Basic Law and the more abstract notion of ‘attachment’ to the Constitution. In the short term, the public may view deviation from a Basic Law in a more negative light than deviation via the NM. In the long term, the attachment of the public to the Constitution may depend upon many factors, and it is not clear that the number of amendments to Basic Laws is one of them. E. The Temporary Nature of the Deviation Both in Canada and in Israel, constitutional deviations are (at least formally) tempor ary.49 The common view is that this is not a merely formal feature of the deviations, but is rather a way to ensure that the deviation does not stay in force due to mere inertia. If the deviation is renewed, a fresh public discussion would ensue and the deviator would have to bear the political cost that may be associated with the deviation.50 However, the temporary nature of the deviation lends itself to legislation rather than to constitutional instruments. The first response to this argument is that sometimes Constitutions do include temporary provisions. They are common in transitional situations.51 As this chapter shows, the story of the Meat Law is a story of constitutional transition.52 Initially the NM was used to allow for a ban on meat importation for four years, but then the ban 48 To the best of my knowledge, there is no study about the attitudes of Israelis towards the Basic Laws, or towards the BLH. My uneducated guess is that most Israelis are not informed about the Basic Laws. This may have many causes, but one of them may be that the term ‘Constitution’ (either in Hebrew or in English) connotes something grand and dignified, whereas the term ‘Basic Law’ does not. 49 It is true that in the specific story of the BLO and the Meat Law, the stipulation that deviations are temporary turned out to be a sham, but one should not jump to conclusions based on one case. 50 See, eg L Weinrib, ‘Learning to Live with the Override’ (1990) 35 McGill Law Journal 541, 561–62. 51 Eg, s 10 of the BLO states that laws passed before its commencement will remain in force until March 2002. After that date, this provision had no legal effect. Adopted in 1998, it had a life span of four years. 52 It is possible, of course, to view the story of the BLO and the Meat Law as a complex transition from legislative supremacy, (before the BLO) to traditional constitutionalism, (after the BLO) to weak-form constitutionalism (after the addition of the NM).
The Notwithstanding Mechanism in Israel 89
became permanent. However, even if the NM is used in a non-transitional fashion, the same rationale for using a temporary amendment in a transitional context may apply to NM invocations. In both types of situations, there is a deadlock in constitutional discussions, both parties are interested in resolving the issue, and the compromise is an agreement on future revision. Of course, the desire to avoid temporary provisions may also originate from an attempt to have a majestic Constitution, but that argument is invalid with regard to temporary constitutional provisions for the same reasons it is invalid with regards to specific constitutional provisions. The more fundamental answer is that the temporary nature of the deviation is only a defining feature of the NM under the empowered legislature model. In Canada, where this model reigns, the time limit on the deviation ensures that the legislature will not be too empowered vis-à-vis the Constitution. However, in Israel, where deviations are to be done by the constituent assembly, it should be able to choose to either make them permanent or temporary. In some cases, when a compromise is necessary between supporters of the deviation and its opponents, the compromise may be a temporary deviation. When there is an agreement in the constituent assembly about the need to make the deviation permanent, as was the case with the Meat Law, it would be a permanent deviation. The Meat Law story exemplifies this point as well. Initially, it was created as a temporary deviation, but then it became permanent. That a deviation may be either temporary or permanent, like any other constitutional amendment, brings the argument full circle. Once we conceptualise the deviation as a constitutional amendment, there is no need to classify it as a deviation anymore. This has been my point throughout the chapter: in Israel, there is no need for a mechanism for legislative deviations, given that the Knesset may easily create constitutional deviations. V. CONCLUSION
One day Israel may have a constitutional Bill of Rights, which would make it impossible for the Knesset to amend Basic Laws with ease. Until then, the fact that in Israel no special majority is required to amend the rights-protecting provisions of the Constitution is no mere detail. Even if one laments the lack of entrenchment of Israeli Basic Laws (I do) and even if one predicts that it will be remedied soon (I do not), one must analyze constitutional settings and institutional designs as they are, with all their peculiarities and shortcomings. The ease with which the Knesset may amend Basic Laws should be analyzed, not ignored. The way I analyzed it in this chapter was to say that deviation from constitutional rights should be done directly within the Basic Laws. There is an irony about the importation of the NM from Canada to Israel. Its initial adoption in Canada, and certainly some of its uses, came to symbolise ugly constitutional politics, so much so that a Canadian Prime Minister pledged to bring about a constitutional amendment that will abolish the Federal Government’s power to invoke it.53 That mechanism, understood in Canada as the sign of constitutional ugliness, was imported to Israel to facilitate constitutional niceness.
53 See CBC News Canada, ‘Martin Wraps Campaign in Constitutional Pledge’, 10 January 2006 www.cbc. ca/story/canadavotes2006/national/2006/01/09/elxn-debates-look.html.
90 Tsvi Kahana It may be that the deeper explanation of the ‘niceness’ approach – the one that prefers the general kitchen discourse over the specific chicken discourse, and does not want to use the word kashrut in the Constitution – goes back to the other thing Motl tells us about his mother – that ‘half of what she says is in American’.54 A lot has been written about the language of America being a language of constitutional rights. Some say that this is good, some say that this is bad, but all agree that this is the language. Now, there is nothing wrong with speaking half of the American language. The Canadian adoption of the NM did exactly that. The NM was understood at the time of its adoption in Canada as a compromise between the British idea of parliamentary supremacy and the American idea of judicial supremacy. It becomes wrong, though, when this half is merely the ‘speaking’, and the emphasis is put not on the content but on the language. Finally, the real language of Canada – its real constitutional culture – is not only the NM. It is of a legal system which designed a mechanism that would befit its special needs. If Israel needs to address non-typical topics in its Constitution, it should do so directly. In Israel, a reference to kashrut in the Constitution should be kosher.
See n 1 above and the accompanying text.
54
7 Constitutional Adjudication and Political Accountability: Comparative Analysis and the Peculiarity of Israel YOAV DOTAN
T
HE ASPIRATION TO protect basic human rights against majoritarian tyranny forms the most fundamental justification for judicial review in modern demo cracy. On the other hand, the aspiration to guarantee that judicial supervision over majoritarian decision-making would not, by itself, create a tyrannical judiciary, is one of the most fundamental concerns regarding the institution of judicial review. The need to balance these two competing goals serves as one of the principal interests that shape the structure of constitutional judicial review. The present chapter discusses, from a comparative perspective, different models of constitutional adjudication in several national systems. Each system has found its own way to deal with the above dilemma of who will guard the guardians. Special attention is paid to the Israeli system, in which current constitutional institutions are yet to pro vide a stable and acceptable solution for this fundamental dilemma.
I. INTRODUCTION: THE DILEMMA OF JUDICIAL ACCOUNTABILITY
Judicial review is a political institution by which unelected officials (judges) retain power to second-guess decisions made by the elected branches of governments. While the democratic legitimacy of judicial review has raised various objections in academic scholarship,1 it is now a commonplace in most modern constitutional systems. The most powerful justifications for judicial review from the democratic point of view are the need to defend fundamental human rights, and the assumption that unelected judges would use their constitutional authority to counterbalance majoritarian tendencies to infringe
1 See, eg J Waldron, ‘The Core of the Case Against Judicial Review’ (2006) 115 Yale Law Journal 1346 (arguing that in reasonably democratic societies there is no reason to assume that courts would do a better job to protect fundamental rights than legislators); RH Fallon, ‘The Core of an Uneasy Case for Judicial Review’ (2008) 121 Harvard Law Review 1693 (arguing that even if courts are not better overall at identifying rights violations than are legislatures, courts have a distinctive perspective that makes them more likely than legisla tures to apprehend serious risks of rights violations in some kinds of cases).
92 Yoav Dotan on the rights of minorities, particularly at periods of national crisis.2 In recent years, the question whether – and if so, to what extent – courts are in fact fulfilling their countermajoritarian role has been much debated by lawyers and social scientists.3 It is, however, beyond question that judges, as unelected officials, carry the power to reverse decisions and policies made by the elected branches. Accordingly, the very existence of such power raises the question as to how it should be constrained. Without suitable democratic con straints, the judicial power to offset majoritarian whims arguably has a serious inherent danger to democratic accountability. The question of the democratic legitimacy of judi cial review is therefore intertwined with the question of who will guard the guardians. We want to give judges power to constrain majoritarian whims and defend fundamental human rights, but we must cope with the danger that judicial power itself may become the tyranny of the minority. Constitutional systems thus face the need to create mechan isms to deal with the democratic accountability of their constitutional judges. The discussion proceeds as follows: in part II of this chapter I discuss the various mechanisms by which constitutional systems deal with the above dilemma. I distinguish in this respect between the vehicle of the constitutional text on the one hand, and mecha nisms relating to the appointment process and tenure of constitutional judges, on the other hand. In part III, I review three models of constitutional review: the United States model, the European continental model and the common law model. I argue that each of these models strikes somewhat differently the balance between counter-majoritarianism and democratic accountability of constitutional courts. In part IV, I discuss the case of Israel. I argue that the Israeli system corresponds with the common law model of consti tutional review and that the ‘Constitutional Revolution’ has not changed this state of affairs. Accordingly, I argue that any proposal to reform judicial review in Israel should bring into consideration the need to preserve the balance between the competing values of counter-majoritarianism and democratic accountability. II. CONSTITUTIONAL MECHANISMS OF JUDICIAL ACCOUNTABILITY
The first and obvious vehicle to deal with political accountability of judicial review is the constitutional text. Arguably, judicial review is a process by which judges apply consti tutional commands to cases brought before them. Therefore, presumably, the question of the political accountability is trivial. All judges are doing is to simply apply pre- existing provisions made by the constitutional assembly (or any other body that formed 2 See, eg AM Bickel, The Least Dangerous Branch: The Supreme Court and the Bar of Politics (Indianapolis, Bobbs-Merrill, 1962); JH Ely, Democracy and Distrust: A Theory of Judicial Review (Cambridge, Harvard University Press, 1980) 43–72. 3 See, eg B Friedman, ‘The History of the Countermajoritarian Difficulty, Part One: The Road to Judicial Supremacy’ (1998) 73 New York University Law Review 333 (arguing that the counter-majoritarian difficulty is chiefly ‘an academic obsession’ since in reality judicial review tends to ratify popular preferences); B Friedman, ‘The Counter-Majoritarian Problem and the Pathology of Constitutional Scholarship’ (2001) 95 North Western Law Review 933 (same); GA Caldeira and JL Gibson, ‘The Etiology of Public Support for the Supreme Court’ (1992) 36 American Journal of Political Science 635 (arguing that courts can occasionally stand against the winds of public opinion but still maintain a high level of ‘diffuse’ public support); W Mishler and RS Sheehan, ‘The Supreme Court as a Counter-Majoritarian Institution? The Impact of Public Opinion on Supreme Court Decisions’ (1993) 87 American Political Science Review 87, 96 (arguing that public opinion influences Supreme Court decisions but its impact occurs at a moderate lag).
Constitutional Courts and Political Accountability 93
the constitution) and thus, their political accountability is based on the fact that they do not impose their own values and preferences on the polity, but rather they effectuate, by means of legal interpretation, decisions and policies made by other (presumably account able) constitutional organs. While this argument is ostensibly alluring, textual constraints can hardly provide an effective answer to the question of judicial accountability. Constitutions are usually formed for the purpose of serving the relevant communities for many years to come, and to accommodate various changes in political policies, cultural preferences and social values. Accordingly the constitutional text is often broad and vague. It refers to general terms such as ‘liberty’, ‘equality’, or ‘dignity’ without specifying the accurate, practical meaning of such general concepts. It is designated to leave ample latitude for those who read and apply it (ie judges) to answer changing needs, interests and ideologies through out the long life of the constitution. It thus leaves judges with substantial liberty to express their own views, preferences and even ideology with regard to the process of applying social values into the written text. At one historical period, the term ‘equal protection of the laws’ could be read as legitimizing policies of racial segregation, while at a later period the reading of the term was radically changed to denounce such policies as both immoral and unconstitutional.4 And, as empirical works on constitutional interpretation suggest, constitutional text is usually regarded by judges as a vehicle to empower judicial capacity, rather than to constrain such powers.5 Therefore, the consti tutional text itself can hardly serve as a meaningful constraint on judicial discretion in a way that would give an adequate answer to the problem of judicial accountability. More effective vehicles to increase political accountability in judicial review refer to the procedures of appointing constitutional judges. Judges are usually not directly elected by the public (except in some US states),6 because electing judges seems to con tradict the fundamental idea of their counter-majoritarian role. In many jurisdictions, however, the political branches are involved in the process of judicial selection. Federal judges in the United States, for example, are appointed by the President and their appointment is subject to confirmation by the Senate, by a process which is political in nature.7 In Europe, members of constitutional courts are often elected by Parliament, either exclusively or with some involvement on behalf of the executive.8 An additional accepted set of political constraints on constitutional judges refers to the judicial tenure. While in some systems (such as the federal judiciary in the US and in some common law jurisdictions) there are no mandatory limitations regarding judicial tenure, in various European countries constitutional judges are elected for fixed (and limited) periods. Thus, the political branches that control the appointment process also retain the power to make changes in the composition of the judiciary, to enhance the level of its political accountability. The combination of these institutional constraints serve in almost all constitutional jurisdictions to deal with the tension between the need for effective counter-majoritarian review in democracy, and the wish to retain some political accountability on behalf of 4 See the US Constitution, amend XIV, §1; Plessey v Ferguson 163 US 537 (1896); Brown v Board of Education 347 US 483, 493–95 (1954). 5 Y Dotan, ‘The Spillover Effect of Bills of Rights’ (2005) 53 American Journal of Comparative Law 293, 340. 6 See below, at section IIIA and accompanying notes. 7 ibid. 8 See below, at section IIIB and accompanying notes.
94 Yoav Dotan the constitutional judges. I now review in some detail three typical models of constitu tional adjudication in this respect: the United States model; the continental European model; and the common law model. III. THREE MODELS OF CONSTITUTIONAL ADJUDICATION
There are three main models of judicial review adopted by most legal systems around the world:9 the American model, the continental European model, and the common law model. I now briefly review the main characteristics of each one, focusing on the rela tions between constitutional adjudication and democratic accountability. A. The United States Model The United States model of constitutional judicial review is the most established and well-known. This model has different variants that exist at the federal level and in the various state systems. All these systems, however, share some general characteristics. Constitutional adjudication in the US is performed in the regular courts (either federal or state courts) and by regular ‘Article III judges’ (that is, judges at all levels, not only judges of the highest echelons). Ever since the famous Marbury v Madison decision10 was handed down, it has been established that the US judiciary holds the power to strike down laws that are unconstitutional. This power of the judiciary is vast, and subject only to the possibility of reversal by an amendment to the Constitution (in itself an extremely onerous procedure that very rarely occurs).11 The power vested in the judi ciary is considered to be a paramount pillar of the ‘checks and balances’ concept which is central to the American system of government. The immense power held by the regular judiciary in the United States to review and strike down legislation as unconstitutional is subject, however, to considerable political constraints at the appointment level. The constraints are designed to ensure the political accountability of those who hold immense judicial powers. While judges in the US are appointed for life (or at least for very long tenures) the appointment procedure itself enjoys a very high level of political accountability. In some states judges are directly elected either in partisan or non-partisan elections by the public, like other officials (or legislators).12 In other states (such as California) judges are appointed but their appointment is subject to approval by various procedures of referenda (retention election).13 The most common 9 A Mavcic, ‘Constitutional Courts: Historical Steps in the Developments of Systems of Constitutional Review and Particularities of their Basic Models’ in The Constitutional Review (The Netherlands, Book World Publications, 2001) 18. 10 Marbury v Madison 5 US 137 (1803). 11 Amendments to the US Constitution require a two-thirds majority in both Houses of Congress as well as ratification by the legislatures of three-quarters of all states, see the US Constitution, Art V. During the history of over 200 years the Constitution was amended 27 times. 12 See RA Karp and R Stidham, Judicial Process in America, 7th edn (Washington DC, CQ Press, 2007) 102–03; L Baum, The Supreme Court, 6th edn (Washington DC, CQ Press, 1998) 43; SP Croley, ‘The Majoritarian Difficulty: Elective Judiciaries and the Rule of Law’ (1995) 62 University of Chicago Law Review 689, 725–26. 13 See BM Dann and RM Hansen, ‘Judicial Retention Election’ (2001) 34 Loyola of Los Angelese Law Review 1429.
Constitutional Courts and Political Accountability 95
procedure of judicial selection, however, is a procedure resembling the one accepted at the federal level. It is based on appointment by the head of the executive (ie the President of the US or the Governor of the state) often after a hearing and a vote of approval by the legisla ture (or some of its organs, such as the Senate, in the federal system).14 This procedure of appointment is considered political in nature. Candidates for judicial posts are selected by the head of the executive on the basis of their known ideology and value preferences. Within the selection process, the candidates’ political affiliation is deemed a perfectly legit imate consideration. Confirmation by the legislature (ie the Senate on the federal level) is performed in public, with extensive media coverage. Candidates are asked questions regarding their political convictions and ideological preferences, as well as their profes sional and personal background. The confirmation vote (at least at the federal level) is often – though not always – dominated by partisan division, and the name of the nominat ing President as well as the political affiliation follow the appointed justice as a natural ‘mark’ near his name in any constitutional law book published after the appointment.15 Thus, the highly political nature of the appointment procedure is designed to ensure the judges’ political accountability and to solve the dilemma of democratic accountability, given the vast constitutional powers conferred on such judges. B. The Continental European Model The institute of judicial review is relatively new to the European constitutional culture. It was adopted in most European systems only after World War II (or with the Iron Curtain’s descent across Eastern Europe) as a direct response to the failure of those des potic regimes to protect fundamental human liberties, and as an acknowledgment of the importance of judicial review in this respect.16 The question that modern constitutional designers in Europe faced, however, was in which judges the constitutional power of judicial review should be vested. Unlike in the United States, appointments for the regular judiciary in continental Europe are bureau cratic in nature. Candidates are selected to serve in the judiciary at a very early stage of their professional career (usually after graduating law school and receiving special train ing to serve in the judiciary). Their service is a lifetime career during which candidates are promoted according to their professional credentials and on the basis of professional exams.17 See Karp and Stidham (n 12) 104–05. See, eg DS Law and LB Solum, ‘Positive Political Theory and the Law: Judicial Selection, Appointments Gridlock and the Nuclear Option’ (2006) 15 Journal of Contemporary Legal Issues 51, 57; G Gunther and KM Sullivan, Constitutional Law, 13th edn (New York, Foundation Press, 1997) app 3. 16 The first European country to adopt the institution of constitutional judicial review was Austria which adopted the constitutional model introduced by Hans Kelsen in 1929 (see Mavcic (n 9)). In Germany and Italy judicial review has been adopted as a constitutional institution after the fall of the fascist regimes after the War (see ibid 21), while in France judicial review was only adopted in the Constitution of the Fourth Republic in 1958 (Mavcic (n 9)). In Spain, Portugal and Greece judicial review was adopted after the fall of the fascist regimes in the 1970s and in Eastern European countries such as Poland, the Czech Republic, Hungary, Slovakia and Russia judicial review was part of the liberal constitutional reforms during the 1990s (ibid and see also J Ferejohn, and P Pasquino, ‘Constitutional Adjudication: Lessons from Europe’ (2004) 82 Texas Law Review 1671, 1675). 17 See DP Currie, The Constitution of the Federal Republic of Germany (Chicago, University of Chicago Press, 1994) 156–57; M Tushnet and V Jackson, Comparative Constitutional Law, 2nd edn (New York, Foundation Press, 2006) 470–71; GF De Andrade, ‘Comparative Constitutional Law: Judicial Review’ (2001) 3 University of Pennsylvania Journal of Constitutional Law 977, 984. 14 15
96 Yoav Dotan Therefore, when the framers of European constitutions after World War II dealt with the need to adopt the institution of judicial review, they all rejected the idea of bestow ing the constitutional authority of judicial review on their regular judiciary, due to its low level of democratic accountability. Instead, they created special judicial institutions separated from the regular judicial system to perform the mission of constitutional review. Constitutional review in most European systems is thus conducted by special constitutional courts which enjoy exclusive power to review the constitutionality of leg islation and which operate as judicial tribunals, discrete from the regular judiciary.18 Unlike the general judiciary, constitutional courts in Europe enjoy a high level of political accountability which is ensured by various constitutional measures. Most importantly, the appointment of constitutional judges is political in nature and is done by election of the judges by Parliament (exclusively, or with some involvement of the heads of the executive). In Germany, for example, half the members of the Constitutional Court are elected by the Upper House of Parliament (the Bundesrat) and half by the Lower House (the Bundestag) in a process which is political in nature.19 In France, three of the nine members of the Conseil Constitutionnel are appointed by the President of the Republic, three are elected by the Chairman of the National Assembly, and three by the Chairman of the Senate. Likewise, in Italy and Spain most constitutional judges are elected by Parliament, while others are appointed by the government or the judiciary.20 In addition, unlike the case of the regular judiciary, constitutional judges are elected for fixed and limited tenure, and thus the political branches retain some influence on the ideology of the court by changing their composition.21 Last, candidates for constitu tional tribunals do not normally come from the ranks of the regular judiciary but are rather legal academics, or even professional politicians.22 C. The Common Law Model The third and last model of judicial review is the one found in the United Kingdom and its progenies (ie systems that follow the common law model, such as Canada, Australia, New Zealand, India and Pakistan).23 Unlike the case in continental Europe, there are no special constitutional courts in common law jurisdictions. Rather, judicial review is 18 See Tushnet and Jackson (n 17) 467–68; M Cappelletti, Judicial Review in the Contemporary World (Indianapolis, Bobbs-Merrill, 1971) 68–75. 19 Grundgesetz für die Bundesrepublik Deutschland (Grundgesetz) (GG) (Basic Law), 23 May 1949, BGBl I, Art 94(1) (Ger); see Currie (n 17) 155–56. 20 See references below (n 21). 21 Supreme Court tenures are for 9 years in France, Italy, Spain, Portugal and Hungary, 10 years in the Czech Republic and 12 years in Germany. See N Dorsen and others, Comparative Constitutionalism: Cases and Materials (St Paul, Thomson West, 2003) 132; Ferejohn and Pasquale, ‘Constitutional Adjudication’ (2004) (n 16) 1681–82. 22 See Currie (n 17) 194. This, however, does not mean that the considerations for appointing judges are exclusively or even primarily political. In reality, judicial appointments are often made on the basis of profes sional expertise and are not partisan. See Ferejohn and Pasquino (n 16) 1681–82. 23 For an extensive discussion of this model of constitutionalism, see S Gardbaum, ‘The New Commonwealth Model of Constitutionalism’ (2001) 49 American Journal of Comparative Law 707, 739; PJ Yap, ‘Rethinking Constitutional Review in America and the Commonwealth: Judicial Protection of Human Rights in the Common Law World’ (2006–07) 35 Georgia Journal of International and Comparative Law 99, 107–25; M Tushnet, ‘New Forms of Judicial Review and the Persistence of Rights- and Democracy-Based Worries’ (2003) 38 Wake Forest Law Review 813.
Constitutional Courts and Political Accountability 97
conducted by the regular courts as in the United States. Unlike in the United States, how ever, judges in these systems are nominated in a process that is not considered to be political in its nature. In the United Kingdom, judges are presently appointed by the Minister of Justice, following recommendations made by the Judicial Appointment Commission (whose members belong in part to the judiciary, the legal professions, or other representatives of the public). Judicial appointments in Canada are made by the Prime Minister or the Minister of Justice;24 in Australia judicial appointments are made by the Cabinet (on the recommendation of the Attorney General);25 in New Zealand, judicial appointments are made by the Governor-General26 and in India they are made by the President after consultation with the justices of the Supreme Court.27 Although in all those systems it is officially the head of the executive who holds the power of appoint ment, in practice the process is based largely on the candidates’ professional expertise, and the nominating executive is largely bound by the recommendations made by the recommending authorities (such as the Appointment Commission in the UK). In other words, unlike the case of the US, the appointment of judges in common law systems is regarded as an apolitical process, at least in the sense that the ideological (let alone polit ical) affiliations of the candidates are officially regarded as irrelevant to the process. How then do common law systems retain political accountability in judicial review? The answer seems to be straightforward. These systems put significant limitations on the very constitutional power bestowed on the judiciary in the first place. This means that, unlike in the case of the US and Europe, the judiciary in common law systems does not possess full power to review the constitutionality of legislation. Rather, they hold some powers that are much more limited, and which allow the political branches lati tude to constitutionally respond to, and potentially nullify, any decision in judicial review. The first significant element that limits the power of judicial review in common law jurisdictions is the fact that some of these jurisdictions are yet to adopt a formal compre hensive constitution; the principle of parliamentary sovereignty is still the paramount pillar of the constitutional system. Thus, even today there is no constitution at all in the UK and New Zealand. In Australia the Federal Constitution does not include a Bill of Rights (thus limiting judicial review to rare cases of structural conflicts between the Territories).28 Likewise, in Canada, the Federal Constitution did not contain a Bill of Rights until the adoption of the Canadian Charter of Human Rights and Freedoms in 1982. Accordingly, in all the above-mentioned jurisdictions, the judiciary does not have the power to review and strike down statutes made by Parliament, as in the case of the 24 PW Hogg, Constitutional Law of Canada, 5th edn (Scarborough, Thomson Carswell, 2007) 245; MC Miller, ‘A Comparison of the Judicial Role in the United States and in Canada’ (1998) 22 Suffolk Transnational Law Review 1, 17–18. 25 M Bhattacharya and R Smyth, ‘The Determinations of Judicial Prestige and Influence: Some Empirical Evidence from the High Court of Australia’ (2001) 30 Journal of Legislative Studies 223, 229; DAR Williams, ‘The Judicial Appointment Process’ (2004) New Zealand Law Review 39, 65. 26 See the official New Zealand Government website at: www.courtsofnz.govt.nz/about/judges/appointments. 27 MP Singh, ‘Securing Independence of the Judiciary – The Indian Experience’ (2005) 10 Indian International and Comparative Law Review 245, 252. High Court appointments are also made by the President (ibid 254) while subordinate courts appointments are made by the Governors of the states (ibid 255). 28 Some Australian Territories, however, adopted Bills of Rights, see, eg the Victorian Charter of Human Rights and Responsibilities Act 2006 and see R Masterman, ‘Interpretations, Declarations and Dialogue: Rights Protection under the Human Rights Act and Victorian Charter of Human Rights and Responsibilities’ (2009) PL 111.
98 Yoav Dotan US or Europe.29 Instead, in some of those systems, relatively recent constitutional devel opments bestowed some partial powers of judicial review on the general courts. In New Zealand, under the Bill of Rights Act of 1990 (which is a regular statute by itself) state agencies (including the courts) are under a duty to interpret the law in accordance with the Bill. In the UK today, under the Human Rights Act 1998 (HRA) all government authorities are bound to respect the fundamental rights that are entrenched in the European Convention on Human Rights.30 Accordingly, UK courts have the power, under the HRA, to review Acts of Parliament and to declare them incompatible with those fundamental rights in the proper cases. Such a declaration of incompatibility does not invalidate the infringing statute, however, and it is up to Parliament to decide whether or not it wants to vacate the statute or amend it in accordance with the judicial declaration.31 In Canada, under the Charter, the framers bestowed even wider powers on the judiciary. The Charter expressly provides that Canadian courts hold the power to strike down legislation which is incompatible with the rights and freedoms enlisted by the Charter. This judicial power is, however, subject to the possibility of reversal by the legislature under the notwithstanding provisions of section 33. To sum up, common law systems do not entrench strong mechanisms to hold their constitutional judges under strict demands of political accountability. Rather, they limit the power of judicial review bestowed on the judiciary and subject judicial decisions to the possibility of reversal by the legislature. * * * We can now summarise in the following table the various solutions for achieving a bal ance between democratic counter-majoritarian aspirations and the judiciary’s political accountability:
Model
US
Europe
CL
Judicial Appointments
Political
Political
Bureaucratic/ professional
Tenure
Unlimited
Limited
Unlimited
JR Power
Full
Full
Limited
Table: Mechanisms of Political Accountability 29 India serves as a notable exception in this respect. The Federal Constitution of India bestows full judicial review power on the judiciary. See Indian Constitution, Art 13, §2; see also SP Sathe, ‘Judicial Activism: The Indian Experience’ (2001) 6 Washington University Journal of Law and Policy 29, 38–40 . 30 Convention for the Protection of Human Rights and Fundamental Freedoms (European Convention on Human Rights, as amended) (ECHR). 31 In reality however, almost in all cases in which the UK courts made such declarations, Parliament was quick to follow suit, see R Clayton, ‘The Human Rights Act Six Years Later: Where Are We Now?’ (2007) European Human Rights Law Review 11, 13. This is by no means a surprise since, according to the ECHR, the relevant parties are entitled to challenge the statute before the European Court of Human Rights whose deci sions are binding on UK authorities according to the Convention. See ECHR, Arts 32–35.
Constitutional Courts and Political Accountability 99
On the basis of the above conceptual framework, I now examine the case of judicial review in Israel. IV. THE CASE OF ISRAEL
A. The Early Days Similarly to the United Kingdom and New Zealand, Israel is one of the few countries in the world that is yet to adopt a formal, complete, unified constitution. Rather, the Israeli ‘constitution’ has been developed by a series of partial moves that are yet to be com pleted. After the establishment of the State of Israel, all formal constitutional ties with Great Britain were abolished. Nevertheless, for many years after statehood, the Israeli judicial system continued to be heavily influenced by the British legal order. While British rule ended immediately with the declaration of statehood, Israel preserved 99 per cent of the elements of the former system created by the British authorities before state hood (with the exception of some statutes that inhibited Jewish immigration to Palestine, or that in any other way conflicted with the interests of the new Zionist Government). The implication is that the structure of the court system established under the British Mandate remained largely intact. Apart from preserving the material body of law, the Israeli legal system also retained its professional affiliation with the English legal tradi tion. Therefore, English common law continued to be an important legal source for Israeli judges long after independence, and British court decisions were often the main source of reference for judges in the newborn state.32 The influence of the common law tradition has also been paramount with regard to judicial appointments in Israel. As in the case of the UK, judicial appointments in Israel are made by the Minister of Justice, acting upon mandatory recommendations by a spe cial committee. According to a statute enacted shortly after establishment, the commit tee is composed of nine members, of whom three are Supreme Court justices, two are ministers (one of whom is the Minister of Justice herself), two Knesset (Israeli Parliament) Members (one of whom is traditionally a member of the opposition) and two represen tatives of the Israeli Bar.33 The committee convenes behind closed doors and neither the political branches nor the general public are directly involved in the process. This com position of the committee as well as the procedures of appointment are designed to ensure that the process is based exclusively on considerations of professional expertise, and is detached from any reference to the candidates’ political affiliations or ideological preferences.34 The lack of a comprehensive constitutional document and the strong influence of the British common law tradition have also shaped the development of judicial review.35 The process of judicial review of governmental action was developed within a constitutional 32 A Zysblat, ‘Introduction: The System of Government’, in I Zamir and A Zysblat (eds), Public Law in Israel (Oxford, Oxford University Press, 1996) l, 4–5. 33 Basic Law: The Judiciary, s 4(b); the Courts Law (Consolidated Version), 5744-1984, c 1, pt 3. 34 See Zysblat (n 32) 14–15; Y Dotan, ‘Does Israel Need a Constitutional Court’ (1999) 5 Law and Government 117, 137–40 (in Hebrew). 35 See D Barak-Erez, ‘From an Unwritten to a Written Constitution: The Israeli Challenge in American Perspective’ (1995) 26 Columbia Human Rights Law Review 309, 317; A Rubinstein and B Medina, The Constitutional Law of Israel, 5th edn (Tel-Aviv, Schocken, 1996) 367 et seq (in Hebrew).
100 Yoav Dotan framework which was partial, without a comprehensive and clear mandate for the judi ciary to second-guess majoritarian decisions, and under the strong influence of the English tradition of the supremacy of Parliament. Israel’s ‘partial constitution’ began with the Declaration of Independence of 194836 that specifically stipulated the formation of a consti tutional assembly which would form a constitution. When the assembly was elected it decided to turn itself into the First Knesset and began to discuss several issues concerning the constitution. The political debates, however, soon revealed deep divisions on central issues, and the decision was taken – in the famous ‘Harari Decision’ of 1950 – to postpone the adoption of a constitution as one document and to prepare it piecemeal – chapter by chapter – in a series of Basic Laws that would eventually be brought together to form the constitution.37 Over the years that followed, the Knesset gradually adopted some Basic Laws, but they dealt mostly with formal and structural aspects of the governmental system rather than with issues of citizens’ rights.38 The Israeli Supreme Court – also faithful to the premises of parliamentary supremacy – was willing to acknowledge the constitutional supremacy of these Basic Laws only on rare occasions and on narrow grounds. It did so in a series of cases during the 1970s and 1980s where it acknowledged the power of the Knesset to entrench certain clauses in Basic Laws by which any legislation deviating from these clauses should meet the requirements for a special majority specified in the entrenched clauses.39 In the meantime, however, and despite the lack of a formal Constitution and a Bill of Rights, Israeli courts have been active in developing a rich jurisprudence of human rights. Accordingly, on many occasions the Supreme Court has raised the concept of an ‘unwritten’ Bill of Rights. While the concept was not cited as a justification to invalidate primary legislation, it allowed the courts to adopt a vigorous interpretative doctrine in favour of human rights, and served as a strong vehicle for controlling administrative actions infringing on human rights.40 Within this framework, the Supreme Court acknow ledged the special status of fundamental freedoms such as freedom of speech and press, freedom of association, and freedom of demonstration and procession.41 B. The ‘Constitutional Revolution’ Since the early 1980s, there has been a major shift in the policies and practices of the Israeli Supreme Court regarding judicial review of governmental action. The Court Declaration of the Establishment of the State of Israel, 1 LSI 7 (1948). See R Rubinstein, ‘The Controversy over Israel’s Bill of Rights’ (1985) 15 Israeli Year Book of Human Rights 113, 147–49; I Zamir, ‘Rule of Law and Civil Liberties in Israel’ (1987) 6 Civil Justice Quarterly 64, 66; Zysblat (n 32) 3–4. 38 See, eg Basic Law: The Knesset; Basic Law: The President of the State; Basic Law: The Government. 39 HCJ 98/69 Bergman v Minister of Finance 23(1) PD 693 [1969] (in Hebrew); HCJ 148/73 Kaniel v Minister of Justice 27 (1) PD 794 [1973] (in Hebrew) (for the English version see I Zamir and A Zysblat (eds), Public Law in Israel (Oxford, Oxford University Press, 1996) 310, 312; HCJ 246/81 Agudat Derekh Eretz v Broadcasting Authority 35(4) PD 1 [1981] (in Hebrew); HCJ 141/82 Rubinstein v Chairman of the Knesset 37(3) PD 141 [1983] (in Hebrew). 40 See Barak-Erez (n 35) 315–17. 41 See, eg HCJ 73/53 Kol Ha’am v Minister of Interior 7 PD 871 [1953] (in Hebrew) (for the English version see Zamir and Zysblat (eds) (n 39) 55, 68) (acknowledging the right of free speech as a fundamental right that cannot be infringed upon unless the danger to public order is ‘nearly certain’); FH 16/61 Company Register v Kardosh 16 PD 1209 [1962] (in Hebrew) (acknowledging the right of association as a fundamental right); HCJ 148/79 Saar v Minister of Interior and Police 34 PD 169 [1979] (in Hebrew); HCJ 153/83 Levi v Police Commissioner for the Southern District 38(2) PD 393 [1984] (in Hebrew) (acknowledging the fundamental right for demonstration and procession). 36 37
Constitutional Courts and Political Accountability 101
significantly expanded the access of NGOs, political parties and politicians to use judi cial review to challenge government policies by reforming its doctrine of standing and acknowledging the concept of ‘public law petitioner’.42 It also changed its position with regard to the doctrine of justiciability (political question) to allow petitioners to chal lenge in judicial review almost any governmental decision regardless of its political nature or implications.43 At the same time, the Court also significantly expanded the scope of judicial review in areas such as military actions, national security, foreign relations, and prosecutorial discretion.44 In 1992 the Knesset passed two new Basic Laws. The first, and most important of the two, was Basic Law: Human Dignity and Liberty.45 Contrary to the Basic Laws previ ously enacted by the Knesset, this one deals with human rights, though the list of rights it includes is partial. It grants constitutional protection to some fundamental freedoms such as human dignity, freedom of movement, privacy, and property rights. It fails, however, to refer to some of the most fundamental political freedoms such as freedom of expression, association, procession, freedom of religion and the basic right to equality. The omission was by no means a coincidence. Rather, this fragmented Bill of Rights was the result of a political compromise between the liberal forces in the Knesset that push for the adoption of a comprehensive Bill of Rights, and the opposition to this idea (formed mainly by the religious parties) that opposed the idea of the Bill altogether.46 The outcome of this struggle is a Bill of Rights that is not only partial, but in addition its constitutional status and superiority over regular legislation was not clearly provided by the text of the Basic Law itself. The text does require any future legislation to conform with the requirements of the Basic Law and to respect the fundamental rights listed therein. It also provides the judiciary with the power to review the compatibility of such legislation with the Basic Law. It does not, however, expressly entrench the provisions of the Basic Law from the possibility of being overridden by regular acts of the Knesset. Despite the limited scope of the text of Basic Law: Human Dignity and Liberty, and its unclear constitutional status, the Israeli judiciary was quick to use it as leverage to expand judicial review and to develop an ambitious concept of constitutional judicial review. In the most celebrated case of United Mizrahi Bank47 the Supreme Court declared that the adoption of the new Basic Law amounts to the transformation of Israel into a constitu tional democracy48 and that the new Basic Laws should be regarded as a constitutional Bill 42 See HCJ 910/86 Ressler v Minister of Defence 42(2) PD 441 [1988] (in Hebrew) (for the English version see Zamir and Zysblat (eds) (n 39) 275, 286–88; Y Dotan and M Hufnung, ‘Interest Groups in the High Court of Justice: Measuring Success in Litigation and in Out-of-Court Settlements’ (2001) 23 Law and Policy 1, 7–11. 43 Ressler v Minister of Defence (n 42), Dotan and Hofnung (n 42) (discussing the rise in the number of peti tions by NGOs after the decision). 44 For an overview of the phenomenon of judicial activism in Israel see, eg Y Dotan, ‘Judicial Accountability in Israel: The High Court of Justice and the Phenomena of Judicial Hyperactivism’ (2002) 8 Israeli Affairs 87; R Hirschl, ‘The Political Origins of Judicial Empowerment through Constitutionalization: Lessons from Four Constitutional Revolutions’ (2000) 25 Law and Social Inquiry 91, 98–101; GI Seidman and EA Nun, ‘Women, the Military and the Court: Israel at 2001’ (2001) 11 Southern California Review of Law and Women’s Studies 91, 101–09; G Dor, ‘Governmental Avoidance Versus Judicial Review: A Comparative Perspective in Israeli Decision-Making Strategies in Response to Constitutional Adjudication’ (1999) 13 Temple International and Comparative Law Journal 231, 232. 45 The second Basic Law is the Basic Law: Freedom of Occupation. 46 See J Karp, ‘Basic Law: Human Dignity and Liberty – A Biography of Power Struggles’ (1992) 1 Law and Government – Law and Government in Israel 323, 326–28, 339 (in Hebrew); Dotan, ‘The Spillover Effect’ (n 5) 302–03. 47 See CA 6821/93 United Mizrahi Bank Ltd v Migdal Cooperative Village 149(4) PD 221 [1995] (in Hebrew). 48 ibid para 1 of Court President Barak’s opinion.
102 Yoav Dotan of Rights. Accordingly, the Court ascertained the duty of the Knesset (as the legislature) to abide by the newly adopted constitutional framework and established the power of the courts to exercise judicial review over such legislation.49 During the past two decades, the Court took various steps to develop the constitutional scope of the new Basic Laws in order to expand judicial review over legislation. Originally, in United Mizrahi Bank, the Court refrained from providing that any Basic Law, as such, enjoys full constitutional superiority over future legislation (since most Basic Laws do not include any formal entrenchment). A few years later, however, the Court reversed its position and ruled that regular legislation of the Knesset cannot be incompatible with any provision included in any Basic Law.50 The Court also significantly expanded, by interpretation, the list of con stitutionally protected rights. Despite the fact that the new Basic Laws did not refer to fundamental political rights (such as expression) or to the right to equality, the Court ruled that flagrant infringements of political rights or equality amounts to a violation of human dignity, and thus broadened the scope of protected rights.51 The Court also used its consti tutional authority to strike down various laws passed by the Knesset as failing to meet the constitutional requirements of the Basic Laws.52 C. Democratic Deficit? The above described developments pose interesting questions with regard to the demo cratic legitimacy of constitutional judicial review in Israel. As I suggested in part III above, different constitutional systems find ways to balance counter-majoritarian aspi rations with the need to ensure democratic accountability in judicial review. Both the historic developments as well as the structural characteristics of the Israeli system clearly suggest that it belongs to the third model of common law systems. It is also clear that as far as political accountability of the judiciary is concerned Israel is located well within the boundaries of this model even after the ‘Constitutional Revolution’. Judicial review is conducted by regular judges (not by a special constitutional court) who are appointed for lifetime tenure (with a mandatory retirement age of 70) in a process that is bureau cratic and apolitical in nature, with very limited influence on the part of the political branches.53 Therefore, the comparative framework presented above suggests that Israel United Mizrahi Bank (n 47), and see references at n 44. HCJ 1384/98 Avni v Prime Minister 52(5) PD 206, 209 [1998] (in Hebrew); HCJ 212/03 Herut v Chairman of the Central Elections Committee 57(1) PD 750 [2003] (in Hebrew); A Bendor, ‘Four Constitutional Revolutions?’ (2003) 6 Law and Government 305, 307 (in Hebrew). 51 See eg HCJ 453/94 Israel Women’s Network v Government of Israel 48(5) PD 510 [1994] (in Hebrew); HCJ 5394/92 Hupert v Yad Vashem 48(3) PD 353 [1994] (in Hebrew). 52 See, eg HCJ 1715/97 Investment Managers Bureau v Minister of Finance 51(4) PD 367 [1997] (in Hebrew); HCJ 6055/95 Zemach v Minister of Defence 53(5) PD 241 [1999] (in Hebrew); HCJ 1030/99 Oron v Speaker of the Knesset 56(3) PD 540 [2002] (in Hebrew); HCJ 6427/02 Movement for Quality Government v Knesset 61(1) PD 619 [2006] (in Hebrew); HCJ 1661/05 Hof Aza Regional Council v Knesset 59(2) PD 481 [2005] (in Hebrew); HCJ 2605/05 Academic Center of Law and Business v Minister of Finance (19 November 2009), Nevo Legal Database (by subscription) (in Hebrew). 53 Recently, some Knesset Members from right-wing coalition parties proposed a Bill under which judicial appointments would be subject to public hearing and confirmation in the Constitution, Law and Justice Committee of the Knesset. See: Draft Bill Amending the Courts Law (Transparency of Procedures to Appoint Supreme Court Justices and the President and Deputy President of the Supreme Court), 2011, P/18/3423. English translation available at: www.acri.org.il/en/wp-content/uploads/2011/11/Hearings-for-Justices-BillENG.pdf. The proposal raised strong opposition from legal circles and the Attorney General expressed his 49 50
Constitutional Courts and Political Accountability 103
must stick to the common law framework of limited power of judicial review in order to remain within the boundaries of this model. From the purely formal point of view, it seems that Israel still keeps within these boundaries today. This is because the Knesset is still formally free to amend (or even repeal) the Basic Laws or to change the constitutional framework in a way that would constrain the constitutional powers of the courts (setting aside the question whether such a move would be possible at the political level). Recent developments in case law (as well as some statements by prominent judicial figures) suggest, however, that the Israeli judiciary is nearing the adoption of jurisprudence that will deny the Knesset’s very power to initiate any changes in the constitutional framework.54 The question whether or not further expansion of judicial review is needed, is beyond the scope of this chapter. However, the comparative framework presented here indicates that any move in this direction requires an answer to the question of political accountability of judicial review. Otherwise the Israeli system might face a serious problem of democratic deficit.55
doubts regarding its constitutionality. As a result, the proposal was removed after Prime Minister Netanyahu backed off from his previous endorsement of the Bill, see Y Lis and T Zarchin, ‘A Bill: Hearing in the Knesset for Judicial Candidates’, Haaretz, Tel-Aviv, 10 November 2011; Y Lis and T Zarchin, ‘Netanyahu Ordered to Drop the Bill for Hearing to Judicial Candidates’, Haaretz, Tel-Aviv, 16 November 2011. 54 Thus, eg in 2008 the former Court President Aharon Barak accused the then Minister of Justice Daniel Friedmann with intentions to ‘destroy’ the ‘judicial project’ after the latter raised proposals to initiate reforms that would extend to some extent the political accountability of the judiciary. See A Shavit, ‘Aharon Barak: A Worried Citizen’, Haaretz, 9 April 2008, and see n 55. 55 In April 2012 the Ministry of Justice introduced a draft for Basic Law: Legislation. The draft officially acknowledges the power of the Supreme Court – sitting as a constitutional court – to strike down any legisla tion that it finds incompatible with any Basic Law. See Bill Memorandum of Basic Law: The Legislation, 2012, published by the Ministry of Justice: www.tazkirim.gov.il/Tazkirim_Attachments/41283_x_AttachFile.doc (in Hebrew). On the other hand, the draft bestows on the Knesset the power to re-legislate any statute struck down by the Court provided that the legislation would be supported by 65 (out of 120) Members. It also provides that the overriding statute would be valid for five years unless the Knesset extends its validity for additional periods of five years. The draft encountered ample opposition from all sides of the political map, as well as from the bench, and its chances of passing into legislation seem unclear at best. See, eg Y Yoaz, ‘Grunis: Fix the Bill for Basic Law: Legislation so that We Shall not Regret It’, Globes, Rishon Letzion, 17 April 2012.
8 Justifying Judicial Review: The Changing Methodology of the Israeli Supreme Court JOSHUA SEGEV
I. INTRODUCTION
T
HIS CHAPTER EXAMINES the developments in jurisprudence of the Israeli Supreme Court regarding its authority to review the Knesset’s (Israeli Parliament) legislation (hereinafter: judicial review). These developments were the subject of an extensive academic examination and also triggered a hot public debate about the proper scope of the Israeli Supreme Court’s constitutional powers. Quite naturally, most of the legal writings centred on the question of legitimacy, namely, whether judicial review was justified under Israel’s constitutional circumstances. However, only in rare cases did the writers conduct their examination with full awareness of the method used by the Israeli Supreme Court, and in fact the method they themselves have used in their writings, in order to provide the answer to the question of legitimacy. This chapter tries to fill this void. This chapter is not aimed at answering the important (and done to death) question of whether the adoption of judicial review by the Israeli Supreme Court was justified. Similarly, this chapter should not be viewed as an additional (futile) attempt to convert1 the faith of supporters of judicial review, nor is it intended to defend a specific model of judicial review (activist, restraint, centralised or dispersed). This chapter examines the constitutional methodology of the Israeli Supreme Court regarding the proper way to justify judicial review. This view point will reveal three methodological phases of the legitimacy of judicial review: the pragmatic phase, the radical phase and the conservative phase. This distinction, this chapter contests, provides a deeper and accurate understanding not only of developments of the jurisprudence but also about the premises and boundaries of present constitutional discourses regarding the legitimacy of judicial review and the proper model of judicial review. The discussion proceeds as follows: first, it describes the developments and controversy about judicial review in Israel (part II). Second, it analyzes the nature and essence of the search for the constitutional legitimacy of judicial review (part III). Based on this discussion, the chapter identifies three key methods of justifying judicial review (parts IV, V and VI). This distinction will show how the Israeli Supreme Court has changed S Levinson, ‘The Constitution in American Civil Religion’ (1979) Supreme Court Review 123, 150.
1
106 Joshua Segev time and again the methodology, that is, the rules and nature of the ‘justification game’ in order to establish judicial review in Israel, and it will alert us to the danger of methodological manipulation. The danger of methodological manipulation is still looming over Israel’s constitutional future as Israel encounters old (activist versus restraint) and new (centralised versus dispersed) fronts of constitutional designs and options. II. HISTORICAL BACKGROUND: THE ADOPTION OF JUDICIAL REVIEW IN ISRAEL
The question of whether the Israeli Supreme Court should be trusted with the authority to review Knesset legislation was under intense controversy as part of the overall constitutional circumstances, which have prevented the adoption of a constitution in the formative era.2 The opponents of judicial review feared that entrusting the Court with the authority to review primary legislation would introduce a hindering force that would prevent progress and justice.3 Others, however, favoured judicial review and viewed it as a reasonable mechanism to guarantee human rights and civil liberties from the encroachment of the Government.4 These disagreements, however, were not confined to the normative propriety of judicial review. Disagreements regarding the authority of the Supreme Court to review primary legislation have often been intertwined with disagreements regarding the normative basis for such a review – namely, disagreements over the legal norms and principles that are considered superior to norms laid down in legislation, and that provide the basis for claims of unconstitutionality. Recognizing the power of the Court to review primary legislation, without resolving the disputes about the proper normative basis for such review, was unacceptable not only to those who opposed judicial review altogether but also to many of its supporters. Unable to adopt a constitution, the political parties devised a political compromise, later to be known as the ‘Harari Decision’.5 According to the Harari Decision, the Knesset would enact the Constitution gradually in incremental steps that would eventually be harmonised to form a complete constitution. However, the Decision did not address the normative status of Basic Laws vis-à-vis ordinary legislation,6 nor their potential enforceability. Nevertheless, as known to all, these circumstances did not bring an end to the adoption of judicial review or the controversies accompanying it. A short time after the establishment of the State of Israel, the newly constituted Supreme Court rejected attempts made in several cases – Leon v Acting District Commissioner of Tel Aviv (Gubernik),7 2 On Israel’s failure to adopt a constitution in the early years of the state, see J Segev, ‘Who Needs a Constitution? In Defense of the Non-decision Constitution-making Tactic in Israel’ (2007) 70 Albany Law Review 409, 412–40; R Gavison, ‘The Controversy Over Israel’s Bill of Rights’ (1985) 15 Israeli Year Book of Human Rights 113, 135; A Shapira, ‘Why Israel has No Constitution’ (1993) 37 St Louis University Law Review 283, 285. 3 Among the opponents of judicial review was Ben-Gurion, Israel’s first Prime Minister. See DK 4 (1950) 816; S Aronson, ‘David Ben-Gurion and the British Constitutional Model’ (1998) 3 Israel Studies 193. 4 DK 4 (1950) 717. See also Segev (n 2) 422–23. 5 DK 5 (1950) 1743. 6 A Shapira, ‘Judicial Review without a Constitution: The Israeli Paradox’ (1983) 56 Temple Law Quarterly 405, 410. 7 HCJ 5/48 Leon v Acting District Commissioner of Tel Aviv (Gubernik) 1(1) PD 58 [1948] (in Hebrew). English translation available at: elyon1.court.gov.il/files_eng/48/050/000/Z01/48000050.z01.htm.
Justifying Judicial Review 107
Zeev v Acting District Commissioner of the Urban Area of Tel Aviv (Gubernik),8 and Al-Karabutli v Minister of Defence9 – to assume the authority to review primary legislation. Many scholars10 have criticised the Court’s decisions in Leon, Zeev and Al-Karabutli, arguing that the Court had missed an historic opportunity to constitute a fully-fledged judicial review according to the principles set by the Declaration of Independence.11 In the late 1960s the Supreme Court assumed de facto the authority to review the Knesset legislation based on Basic Laws in Bergman v Minister of Finance.12 This and other subsequent decisions of the Supreme Court,13 also received harsh criticism on various grounds by the academic community.14 There is no doubt that the most important decision regarding judicial review of recent years is the 1995 United Mizrahi Bank Ltd v Migdal Cooperative Village.15 In the United Mizrahi Bank decision, the Israeli Supreme Court followed the steps of its American counterpart and assumed an unrestricted authority to review primary legislation notwithstanding the absence of a constitution in the conventional sense.16 This and other subsequent decisions of the Supreme Court17 have also aroused political and academic disapproval.18 In recent years a number of legislative Bills were presented before the Knesset seeking to nullify,19 or at 8 HCJ 10/48 Zeev v Acting District Commissioner of the Urban Area of Tel Aviv (Gubernik) 1(1) PD 85 [1948] (in Hebrew). English translation available at: elyon1.court.gov.il/files_eng/48/100/000/Z01/48000100. z01.htm. 9 HCJ 7/48 Al-Karabutli v Minister of Defence 2(1) PD 5 [1949] (in Hebrew). 10 See, eg P Lahav, Judgment in Jerusalem: Chief Justice Simon Agranat and the Zionist Century (California, University of California Press, 1997) 92; E Salzberger, ‘The Emergency Regulations of the British Mandatory Government – Leon v Gubernik and Zeev v Gubernik’ in D Barak-Erez (ed), Reflections upon Decisions of the Israeli Supreme Court During the First Year of Israel’s Independence (Bnei-Brak, Hakibbutz Hameuchad, 1999) 41, 45–47 (in Hebrew); D Kretzmer, ‘Judicial Review of Knesset Decisions’ (1988) 8 Tel Aviv Studies in Law 95, 97–98. 11 Declaration of the Establishment of the State of Israel, 1 LSI 7 (1948). The petitions attacked the emergency regulations of the British Mandatory Government, enacted to suppress the Jewish community in Palestine, which were used by the new Israeli Government to confiscate apartments in Tel Aviv for the private use of public officials. One of the major arguments, presented before the Supreme Court to persuade it that it possessed the authority to invalidate these regulations, was that the emergency regulations were inconsistent with the principles of ‘freedom, justice, and peace as envisaged by the prophets of Israel’, included in the Declaration of Independence and intended to guarantee personal freedom (Zeev (n 8) 88; Al-Karabutli (n 9) 13). The Supreme Court rejected this argument stating that while the Declaration of Independence defines the basic credo of the state, it is not constitutional law and does not determine whether ordinance and laws are valid or invalid (Zeev (n 8) 88–89). 12 HCJ 98/69 Bergman v Minister of Finance 23(1) PD 693 [1969] (in Hebrew). 13 HCJ 141/82 Rubinstein v Chairman of the Knesset 37(3) PD 141 [1983] (in Hebrew); HCJ 246/81 Agudat Derekh Eretz v Broadcasting Authority 35(4) PD 1 [1981] (in Hebrew); HCJ 60/77 Ressler v Chairman of the Elections commission 31(2) PD 556 [1977] (in Hebrew); HCJ 148/73 Kaniel v Minister of Justice 27(1) PD 794 [1973] (in Hebrew). 14 For representative critical writings, see C Klein, ‘A New Era in Israel’s Constitutional Law’ (1971) 6 Israel Law Review 376; MB Nimmer, ‘The Uses of Judicial Review in Israel’s Quest for a Constitution’ (1970) 70 Columbia Law Review 1217, 1222; RA Burt, ‘Inventing Judicial Review: Israel and America’ (1989) 10 Cardozo Law Review 2013. 15 CA 6821/93 United Mizrahi Bank Ltd v Migdal Cooperative Village 49 (4) PD 221 [1995] (in Hebrew). 16 See Gavison, ‘The Controversy’ (n 2) 128; H Sommer, ‘Richard Posner on Aharon Barak: The View from Abroad’ (2007) 49 The Lawyer 523, 529–30 (in Hebrew). 17 HCJ 1715/97 Investment Managers Bureau v Minister of Finance 51(4) PD 367 [1997] (in Hebrew); HCJ 6055/95 Zemach v Minister of Defence 53(5) PD 241 [1999] (in Hebrew); HCJ 212/03 Herut v Chairman of the Central Elections Committee 57(1) PD 750 [2003] (in Hebrew). 18 See Sommer (n 16) 529–30; Segev (n 2) 468–73. Although see G Sapir, Constitutional Revolution in Israel: Past Present and Future (Haifa, University of Haifa Press, 2010) 132–34 (in Hebrew) arguing that the academic community in general did not criticize the Court’s Constitutional Revolution. 19 See Draft Bill Amending Basic Law: The Judiciary (Judicial Review of Legislation), 2007, P/17/1975; Draft Bill Basic Law: Judicial Review, 2006, P/17/1864.
108 Joshua Segev least to regulate and to limit, the scope of the Court’s authority to review the Knesset’s legislation.20 Thus, it seems that as far as it concerns the adoption of judicial review the Israeli Supreme Court has always been in a situation of: ‘Damned if you do; damned if you don’t’. As aforesaid, the centre of this chapter is the methodological question of the justification of judicial review and not its legitimacy or whether its adoption was proper. However, since talks of ‘the legitimacy of judicial review’ or ‘justifying judicial review’ are deplorably vague and ambiguous, some explanation regarding the legitimacy of judicial review is necessary before one can explain the methods used by the Supreme Court to establish it. III. LEGITIMACY AND JUDICIAL REVIEW
The concept of legitimacy has at least three different meanings.21 The first understands the legitimacy in terms of ‘legality’, namely, whether the court was empowered or authorised by law to assume the authority to review the Knesset’s legislation. On this understanding, the Supreme Court would be considered legitimate in assuming judicial review if the materials of legal analysis – the accepted sources of law (such as the Basic Laws and constitutional principles) and the accepted methods of working with those sources (such as deduction and analogy) – are sufficient to conclude that the Court was authorised to assume this new role. The second way to understand legitimacy is in sociological terms.22 Namely, the legitimacy or illegitimacy of a given institution is contingent upon social norms and public attitudes toward this institution.23 Accordingly, the Supreme Court would be considered legitimate in assuming judicial review if it gained support and approval by the Israeli public. Concerns about the legitimacy crisis24 of the Supreme Court25 is often understood in this way – that is, in terms of the unique difficulties faced by the Supreme Court in preserving and maintaining public trust and confidence. The third way to understand legitimacy is in moral or political terms. Constitutional scholars and lawyers often associate the problem of the legitimacy of judicial review with the ‘counter-majoritarian difficulty’, noted by Bickel, according to which the ‘root difficulty is that judicial review is a counter-majoritarian force’,26 which needs to be reconciled with
Draft Bill Amending Basic Law: The Judiciary (Judicial Review of Legislation), 2010, P/18/2056. See also RH Fallon, ‘Legitimacy and the Constitution’ (2005) 118 Harvard Law Review 1787, 1794–801. 22 AJ Simmons, ‘Justification and Legitimacy’ in Justification and Legitimacy: Essays on Rights and Obligations (Cambridge, Cambridge University Press, 2001) 122, 131–35. 23 M Weber, The Theory of Social and Economic Organization (New York, Free Press, 1947) 114; SM Lipset, Political Man: The Social Bases of Politics (New York, Doubleday, 1960) 77; TR Tyler, ‘Procedural Justice, Legitimacy, and the Effective Rule of Law’ (2003) 30 Crime and Justice 283, 307. 24 On the legitimacy crisis characterizing modern states, see C Taylor, ‘Legitimacy Crisis?’ in C Taylor (ed), Philosophy and the Human Science, Philosophical Papers, vol 2 (Cambridge, Cambridge University Press, 1985) 248. 25 For concerns regarding the sociological legitimacy of the Israeli Supreme Court, following the adoption of fully-fledged judicial review, see E Yuchtman-Yaar and Y Peres, Between Consent and Dissent: Democracy and Peace in the Israeli Mind (Lanham, Rowman & Littlefield Publishers, 2000) 58; compare G Barzilia, E Yuchtman-Yaar and Z Segal, The Israeli Supreme Court and the Israeli Public (Tel-Aviv, Tel-Aviv University Press, 1994). 26 AM Bickel, The Least Dangerous Branch: The Supreme Court at the Bar of Politics (New Haven, Yale University Press, 1986) 16. 20 21
Justifying Judicial Review 109
representative democracy.27 Political philosophers will probably associate the problem of legitimacy with the problem of adopting criteria by which one can judge social institutions.28 According to this account of legitimacy, the Supreme Court acted legitimately in assuming judicial review if there were grounds by which one could reconcile the new role and the scheme of representative democracy in Israel or if there were grounds for finding a moral requirement for the Supreme Court to take on this role. The moral and political concept of legitimacy is connected to the problem of adopting criteria by which one can judge social institutions (what makes an institution ‘good’, ‘right’, ‘just’, ‘authoritative’ or ‘legitimate’)29 – it asks us to come up with standards and principles for judging judicial review. Political philosophy has been debating the issue, that is, the proper criteria for the assessment of social institutions, almost from the dawn of humanity with no solution in sight.30 The question has always been not so much what these standards and principles are, as how do we defend these standards and principles, since the ‘criterion requires a criterion’.31 While the problem of adopting criteria for judging social institutions is well known in political philosophy, the same is true about legal philosophy (since any legal criterion requires a criterion) and sociology (since any social criterion requires a criterion). Thus any attempt to justify (legally, sociologically or politically) judicial review rests on a two-phase process. First we must adopt criteria to judge judicial review and justify them rebutting objections made by possible sceptics arguing against the validity or propriety of these criteria. Second we must show that the criteria adopted support the adoption and preservation of judicial review. Showing that often requires us to meet or rebut certain kinds of fundamental objections (based on the standards adopted) to the institution of judicial review:32 either ‘comparative’ objections that alternative arrangements are legally, socially or morally preferable, or ‘non-comparative objections’ that the
27 J Waldron, ‘Book Review: We the People’ (1993) 90 Journal of Philosophy 149, 151. Among the recent well-known attempts to legitimize judicial review one may find: JH Ely, Democracy and Distrust: A Theory of Judicial Review (Cambridge, Harvard University Press, 1980) 87–104; BA Ackerman, We the People: Foundations (Cambridge, Harvard University Press, 1991) 16–24; S Freeman, ‘Constitutional Democracy and the Legitimacy of Judicial Review’ (1990) 9 Law and Philosophy 327. For critiques of these attempts, see R Dworkin, ‘The Forum of Principle’ (1981) 56 New York University Law Review 469; Waldron, ‘Book Review’ (n 27) 149, 151. 28 Either way, the legitimacy in the third sense is independent of any legal or social legitimacy that exists, or does not exist, in regard to the new role assumed by the Israeli Supreme Court to review the Knesset’s legislation. The existence of a legal justification, or a social requirement, never establishes (by itself) a moral requirement, although the existence of such a legal justification may be relevant to a moral justification. See AJ Simmons, Moral Principle and Political Obligation (Princeton, Princeton University Press, 1979) 21. 29 While some philosophers have used some of these terms interchangeably, others have often distinguished between these different concepts. 30 For a fundamental discussion, see J Rawls, A Theory of Justice, revised edn (Cambridge, Harvard University Press, 1999) 3–4 which defines ‘justice’ as the first virtue of social institution. See Sandel for a critique of Rawls and on the problems that accompany any philosophical efforts to come up with standards for the assessment of social institution: MJ Sandel, Liberalism and the Limits of Justice, 2nd edn (Cambridge, Cambridge University Press, 1998) 15–65. 31 Y Agassi, Letters to My Sister (Tel-Aviv, YBooks, 2000) 151 (in Hebrew); Rawls, A Theory of Justice (n 30); B Barry, Justice as Impartiality (Oxford, Oxford University Press, 1995) 160–88; Sandel (n 30); Taylor (n 24) 15–90; J Rawls, Political Liberalism (New York, Columbia University Press, 2005) 54–58; J Waldron, Law and Disagreement (Oxford, Clarendon Press, 1999); R Unger, Knowledge and Politics (New York, Free Press, 1975) 241. 32 AJ Simmons, ‘Original-Acquisitions: Justification of Private Property’ in AJ Simmons (ed), Justification and Legitimacy: Essays on Rights and Obligations (Cambridge, Cambridge University Press, 2001) 197, 200–04.
110 Joshua Segev institution of judicial review breaches the criteria adopted (ie rights, self-government, representative government, public trust or other values we hold dear). Four conclusions follow this analysis regarding the nature of the search for judicial review’s legitimacy. The first is that locating the legitimacy of judicial review does not equal an attempt to justify all kinds of judicial review and or every decision made by a court implementing judicial review. As far as I know, no one argues that courts performing judicial review are free to act as they please arbitrarily or subjectively motivated by narrow or self-interests.33 Hence, justifying judicial review is an attempt to justify a certain type (a general type or a particular type) of judicial review. Very often the search for the legitimacy of judicial review is conducted in very narrow borders, where scholars and judges try to justify a ‘local type’ of judicial review as it is practiced in the United States34 or in Israel.35 In other words, the supporters of judicial review argue that at least some kind of judicial review, ideal (ie liberal, republican, restraint, activist, centralist or dispersed) or local (American, Canadian, Israeli or European) are preferable to other arrangements or institutions, or permissible because of qualities, advantages and benefits it possesses to a given society. The second conclusion is that the search for the legitimacy of judicial review is independent of concrete or specific legal, social or moral judgments which favour a specific decision made by a court practicing judicial review. Accordingly, justifying judicial review is content-independent since it is not based directly on the specific value judgement in a specific case.36 Hence, the legitimacy of the American institution of judicial review is independent of landmark cases such as Brown v Board of Education.37 Racial segregation is a legal, social and moral wrong, and should be nullified regardless of the question of the legitimacy of judicial review. The obsession of American theories38 of judicial review trying to explain and justify Brown or trying to justify judicial review by appealing to Brown is understandable. However, the Brown decision should be followed 33 Of course more than one of the opponents of judicial review tried to depict judicial review along the lines of a court acting arbitrarily to glorify its powers and prestige. See, eg in the American context: RH Bork, ‘Our Judicial Oligarchy’ (1996) 67 First Things 21. However, the attempts to justify judicial review are not aimed at justifying judicial dictatorship, but rather at rebutting the analogy between judicial review and Plato’s ‘philosopher-king’. See, eg A Harel, ‘Rights-Based Judicial Review: A Democratic Justification’ (2003) 22 Law and Philosophy 247. Obviously, the supporters of judicial review cannot simply allow judges to rely upon their unconstrained moral views in providing justification on principled grounds. Instead, their moral views have to be embedded in a legal and political theory that indicates when they may be invoked by the judge practicing judicial review. See, eg R Dworkin, ‘The Judge’s New Role: Should Personal Convictions Count?’ (2003) 1 Journal of the International Court of Justice 4. 34 See, eg Ackerman, We the People (n 27) 296; JM Farber, ‘Justifying Judicial Review: Liberalism and Popular Sovereignty’ (2003) 32 Capital University Law Review 65. For arguments questioning the legitimacy of judicial review as it is practiced in the United States, see RH Bork, The Tempting of America (New York, Free Press, 1990) 139–41, 153–55, 251–59; Waldron, Law and Disagreement (n 31) 287–88. 35 See A Barak, ‘A Judge on Judging: The Role of the Supreme Court in a Democracy’ (2002) 116 Harvard Law Review 19, 24. 36 In regard to general accounts for the duty to obey the law, see P Soper, ‘Legal Theory and the Claim of Authority’ in WA Edmundson (ed), The Duty to Obey the Law: Selected Philosophical Readings (Lanham, Rowman & Littlefield Publishers, 1999) 213, 221–22; HLA Hart, Essays on Bentham (Oxford, Clarendon Press, 1982) 254–55; J Raz, The Morality of Freedom (Oxford, Oxford University Press, 1986) 35. In regard to judicial review see J Waldron, ‘The Core of the Case against Judicial Review’ (2006) 115 Yale Law Journal 1346. 37 Brown v Board of Education 347 US 483 (1954). 38 See, eg the three leading theories of Ely, Ackerman and Dworkin, all seeking to explain and justify Brown along the fault line of their model of judicial review: Ely (n 27) 65–66, 119–20; Ackerman, We the People (n 27) 133–40; R Dworkin, Law’s Empire (Cambridge, Harvard University Press, 1986) 131.
Justifying Judicial Review 111
because of its independent judgment and not just because the United States Supreme Court ordered the desegregation of schools. The legitimacy of judicial review is a negligible legal, social or moral addition to the requirement to avoid racial discrimination. The third conclusion derived from the analysis above, is that the legitimacy of judicial review is in large measure a ‘defensive project’, in that we ask for the justification of judicial review against a background presumption of possible objections: either comparative or non-comparative objections. Hence theories of judicial review are in nature apologetic,39 since they are required to provide a defence against the background of possible objections undermining its legitimacy. Thus we justify judicial review against objections arguing that it is undemocratic,40 counter-majoritarian,41 unjust,42 breaching the right of equal political participation,43 based on extra-constitutional standards (noninterpretivism),44 and the like. The fourth conclusion is about the ‘level’ of the justification of judicial review that is at issue. Theories concerning the legitimacy of judicial review may appear in two versions. The first is optimal theories attempting to defend the comparative judgement that judicial review is superior to alternative constitutional arrangements. Ronald Dworkin’s claim in his book Law’s Empire – that ‘the United States is a more just society than it would have been had its constitutional rights been left to the conscience of majoritarian institutions’45 – is an optimal claim and has to be supported by an optimal theory of 39 The claim about the apologetic nature of judicial review was made in the past by Professor Eisgruber, arguing that most theories of judicial review apologize for the constitution, and the only controversy is about what kind of an apology (activist or restraint). CL Eisgruber, ‘Justice and the Text: Rethinking the Constitutional Relation Between Principle and Prudence’ (1993) 43 Duke Law Journal 1, 3; and compare S Levinson, ‘Judicial Review and the Problem of the Comprehensible Constitution’ (1981) 59 Texas Law Review 395. However, Eisgruber argued that constitutional theory is on the verge of abandoning its apologetic nature since a growing number of scholars are trying to explain the authority of the constitution itself as an ultimate criteria for judging American politics. See also J Rubenfeld, Freedom and Time: A Theory of Constitutional Self-Government (New Haven, Yale University Press, 2001). In light of the analysis above, constitutional theories (and theories of judicial review as one kind of constitutional theory) cannot escape the apologetic nature, since justification always involves rebutting a presumption of possible objections. Hence, since even ‘new’ theories of judicial review, which take the constitution to be the ultimate criteria to judge politics, need to provide a defence (apology) why should we take the constitution as a normative standard to judge politics? 40 JB Thayer, ‘The Origin and Scope of the American Doctrine of Constitutional Law’ (1893) 7 Harvard Law Review 129, 154–55; EV Rostow, ‘The Democratic Character of Judicial Review’ (1952) 66 Harvard Law Review 193, 193–94; C McClesky, ‘Judicial Review in a Democracy: A Dissenting Opinion’ (1966) 3 Houston Law Review 354, 357–64. 41 Bickel, The Least Dangerous Branch (n 26). The counter-majoritarian difficulty gave birth to two kinds of theories of judicial review: 1) majoritarian theories rebutting the claim that judicial review is counter- majoritarian in any meaningful sense (RA Dahl, ‘Decision-Making in a Democracy: The Supreme Court as a National Policy-Maker’ (1957) 6 Journal of Public Law 279, 285; MJ Klarman, ‘Rethinking the Civil Rights and Civil Liberties Revolutions’ (1996) 82 Virginia Law Review 1; MJ Klarman, ‘Majoritarian Judicial Review: The Entrenchment Problem’ (1997) 85 Georgia Law Journal 491). 2) Anti-majoritarian theories which acknow ledge the counter-majoritarian nature of judicial review, but justify it on the ground of basic values and rights we hold dear. Accordingly judicial review is aimed at preventing the ‘tyranny of the majority’. (See A de Tocqueville, Democracy in America (HC Mansfield & D Winthrop trs, Chicago, University of Chicago Press, 2000) 239–42; R Dworkin, Taking Rights Seriously (Cambridge, Harvard University Press, 1977) 184–86; R Dworkin, ‘Rights as Trumps’ in J Waldron (ed), Theories of Rights (Oxford, Oxford University Press, 1984) 153, 163–64). 42 R Dworkin, Freedom’s Law: The Moral Reading of the American Constitution (Cambridge, Harvard University Press, 1996) 1–38; L Sager, Justice in Plainclothes: A Theory of American Constitutional Practice (New Haven, Yale University Press, 2004) 70–83. 43 Harel (n 33) 247; Y Eylon and A Harel, ‘The Right to Judicial Review’ (2006) 92 Virginia Law Review 991. 44 Ely (n 27) 1; Dworkin, Taking Rights Seriously (n 41) 131–49. 45 Dworkin, Law’s Empire (n 38) 356.
112 Joshua Segev judicial review.46 At least part of the controversy surrounding the legitimacy of judicial review is in the field of optimal theories of judicial review, where supporters argue for the superiority of some kind of judicial review over non-judicial review or other kinds of judicial review constitutional arrangements.47 The second is permissible theories attempting to defend the legitimacy of judicial review on its ability to satisfy some legal, social or moral requirements and thresholds. Accordingly, judicial review is defensible (and thus it is permissible for adoption) by its ability to answer some legal, social or moral threshold even if it is not legally, socially or morally optimal or clearly superior to all constitutional alternatives. Dworkin’s claim in his book Freedom’s Law – that ‘Democracy does not insist on judges having the last word, but it does not insist that they must not have it’48 – is a permissible claim and may be supported by a (weak) permissible theory of judicial review. While permissible theories of judicial review often scrutinise other constitutional arrangements, it is not aimed to rank the different arrangements within which only the top score passes, but rather to shape and constitute the legal, social or political thresholds which judicial review is required to maintain. After analyzing the nature and essence of the quest for the legitimacy of judicial review let me now turn to the developments in jurisprudence of the Israeli Supreme Court regarding its authority to review the Knesset’s legislation. My aim is to track the constitutional methodology of the Israeli Supreme Court regarding the proper way to justify judicial review by clearing the fog surrounding talks about the legitimacy of judicial review. This view point will reveal three methodological phases of the legitimacy of judicial review: the pragmatic phase, the radical phase and the conservative phase. IV. THE PRAGMATIC PHASE: THE BERGMAN PERIOD
The pragmatist49 methodology, as I define it,50 is characterised by antagonism or at least indifference to the question of the justification of judicial review. The pragmatist judge seeks to resolve the dispute at hand and to avoid deep theoretical questions concerning the justification and legitimacy of legal institutions such as judicial review.51 The leading precedent for a broad-based judicial review, which mandates the oversight of primary legislation, is the Bergman decision,52 which has been cited often as the For a critique of Dworkin’s claim see Waldron, Law and Disagreement (n 31) 287–88. See, eg Freeman (n 27) 327. 48 Dworkin, Freedom’s Law (n 42) 7. 49 The concept of legal pragmatism has been invoked by many scholars: for an elaborated discussion, see RA Posner, Law, Pragmatism, and Democracy (Cambridge, Harvard University Press, 2003). 50 For a similar definition to pragmatism, see DA Farber, ‘Legal Pragmatism and the Constitution’ (1988) 72 Minnesota Law Review 1331. 51 See also RA Posner, ‘Against Constitutional Theory’ in N Dorsen (ed), The Unpredictable Constitution (New York, New York University Press, 2002) 217; CR Sunstein, One Case at a Time: Judicial Minimalism on the Supreme Court (Cambridge, Harvard University Press, 1999). 52 Bergman (n 12). In 1969, the Knesset enacted the Knesset Elections Law and Local Authorities (Funding, Limits on Expenditure and Control), 5729-1969 (hereinafter: Financing Law), which provided for governmental financing of political parties in election campaigns. Dr Aharon Bergman, a lawyer, brought an action, apparently in his capacity as an individual elector, before the Israeli Supreme Court, seeking to restrain the Minister of Finance from paying any expenses under the Financing Law. Dr Bergman’s argument was that the Financing Law was not duly enacted, since it contradicted the principle of electoral equality entrenched in s 4 of the Basic Law: The Knesset, according to which: ‘The Knesset shall be elected by general, national, direct, equal, secret and proportional elections, in accordance with the Knesset Election Law; this section shall not be varied save by a majority of the members of the Knesset’. Dr Bergman’s complaint was that the Financing Law unlawfully 46 47
Justifying Judicial Review 113
Israeli Marbury v Madison.53 However, Bergman is far from being Marbury54 as one examines it in regard to the legitimacy of judicial review. In fact, the question regarding the court’s authority to review Knesset legislation based on the Basic Laws was not raised explicitly in Bergman. Rather Justice Landau in that case, treated the question of justiciability and of the Court’s authority to review primary legislation, as interchangeable, and referred to the Attorney General’s position that it was unnecessary to decide the question of justiciability, since even if the matter was held to be justiciable, the petitioner would nevertheless lose on the merits. Justice Landau took advantage of this position,55 left this issue as ‘requiring further consideration’, and proceeded to resolve the dispute on the merits without expressing any opinion on the underlying constitutional questions.56 Deciding in favour of the petitioner on the merits, Justice Landau ordered the Minister of Finance not to enforce the Financing Law and hence, invalidated it de facto without (again) addressing its authority to invalidate Knesset legislation. In this way, Justice Landau avoided the highly divisive question of whether the Israeli Supreme Court was justified in assuming judicial review. Of course, Justice Landau’s decision, which de facto invalidated the Financing Law without resolving the fundamental questions of justiciability and judicial review, was criticised by many scholars. Robert A Burt even argued that ‘Bergman is actually the least important basis for the Israeli Supreme Court’s contemporary claims for increased judicial authority’.57 The Bergman decision did not offer any (legal, sociological or political) justification to the Court’s authority to review the Knesset’s legislation. Justice Landau did not impose his vision of the proper constitutional framework upon the Government and the Knesset. discriminated against new political parties since it provided public financing only for those parties that were represented in the outgoing Knesset. According to Dr Bergman, such inequality was unlawful mainly since the Financing Law was enacted by the Knesset by less than the required absolute majority of its total members, as is established by s 4 of the Basic Law: The Knesset. In other words, the petition maintained that the last clause of s 4 gives the section its entrenched feature, distinguishes it from ordinary legislation, and authorizes the Supreme Court to invalidate the Financing Law, which contradicts the prescription of equal election declared by this section. 53 Marbury v Madison 5 US 137 (1803). See, eg Justice Zamir in United Mizrahi Bank (n 15) para 1. See also GJ Jacobsohn, Apple of Gold: Constitutionalism in Israel and the United States (Princeton, Princeton University Press, 1993) 124; A Barak, ‘Freedom of Speech in Israel: The Impact of the American Constitution’ (1988) 8 Tel Aviv Studies in Law 241, 243: ‘Our Marbury v Madison is about to be pronounced – or maybe it was several years ago’. But see M Shamgar, ‘On the Written Constitution’ (1974) 9 Isrraeli Law Review 467, 474: ‘Bergman did not pretend to be an Israeli Marbury v Madison but after it, the idea of further development of judicial review of legislation could not be regarded as entirely unexpected’. 54 On the academic value of Marbury, in the teaching and enquiring about the legitimacy of judicial review, see S Levinson, ‘Why I Do Not Teach Marbury (Except to Eastern Europeans) and Why You Shouldn’t Either’ (2003) 38 Wake Forest Law Review 553; EJ Segall, ‘Why I Still Teach Marbury (And So Should You): A Response to Professor Levinson’ (2004) 6 University of Pennsylvania Journal of Constitutional Law 573; S Levinson, ‘Reply: Why I Still Won’t Teach Marbury (Except in a Seminar)’ (2004) 6 University of Pennsylvania Journal of Constitutional Law 588. 55 When the Attorney General was asked by the court what his position would be in the event that the Court determined that there was a violation of s 4 of the Basic Law: The Knesset, the Attorney General responded that ‘in such event, he was ready to be at the disposal of the Court and express his opinion on the question of justiciability’. Bergman (n 12) 696. From this exchange, the court decided to leave the question of justiciability for further consideration, since ‘for obvious reasons, the material problems facing us in these hearings call for speedy resolution, and consideration of the preliminary constitutional questions would necessitate a lengthy hearing on its own’. ibid. 56 ibid. 57 Burt (n 14) 2043. See also Y Zemach, ‘How was Judicial Review in Constitutional Matters Established?’ (1975) 29 The Lawyer 515 (in Hebrew).
114 Joshua Segev The tactic of leaving the question of judicial review open, continued in subsequent decisions. In Kaniel v Minister of Justice,58 the Court rejected the petitioner’s claim that an amendment to a Basic Law should have been enacted by a subsequent Basic Law. In light of this result, the Court, again, found no need to consider the preliminary questions regarding the justiciability of the issue or the authority of the Supreme Court to review Knesset decisions.59 In Agudat Derekh Eretz v Broadcasting Authority,60 in which new parties challenged an amendment to the Financing Law, the Court, once again, alluded to the Attorney General’s consent not to raise the issue of justiciability since he sought the advice of the Supreme Court on the substantive controversy. In Rubinstein v Chairman of the Knesset,61 however, the Attorney General asked for a decision on the merits, avoiding fundamental questions over the jurisdiction of the Court, but the Court took a slightly different position. Justice Levin noted: As the number of petitions concerning constitutional matters, with which the Court agrees to deal increases, the less chance there is that the Court will refrain from dealing with such cases in the future, even if the Attorney General raises these questions and questions similar to these.62
Many scholars believed that Justice Levin’s remark recognised that the point of no return had been reached on the ability of the Supreme Court to review Knesset legislation. By deciding this series of cases, the court has pragmatically assumed authority to review primary legislation according to entrenched provisions included in Basic Laws and made disputes over this question moot.63 It did so, however, not by providing a convincing political, sociological or legal justification: that judicial review makes the Israeli polity a more just society (optimal justification), that it is in accordance with the principles and essence of a democratic regime (permissible justification), that it is legally essential (optimal justification), or that it maintains public trust (permissible justification). Rather, the issue of the Court’s authority became a moot question by deciding the constitutionality of Knesset legislation on a case by case basis. V. THE RADICAL PHASE: THE MIGDAL DECISION
The radical methodology,64 as I define it, is characterised by a belief in the Court’s abilities to engage in supplying reasons that ‘answer the question of legitimacy’65 in a very broad sense. Moreover, the radical methodology requires us to reveal the ‘genetic code’ of the justification of judicial review by appealing – not to traditions and long standing practices – but to abstract legal, social and political reasons and principles. Thus, the role of the Court is far beyond resolving a specific dispute. In fact, its role is even beyond the ‘merely defensive’,66 and is not confined to the invalidation of a particular statute or Kaniel v Minister of Justice (n 13). ibid 798. 60 Agudat Derekh Eretz (n 13). 61 Rubinstein v Chairman of the Knesset (n 13). 62 ibid 148. 63 Kretzmer (n 10) 117. 64 For a basic discussion of radicalism and its elements in philosophy, see Y Agassi, The History of Modern Philosophy: From Bacon to Kant (1600–1800) (Tel-Aviv, Ramot, 1993) 35–56 (in Hebrew). 65 BA Ackerman, Social Justice in the Liberal State (New Haven, Yale University Press, 1980) 311. 66 Rawls (n 31) 235. 58 59
Justifying Judicial Review 115
regulation. It is, in a sense, an exemplar of public reason,67 as it is assigned the role of articulating a theory of state (legal, sociological or political), part of which is a theory justifying judicial review.68 In 1995, following the enactment of the two new Basic Laws, the Court was called upon to decide the United Mizrahi Bank case.69 Many fundamental constitutional questions were raised, discussed, and determined, necessarily and unnecessarily, by a bench of nine Supreme Court Justices. This was a lengthy discussion, consisting of about 360 pages of legal analysis. Court President Barak, joined by a majority of the Court,70 outlined his theory that Israel had achieved since 1992 fully-fledged judicial review, and that Basic Law: Human Dignity and Liberty enjoyed normative superiority; hence new legislation that infringes rights protected by the Basic Law must satisfy the requirement of section 8 – also known as the limitation provision. As aforesaid, one of the fundamental issues discussed in the United Mizrahi Bank decision was the propriety of judicial review. The new Basic Laws do not include an explicit provision regarding the remedies available in cases of a violation of enshrined human rights, which fails to satisfy the limitation provisions. Court President Barak in United Mizrahi Bank presents a four-fold argument in favour of judicial review over primary legislation. The first proposition adopted by Barak is that the basis for judicial review of the constitutionality of primary legislation is the principle of the rule of law.71 The central role of the court in a democratic society is the duty to maintain the rule of law, which includes guarding the law from governmental transgressions. In countries that have a constitution, the rule of law mandates the protection of the constitution. ‘The constitution and the Basic Law themselves’, Barak ibid. R Dworkin, ‘In Praise of Theory’ (1997) 29 Arizona State Law Journal 353, 375. 69 United Mizrahi Bank (n 15). The United Mizrahi Bank decision was given by the Supreme Court presiding as an appellate court on civil matters. Among the decisions appealed in United Mizrahi Bank was an appeal of a District Court decision invalidating a law because it was unconstitutional under the new Basic Laws. The District Court judge had decided that an amendment to the Family Agriculture Arrangements Law, 5752-1992, (more commonly known as the ‘Gal Law’, named after the Knesset Member who promoted it, Gedalya Gal) is void since it transgresses the right of property included in s 3 of the Basic Law: Human Dignity and Liberty. The Gal Law was enacted to relieve the economic crisis of the agricultural sector. It was enacted in 1992, prior to the enactment of Basic Law: Human Dignity and Liberty; hence it was protected from judicial review by s 10, the immunity provision, which states that the validity of legislation enacted before the Basic Law came into force will not be affected. The Gal Law stipulated that a trustee should oversee the restoration efforts of agricultural organizations, and established restrictions on the collection of debts from agricultural organizations and associations as well as other proceedings of a similar economic nature which were meant to relieve the pressures of creditors. Although the Gal Law itself was immune from judicial review, its amendment, the Family Agriculture Arrangements Law (Amendment), 5753-1993, was enacted after Basic Law: Human Dignity and Liberty came into effect, and therefore was exposed to constitutional review in light of the new Basic Law. The amendment to the Gal Law contained two major changes: extension of the period in which debts could be created, which debts were to be handled by the trustee, as well as changes in the definition of debts and their assessment. Among the decisions appealed to the Supreme Court, was the decision that the amendment to the Gal Law transgressed the creditors’ property rights enshrined in s 3 of the Basic Law: Human Dignity and liberty. Since the transgression, according to the District Court judge, did not satisfy the requirement of the limitation provision, the amendment to the Gal Law was declared void by the District Court judge. To make a 300 page story short, the Supreme Court determined that although the amendment to the Gal Law did, in fact, transgress the property rights of creditors and hence violated s 3 of the Basic Law, this transgression was allowed since it satisfied the standards of the limitation provision. The petitions could have been decided on fairly narrow grounds – that the Gal Law did not clash with the Basic Law – which would have made the constitutional deliberation of the United Mizrahi Bank decision unnecessary. The justices, however, were eager to express their opinions on the so-called Constitutional Revolution, its sources, its meaning, and its scope. 70 Justice Cheshin was the only justice who dissented. 71 United Mizrahi Bank (n 15) para 78 of Court President Barak’s opinion. 67 68
116 Joshua Segev writes, ‘legitimise the judicial review of constitutionality’.72 The second proposition introduced by Barak is that judicial review is derived from the principle of separation of powers.73 The role of the legislator is to legislate, while the responsibility of the court is to interpret the constitution when a dispute arises as to the proper interpretation of the constitution. Hence, according to Barak, when a court declares a law unconstitutional and thus void, it implements the principle of separation of powers rather than violating it. The third proposition is that judicial review over legislation is an ‘integral element of the essence of democracy’,74 since democracy is a delicate balance between majority rule and fundamental values (among them protecting minority rights). The fourth proposition, argued by Barak, is that judicial review is an expression of constitutional values. Judicial review, according to Barak, is not a manifestation of the judges’ subjective will, but rather a reflection of fundamental social perceptions provided by the mechanism of judicial objectivity.75 As striking as it may seem, the bench of nine Supreme Court Justices unanimously agreed that the Supreme Court itself possesses the power of judicial review over primary legislation even though the Basic Laws did not explicitly empower it to do so. We have seen in the previous chapter that up until the 1980s the Supreme Court avoided, as much as possible, the question of the legitimacy of judicial review. Although the prevailing view among the academic community was that the decisions from Bergman to Rubinstein v Chairman of the Knesset assumed the authority to review primary legislation, under entrenched provisions in the Basic Laws, the Court has not offered explicitly or unequivocally a theory asserting and legitimizing such authority. Thus the United Mizrahi Bank decision indeed signifies a revolution – a methodological revolution – since the Supreme Court has changed its approach to the way by which questions of judicial review of primary legislation should be answered. Questions about the legitimacy of judicial review should be directly confronted and deliberated upon systematically and profoundly. This has led not only to the expansion of judicial review over primary legislation (which is the direct result of the Constitutional Revolution), but also to the creation of a radical constitutional theory. This theory developed independently of past traditions and practices and conflated the question of judicial review with ideas such as democracy, the protection of human rights, and justice.76 As explained above, this chapter is not meant to examine the strength of the arguments77 put forward by the court to justify judicial review, but rather to draw our attention to the fact that the methodology in justifying judicial review has changed, and alert us to the danger of methodology manipulation. One should notice the following two insights regarding Court President Barak’s methodology. First, Court President Barak confronts the question of the legitimacy of judicial review very broadly, namely, in all the three major fronts: legal legitimacy, moral legitimacy and social legitimacy. Accordingly, the first two arguments should be interpreted in legal terms, that is, that ibid. ibid para 79. 74 ibid para 80. 75 ibid para 81. 76 As Professor Gavison correctly points out, the mere fact that some democracies exist without the institution of judicial review suggests that nothing in the idea of democracy itself requires such a review. See R Gavison, ‘The Role of Courts in Rifted Democracies’ (1999) 33 Israeli Law Review 216, 221. 77 For a critical analysis of these and other arguments in favour of judicial review see ibid 221–25; Waldron, Law and Disagreement (n 31) 282–312. 72 73
Justifying Judicial Review 117
the legality of the Basic Laws provides the legality of judicial review and that the legal principle of separation of powers provides the adoption of judicial review. Hence, he mounts a legal defence against those who question the claim that judicial review is legally required or derived from accepted legal materials.78 The third argument should be understood in moral and political terms, that is, that judicial review is part and essence of the attractive moral ideal of a democratic regime. The fourth argument should be viewed as answering the question of social legitimacy, namely, rebutting the objection that judicial review is a manifestation of the judge’s own subjective perception and is not supported by social and public perceptions. Second, Court President Barak’s arguments are not clear in regard to their level of justification: whether the defence of judicial review in the United Mizrahi Bank decision is an optimal defence or only a permissible defence. This ambiguity is not accidental. It enables Court President Barak to use the radical methodology without confronting head-on the burning legal, social and political questions that arise from the Supreme Court’s shift in deciding to assume a fully-fledged judicial review. The first two arguments and the fourth argument seem to defend judicial review on its ability to satisfy the principles and requirements of the rule of law, the separation of powers and fundamental social perceptions. The third argument seems to aim much higher, since the claim is that judicial review is part and parcel of a true democracy, establishing, arguably, its superiority over non-judicial review constitutional arrangements. Is judicial review ‘good enough’ or ‘the best’ constitutional arrangement according to Court President Barak? One explanation of Court President Barak’s approach is that optimal arguments are much harder to prove than permissible arguments (all things being equal).79 Thus while Court President Barak initially attempted to show that judicial review is the best possible democratic institution, he is pushed to make a much weaker permissible argument in favour of judicial review based on the virtue that it possesses – the rule of law or separation of powers. The second explanation for the ambiguity of Court President Barak’s level of justification is much more sophisticated and attributes to him the motivation (even if it is on a subconscious level) to avoid really hard questions of the legitimacy of judicial review under Israel’s constitutional circumstances. I will explain: while Court President Barak is far from being clear about the ties that link the different arguments together it is important to see that the fact that judicial review has virtues, benefits and advantages will probably provide comfort to free-will participants in the enterprise of judicial review: judges, lawyers and petitioners to the Israeli Supreme Court. Their choice to participate and take part in this enterprise was no thoughtless choice nor was it a choice to participate in an immoral or illegal arrangement. However, Court President Barak’s 78 A second way to understand Court President Barak’s first argument is by interpreting legitimacy in moral and political terms. However, this interpretation is open to serious objections, related to the circumstances of judicial review in Israel. According to the moral interpretation, the moral and political propriety of Basic Laws provides the moral and political propriety of judicial review. First, some scholars and public representatives believe that the process by which Basic Laws have been enacted is not legitimate. See, eg M Tushnet, ‘Book Review: The Universal and the Particular in Constitutional Law: An Israeli Case Study’ (2000) 100 Columbia Law Review 1327. Second, it has also been argued that the new Basic Laws were not meant to endow the Supreme Court, as currently composed, with the full power of judicial review. Third, the overall justness or moral propriety of the new Basic Laws themselves has been attacked. Fourth, it has been argued that justice, as such, does not require the establishment of judicial review. See Waldron, Law and Disagreement (n 31) 287–89. 79 Simmons, ‘Justification and Legitimacy’ (n 22) 124.
118 Joshua Segev argument is not aimed only at wilful participators, but also at legislators, public representatives and ordinary citizens, who consist also of judicial review’s sceptics, that is, those who questioned the legitimacy of judicial review; did not choose it and are subject to it against their free will. A permissible level of justification to judicial review, combined with a radical methodology, seems a complete non sequitur. Showing that it is permissible to adopt judicial review is very different from showing that Israel’s citizenry did adopt judicial review. Showing the last proposition must include some kind of historical data and propositions that would support the conclusion that judicial review, is not only permitted for adoption, but also was in fact (as a matter of real history legal, political or social) adopted by the Israeli citizenry. For example, that the adoption of judicial review is the result of a (real) political compromise, process or transaction. Without that, Justice Barak’s abstract arguments, which are not based on concrete history and long-standing practices, simply beg the question: why should Court President Barak (joined by a bench of eight justices) have the right or authority to decide about the adoption of judicial review for us all? In order to bypass this objection Court President Barak’s theory of legitimacy must move from the permissible level to the optimal level. The optimal level of justification asks us to adopt the best feasible constitutional arrangement: if judicial review is the ‘best’ constitutional arrangement, Israel’s citizenry must choose it even if as a matter of historic fact they did not choose it. Hence, the ambiguity of Court President Barak’s level of justification: a permissible level of justification combined with a radical methodology do not do the job of legitimizing judicial review under Israel’s circumstances; an optimal level of justification combined with a radical methodology can do the job (since it is not contingent upon Israel’s unique constitutional history), but is highly questionable and divisive theoretically.80 VI. THE CONSERVATIVE PHASE
The conservative81 methodology, as I define it, is characterised by scepticism about our ability to base the legitimacy of judicial review on a limited number of goals, patterns and principles. Social institutions – and judicial review is no exception in this regard – are too complicated and unique to be reduced to a theory consisting of few principles and rules. The conservative methodology does not question the legitimacy of judicial review, but rather the human ability to grasp this legitimacy isolated from the social history of a given society. Accordingly, the conservative thought bases the legitimacy of judicial review upon social relationships, norms and conventions which are formed by an historic genealogy of precedents, traditions and shared conventions. Our current constitutional arrangement is rational although its rationality cannot be grasped in isolation from its concrete historic development, facts and circumstances. See Waldron, Law and Disagreement (n 31); Gavison, ‘The Role of Courts’ (n 76). On constitutional conservatism, see and compare: M Tushnet, ‘Conservative Constitutional Theory’ (1985) 59 Tulane Law Review 910; R West, ‘Progressive and Conservative Constitutionalism’ (1990) 88 Michigan Law Review 641; G Lawson, ‘Conservative or Constitutionalist?’ (2002) 1 Georgetown Journal of Law and Public Policy 81; AM Bickel, The Morality of Consent (New Haven, Yale University Press, 1975) 3–30; DA Strauss, ‘Constitutions, Written and Otherwise’ (2000) 19 Law and Philosophy 451; CR Sunstein, ‘Burkean Minimalism’ (2006) Chicago Public Law and Legal Theory Working Paper No 116: www.law.uchicago.edu/files/files/116-crsburke.pdf; E Young, ‘Rediscovering Conservatism: Burkean Political Theory and Constitutional Interpretation’ (1994) 72 North Carolina Law Review 619; TW Merrill, ‘Bork v Burke’ (1996) 19 Harvard Journal of Law and Public Policy 509. 80 81
Justifying Judicial Review 119
In 2003 the Supreme Court was called to decide Bank Leumi v Hazan.82 The Hazan decision dealt with another amendment to the Family Agriculture Arrangements Law, 5752-1992.83 The Law’s previous amendment was at the centre of the United Mizrahi Bank decision. Hence, the Hazan decision possessed the potential of a re-examination (or at least a further evaluation) of the Constitutional Revolution – its benefits and costs, including the costs and benefits of adopting judicial review. Not only that this potential was not fulfilled, but rather we witness a conservative revolution in the Court’s methodology. Justice Ayala Procaccia (joined by Justices Mazza and Englard) opened her constitutional analysis by declaring that Basic Law: Human Dignity and Liberty establishes the protection of the right to property and presupposes the prohibition of infringement upon this right without satisfying the Limitation Clause. The conservative reasoning quickly follows this announcement: The Basic Law, according to its simple linguistic meaning, and according to its interpretation which has put down roots in Supreme Court’s jurisprudence and precedents, establishes its superiority not only over secondary legislation but over primary legislation as well.84
And she continues: The Limitation Clause in the Basic Law establishes the superior status of the Basic Law . . . This ruling was established by Court President Barak in the United Mizrahi Bank precedent . . . a failure to satisfy the constitutional requirements turns the law to be unconstitutional . . . The court may declare its nullification.85
Notice the significance of precedents in Justice Procaccia’s reasoning for establishing the Court’s authority to review the Knesset legislation. Furthermore, the conservative methodology can be identified in Justice Barak’s own judicial reasoning post the Constitutional Revolution. In the Movement for Quality Government in Israel v Knesset86 the petitioners sought the drafting for the military of students of religious institutions called Yeshivot by the nullification of the Knesset’s legislation exempting Yeshiva students from enlistment to the military. Justice Asher Grunis argued in favour of adopting the theory of judicial review presented by John Hart Ely in his book Democracy and Distrust.87 According to Ely’s theory, Grunis explained, the role of the court in performing judicial review is to keep the democratic channels open and protect weak and isolated minorities. A statute which privileges the religious minority on behalf of the secular majority should not be nullified.88 Court President Barak responded by arguing that Ely’s theory cannot explain Israel’s constitutional history: In my opinion, Ely’s theory is wrong from within; it is based on a narrow and improper conception of democracy; it has not been accepted in the United States where it was conceived, and was not adopted by other countries that practice judicial review. It is foreign to our constitutional structure and it contradicts all our constitutional decisions.
LCA 3145/99 Bank Leumi v Hazan 57(5) PD 385 [2003] (in Hebrew). See n 69 above. 84 Hazan (n 82) 397. 85 ibid. 86 HCJ 6427/02 Movement for Quality Government v Knesset 61(1) PD 619 [2005] (in Hebrew). 87 Ely (n 27). 88 Movement for Quality Government (n 86) para 9 of Grunis J’s opinion. 82 83
120 Joshua Segev . . . Ely’s approach . . . asked to explain the interpretation of the American constitution . . . Harsh criticism was leveled against it even in the USA . . . the accepted opinion is that it is not a proper explanation to the status of the American constitution. The American Supreme Court, through all its generation, did not accept it . . . Be Ely’s approach as it may, it could not be anchored in Israel’s constitutional structure. My friend Justice A. Grunis did not explain at all, how can Ely’s approach be adopted in Israel? There is no explanation at all how it could fit this court’s rulings since the Migdal decision. Was the Bank . . . a minority group worthy of protection? Was this the reasoning upon which rest the judgments in that decision?89
In his response to Justice Grunis, President Barak continued to review other court decisions post the Constitutional Revolution arguing that Grunis’s theory cannot explain these decisions.90 VII. CONCLUSION
Former US Vice President Spiro Agnew, complained bitterly after the Watergate scandal blew up: ‘Suddenly the rules are changed in the middle of the game’.91 It is a well-known proposition that the Constitutional Revolution signified a departure from Israel’s past constitutional practices, rules and decisions – and hence the rules of Israel’s constitutional arrangements have changed. However, this chapter comes to the conclusion that some of Israel’s constitutional rules have changed again after the Constitutional Revolution. This chapter examined the constitutional methodology of the Israeli Supreme Court regarding the proper way to justify judicial review. This view point revealed three methodological phases of the legitimacy of judicial review: the pragmatic phase (prior to the Constitutional Revolution), the radical phase (the Constitutional Revolution) and the conservative phase (post Constitutional Revolution). This distinction showed how the Israeli Supreme Court has changed its methodology, that is, the rules and nature of the ‘justification game’ in order to establish judicial review in Israel. It would not be a huge leap to assume that the last change was not a coincidence as well: it was meant to preserve the achievements of the Constitutional Revolution. However, the preservation of these achievements was made at the expense of changing the rules in the middle of the game, which goes against the fundamental concept of constitutional fairness.
ibid paras 75, 78 and 79 of Court President Barak’s opinion. ibid para 79 of Court President Barak’s opinion. Another example of the conservative shift in Court President Barak’s methodology could by identified in his position regarding the model of judicial review that should be adopted: centralized or dispersed review. In early cases and writings Justice Barak wrote in favour of a dispersed form of judicial review, but in recent articles he has argued that the centralized version of judicial review better fits our legal tradition A Barak, ‘Judicial Review of the Constitutionality of Statutes: Centralism v Decentralism’ (2005) 8 Law and Government 13, 21–22 (in Hebrew). Accordingly it would be a sharp turn from a parliamentary democracy to a constitutional democracy if every court could review Knesset legislation. As long as public perception is not used to the idea of judicial review it would be better to concentrate the authority to review Knesset legislation in the hands of the Supreme Court alone. 91 S Alsop, ‘President Agnew: It has Serious Sound?’, Newsweek, New York, 29 July 1973. 89 90
9 The Democratic Case for Diffuse Judicial Review in Israel ORI ARONSON*
I. INTRODUCTION
T
HE DESIGN OF a legal system that includes an element of judicial review of the constitutionality of legislation entails a choice of constitutional jurisdiction. Two general models are most common in the world’s constitutional demo cracies: a diffuse model of judicial review, in which all courts within the judiciary are empowered to hold a statute unconstitutional, and a concentrated model of judicial review, in which the power to strike down legislation is entrusted with a single central ised forum. In Israel, the institutional vehicles for exercising the power of judicial review are yet to be settled in an authoritative way, and are therefore subject to countervailing ideologi cal, intellectual, and jurisprudential currents. In the jurisdictional context, which is the focus of this chapter, Israel has so far avoided making a choice. Its system of judicial review contains elements of both the diffuse and the concentrated models, in formations that are sometimes complementary and at other times contradictory. I will define this condition as the ‘double hybridity’ of Israeli constitutional jurisdiction. As public debate continues over Israel’s evolving constitutional order, the normative discussion as to the ultimate choice between the diffuse and concentrated models of judi cial review has plotted an exceptionally consensual trajectory towards the concentrated end of the jurisdictional spectrum: commentators and stakeholders of different convic tions find themselves in a curious coalition that favours concentrated judicial review for Israel. While actual reform along these lines is yet to happen, the emerging preference of a concentrated system of judicial review warrants attention and critique. In this chapter I offer an analysis of the current jurisdictional state of affairs in the field of judicial review, as well as an evaluation of the political and intellectual discourse that has been reimagining it for the past two decades. These lead to a critique of what I find to be a prevalent capture with a Supreme Court dominated understanding of consti tutional adjudication. This reality, I argue, overlooks the redeeming potential of trial court adjudication for a deliberative, participatory, and pluralistic process of creating * I am grateful to Menny Mautner and Dori Spivak for helpful comments and insights, and to Dana Cohen for research assistance.
122 Ori Aronson constitutional norms and understandings; that is, for a profoundly democratic kind of constitutionalism. Because Israel did not experience a popularly engaging ‘constitutional moment’, and because it is replete with cultural and ideological divisions, the need for democratizing its processes of constitutional lawmaking is especially pronounced. Therefore, as Israel proceeds in its journey towards a fully established constitutional order, I hope to show that the democratic promise of the diffuse model of judicial review should not be missed. I begin, in part II, with some definitions of the main concepts used throughout the chapter. Part III examines the jurisdictional state of judicial review in Israel, describing it as a complex hybrid of diffuse and concentrated elements. Part IV tracks and critiques the ongoing debates in Israeli politics and academe over the future of judicial review. It identifies the curious coalition that has converged in support of a concentrated model of judicial review for Israel. Part V presents a normative intervention in the discourse, which stresses the unique characteristics of trial court adjudication and the democratiz ing potential they might entail for Israeli constitutionalism. The implication is that the diffuse model of judicial review holds a promise that ought not to be neglected. Part VI concludes. II. DEFINING THE CONCEPTS
Following are brief introductory definitions of the central concepts of the discussion that follows. These are not definitive framings, rather generalizations that seem to capture prevailing understandings and usages. Judicial review. ‘Judicial review’ is the act of a court, which strikes down a piece of legislation, having found that the legislative content violates a superior constitutional norm. The definition as I use it here is limited, therefore, to the review of legislative acts alone, and not other governmental acts (eg executive or administrative). The purpose here is to limit the discussion to those judicial acts, which are explicitly ‘counter- majoritarian’ in Bickelian terms.1 Further, the definition concerns only review of legisla tive content – I leave aside challenges to legislative validity that are based on the various (constitutional and other) norms regulating the legislative process. Such claims raise arguably distinct jurisprudential questions that this chapter does not treat. The power of judicial review can emanate from a constitutional text, as it does in many of the world’s current constitutions;2 or it can result from a judicial interpretive act, backed by some social-political mechanism of legitimation, as has been the case in the United States.3 Be that as it may, the notion of judicial review assumes a normative system that subjects regular legislative acts to a higher constitutional order, and hands judges the authority to enforce this hierarchy. It seems beyond dispute that the constitu tional system in Israel today includes at least some version of judicial review.
1 AM Bickel, The Least Dangerous Branch: The Supreme Court at the Bar of Politics (New Haven, Yale University Press, 1962). 2 See, eg Grundgesetz für die Bundesrepublik Deutschland (Grundgesetz) (GG) (Basic Law), 23 May 1949, BGBl I, art 93 (Ger); Canadian Charter of Rights and Freedoms, 17 April 1982, s 24; S African Constitution, 1996, s 167. 3 Marbury v Madison 5 US 137 (1803).
Diffuse Judicial Review 123
Diffuse judicial review. In a diffuse system of judicial review all courts within the sys tem are in principle authorised to review legislation for constitutionality.4 Just like other commonplace acts of legal interpretation, findings of constitutionality and, more con sequentially, of unconstitutionality are made initially by a trial court of original juris diction, and may then proceed up the judicial hierarchy by way of appellate review. Like other legal determinations made by lower courts, a finding of unconstitutionality entered by a trial court applies only to the parties involved in the given case; the reviewed statute, if somewhat wounded, retains in principle its force in all other cases.5 The diffuse system of judicial review accepts, therefore, at least temporary phases of incoherence, uncertainty, and even contradiction in the constitutional order: while one trial court may invalidate a law, another may find it constitutional. These phases are likely to end as cases reach the higher courts of the system and ultimately the Supreme Court, which provides an authoritative resolution to the constitutional question, bind ing all other courts through the power of precedent. The United States and Canada are familiar examples of diffuse systems of judicial review. Concentrated judicial review. Concentrated, or centralised, systems, locate the power of judicial review with a single tribunal, often a separate specialised constitutional court.6 The concentrated system denies inferior (and sometimes supreme) courts the jurisdiction over matters of constitutional interpretation and application, and instead institutes mechanisms for channelling all constitutional challenges directly to the single body charged with the power of judicial review.7 These may include direct petitions to the constitutional tribunal, as well as certification of questions of constitutional validity that arise during the processes of legislation or litigation to the centralised forum.8 Concentrated judicial review works in a once-and-for-all fashion: a constitutional chal lenge arrives, in principle, only once before the constitutional court, and its determina tion is final and binding, in the sense that no other court (nor, arguably, any governmental body) can subsequently revisit the same question.9 Because constitutional courts in a concentrated system of judicial review are often specialised tribunals whose specialization treads the explicitly political, their personal makeup and appointment procedures may (and often do) vary from those of regular courts, and sometimes incorporate more openly political and less legal-professional ele ments.10 Concentrated judicial review is the more prevalent model in the world’s consti tutional jurisdictions. Notable examples are Germany, South Africa, France, and Italy. Types of constitutional challenges notwithstanding. The distinction between diffuse and concentrated judicial review is often comingled with the types of constitutional chal lenges presented to the court: the centralised model is often linked to direct challenges of 4 See generally AR Brewer-Carías, Judicial Review in Comparative Law (Cambridge, Cambridge University Press, 1989) 127–35; M Cappelletti, The Judicial Process in Comparative Perspective (Oxford, Clarendon Press, 1989) 133–35; A Stone Sweet, ‘Why Europe Rejected American Judicial Review and Why it May Not Matter’ (2003) 101 Michigan Law Review 2744, 2770–71. 5 This account is complicated by the remedy of injunction, which may be worded in comprehensive terms, and is assumed to apply to a whole industry or to a governmental body. Class actions in constitutional contexts can also have a similar effect. 6 Brewer-Carías (n 4) 185–94; Cappelletti (n 4) 136–46. 7 VF Comella, Constitutional Courts and Democratic Values: A European Perspective (New Haven, Yale University Press, 2009) 5–6. 8 ibid 7–8. 9 ibid 9. 10 ibid 39–45.
124 Ori Aronson unconstitutionality (ie processes in which the remedy sought is the annulment of the stat ute) as well as to abstract challenges (challenges to the statute as a whole, regardless of how it would be interpreted and enforced); while the diffuse model is often attached to indirect challenges (raised primarily as a defence in the context of a proceeding in which the statute is invoked) and to as-applied challenges (in which a certain effect or application of the statute is put under review). While these correlations are often accurate, they are not necessary and counter-cases can arise: consider for example a diffuse–direct suit for declar atory or injunctive relief; or a concentrated-as-applied petition seeking to limit the scope of a statute even before its entry into force. The implication is that the choice between diffuse and concentrated judicial review – or some combination of the models – cannot be derived from prior jurisprudential or procedural assumptions. Rather it has to rely on our understanding of institutional com petence and political preference. III. JUDICIAL REVIEW IN ISRAEL: A STATE OF DOUBLE HYBRIDITY
Absent deliberate design, the institutional formation of the practice of judicial review in Israel has evolved haphazardly into an intriguing combination of diffuse and concen trated elements, constituting what may be termed ‘double hybridity’. In order to evalu ate this institutional state of affairs, a brief primer on the jurisdictional framework of the Israeli Supreme Court is necessary. I will follow with a separate examination of the diffuse and the concentrated aspects of the Israeli system of judicial review, and then reflect more comprehensively on the overall jurisdictional design that emerges. A. The Israeli Supreme Court: A Case of Mixed Jurisdiction The Israeli Supreme Court consists (currently) of 15 justices, who normally adjudicate cases in mostly random panels of 3. The Court President can decide to enlarge the panel when a uniquely important issue is under consideration, and such enlargements – some times even to panels of 9 or 11 justices – have become more common in the past two decades, along with the deepening role of the Court in deciding politically contested matters and the introduction of judicial review during the same period.11 The Court’s panels sit interchangeably in two kinds of jurisdictional capacities: 1. High Court of Appeals. The Supreme Court is Israel’s court of last resort for all appeals from the civilian court system, which consists broadly of two lower levels of trial and mid-appellate jurisdiction. Depending on subject matter and procedural his tory, the Court hears some of the appeals as a matter of right, while exercising discre tionary jurisdiction over the rest. Supreme Court decisions carry a precedential effect that binds all lower courts,12 such that the Court is the final settler of civil, criminal, and administrative law through its appellate jurisdiction. 2. High Court of Justice. The Supreme Court is also a court of original, non-discretionary, jurisdiction for petitions for writs of mandamus against the state and its organs. Through See S Navot, The Constitutional Law of Israel (Hague, Kluwer Law International, 2007) 139. Basic Law: The Judiciary, s 20(b).
11 12
Diffuse Judicial Review 125
its jurisdiction as High Court of Justice, the Court has over the past six decades devel oped a rich body of constitutional and administrative law, sometimes backed by legisla tion but often through its original creation by way of common law. Since the late 1980s, the justices have lifted most of the standing and justiciability barriers that have tradi tionally checked the flow of direct petitions to the Court. Consequently, the Court cur rently considers about 2,000 HCJ petitions every year, filed by individuals, public interest organizations, and public officials, challenging the full variety of governmental action (legislative, executive, administrative, religious and military), both in the abstract and in the context of specific violations. Each of these petitions is disposed of by a panel of at least three justices. While each of the Supreme Court’s two jurisdictional purviews involves a different set of procedures – notably, appeals are decided based on records developed by inferior courts; while petitions are premised on parties’ submitted (and rarely challenged) affida vits – the practical distinctions between them are not highly significant. Thus, a Court’s panel will regularly hear both sorts of cases during a single sitting, depending on mere scheduling factors; cases of both kinds which deal with similar issues are often joined and heard in tandem; and, as a matter of public perception, the Court’s two functions are often mixed, with the epithet ‘Bagatz’ – signifying the Hebrew acronym for High Court of Justice (HCJ) – used as a general reference for the Supreme Court. B. Israel’s Diffuse System of Judicial Review As a formal jurisdictional matter, Israel currently upholds a diffuse system of judicial review. In the absence of a written constitutional framework constructing the practice of judicial review and defining its jurisdictional boundaries, the power of judicial review has been recognised by the Israeli Supreme Court in a ‘Marburian’ moment in the famous United Mizrahi Bank case of 1995, as the power of any court to consider the constitutionality of a statute it is about to apply.13 Indeed, the very historic case of United Mizrahi Bank – heralding what is known as Israel’s ‘Constitutional Revolution’14 – arrived at the Supreme Court in its capacity as a High Court of Civil Appeals, following the decisions of two lower courts to strike down a statute due to its violation of a consti tutionally guaranteed right to property.15 In United Mizrahi Bank, the Supreme Court classified the newly recognised con stitutional texts – the 1992 rights-protecting Basic Laws – as higher laws; that is, as valid instances of the legislature’s constitutional lawmaking authority.16 Although of a constitutional level, the Basic Laws were understood as pieces of legal matter that were 13 CA 6821/93 United Mizrahi Bank Ltd v Migdal Cooperative Village 49(4) PD 221, 418 [1995] para 77 of Court President Barak’s opinion (in Hebrew). 14 See, eg R Hirschl, ‘Israel’s “Constitutional Revolution”: The Legal Interpretation of Entrenched Civil Liberties in an Emerging Neo-Liberal Economic Order’ (1998) 46 American Journal of Comparative Law 427, 428–31. 15 A right to personal property was included in Basic Law: Human Dignity and Liberty. The Basic Law was determined by the Supreme Court in the United Mizrahi Bank case to constitute an initial Bill of Rights and to empower courts to review the constitutionality of ‘regular laws’ against its rights-guarantees. 16 This of course closely echoes Marshall CJ’s reasoning in Marbury (n 3) 177, which itself drew on Alexander Hamilton’s similar arguments in A Hamilton, ‘Federalist No 78’ in The Federalist Papers 465 (New York, NAL Penguin, 1961).
126 Ori Aronson part of the prevailing legal order, applicable (like all other laws) to any legal dispute, regardless of the forum in which they are invoked. Reflecting this understanding, the Supreme Court in several subsequent cases has sent parties to raise indirect constitu tional challenges in the context of ongoing litigations in inferior courts rather than pres ent a direct and abstract challenge to the High Court of Justice;17 and in a single instance so far it has struck down a law in its capacity as a High Court of Appeals.18 In the absence of any further legislative action in the constitutional field, therefore, the Israeli judicial system has been functioning under the assumption that any and all courts are in principle empowered to strike down statutes due to a violation of constitutional rights-guarantees. The Supreme Court, in this picture, is a mere appellate instance that gets a chance to review constitutionality only pursuant to lower court decisions. Realizing the availability of a new field of legal claims, litigants and attorneys have been regularly arguing the unconstitutionality of various statutes in mostly indirect challenges within criminal, civil, and administrative procedures in the multiple levels of trial and appellate jurisdiction. The various inferior courts have been duly considering the merits of these claims (with jurisdiction normally uncontested) and, as a tentative canvassing of the case law reveals, rejecting their vast majority, save very few instan ces.19 Hence the characterization of the Israeli system of judicial review as only formally diffuse: despite its uncontested legality, diffuse judicial review has been scarcely employed in Israel and has had practically no effect on the validity of any law. Israel’s trial courts retain the jurisdiction to strike down statutes, and have incorporated the relevant discourse in their judicial vocabulary, but so far have shied away from putting their newly discovered power to actual and resonating use. Of the few cases in which trial courts did declare a law unconstitutional, only one made it to public consciousness, and is therefore worth reflecting on in a little detail, as an exhibit of how judicial review by a lower court has been perceived and rewarded. In the 2003 criminal case of Handelman, a magistrate court, which is the lowest of the three main instances in Israel, held unconstitutional a provision of the Income Tax Ordinance (New Version).20 The defendant was acquitted from the relevant charge, and the Government, likely for strategic reasons, did not appeal the constitutional determina tion, effectively limiting its impact to that single individual. Nonetheless, in deciding the defendant’s appeal from his conviction of several other charges, the District Court, sitting as a mid-appellate level, chose to reprimand the magis trate judge for ‘hasting’ to strike down the provision, instead of working harder to reach a 17 See, eg HCJ 8424/01 Handelman v Income Tax Commissioner (30 September 2002), Nevo Legal Database (by subscription) (in Hebrew); HCJ 2055/02 Oubeid v Minister of Defence (12 December 2002), Nevo Legal Database (by subscription) (in Hebrew); HCJ 7190/05 Lobel v Government of Israel (18 January 2006), Nevo Legal Database (by subscription) (in Hebrew). The Court stated in these cases the discretionary nature of its power to grant a remedy when an alternative venue exists for presenting the constitutional claim. In many other cases it has, nonetheless, agreed to consider direct constitutional challenges – these will be described in the fol lowing section. See below, text accompanying nn 23–24. 18 CHR 8823/07 John Doe (Anon) v State of Israel (11 February 2010), Nevo Legal Database (by subscrip tion) (in Hebrew). 19 I am currently aware of only four cases in which an Israeli trial court has effectively struck down inter partes, a statutory provision. See CrC 4696/01 State of Israel v Handelman (14 April 2003), Nevo Legal Database (by subscription) (in Hebrew); SC (PT) 940/07 Taruf v Iberia (8 July 2007), Nevo Legal Database (by subscription) (in Hebrew); SC (Ac) 1457/07 Inbar v Iberia (20 November 2007), Nevo Legal Database (by sub scription) (in Hebrew); NI (Naz) 1822/09 Yedinek v National Insurance Institute (7 January 2010), Nevo Legal Database (by subscription) (in Hebrew). A full empirical survey is still ongoing, however. 20 State of Israel v Handelman (n 19).
Diffuse Judicial Review 127
constitutionally sustainable interpretation.21 In addition, despite the decision’s extremely limited application and its conformity with the diffuse logic of Israeli constitutional juris diction, the fact that a low-level court was allowed to review the validity of a piece of pri mary legislation was greeted with much alarm by prominent politicians, signalling this case as an example of the unchecked judicial power unleashed by the ‘Constitutional Revolution’.22 Although its place in Israeli tax law, or in Israeli constitutional law, was wholly negligible, Handelman turned out to be a telling case of constitutional politics: it revealed some of the mechanisms that may work to dispirit trial-level constitutional activ ism, as well as the processes of education and acculturation that would be required in order to turn the Israeli system of judicial review from formally to genuinely diffuse. C. Israel’s Concentrated System of Judicial Review Stemming from the dual jurisdiction of its Supreme Court, Israel currently upholds a concentrated system of judicial review alongside its diffuse design. When it presides as a High Court of Justice – a court of original (first and last) jurisdiction – the Court regularly considers direct constitutional challenges to the validity of legislation, often (but not neces sarily and not always) brought by public interest organizations and argued in the abstract. Because its jurisdiction is defined in extremely broad terms, including the authority to ‘give relief for the furtherance of justice’ and to order any governmental organ to ‘act or to cease from acting’,23 and because it has historically enjoyed the legitimacy to review the full vari ety of governmental action, the High Court of Justice is assumed to have the power to entertain and to decide direct petitions for judicial review. In this capacity, the Court func tions much like the centralised constitutional courts prevalent in other systems. Thus, of the 10 instances in which the Supreme Court has so far explicitly invalidated a statute on rights-based constitutional grounds since 1995, nine have been under its HCJ jurisdiction. In all nine cases the Court had accepted wholly abstract direct chal lenges to the constitutionality of statutes or provisions therein, and held them invalid erga omnes.24 These nine cases are a mere fraction of the petitions filed with the Court calling it to exercise its power of judicial review. As in the diffuse context of trial 21 CrimA (TA) 70597/04 Handelman v State of Israel (1 December 2005), Nevo Legal Database (by subscrip tion) para 17 (in Hebrew). 22 See, eg Z Lavie, ‘The Citizens of Israel May Have Judges But They Have No Law’, Globes, 24 April 2003: www.globes.co.il/news/article.aspx?did=683107 (in Hebrew) (interview with Knesset Member Michael Eitan); Hadas Magen and Itamar Levin, ‘Rivlin: Aharon Barak’s “Constitutional Revolution” Has Turned Into a Governmental Coup’, Globes, 22 May 2003: www.globes.co.il/news/article.aspx?did=690837 (in Hebrew) (reporting comments by Knesset Speaker Reuven Rivlin). 23 Basic Law: The Judiciary, 5744-1984, ss 15(c), 15(d)(2). 24 HCJ 1715/97 Investment Managers Bureau v Minister of Finance 51(4) PD 367 [1997] (in Hebrew); HCJ 6055/95 Zemach v Minister of Defence 53(5) PD 241 [1999] (in Hebrew); HCJ 1030/99 Oron v Speaker of the Knesset 56(3) PD 640 [2002] (in Hebrew); HCJ 1661/05 Hof Aza Regional Council v Knesset 59(2) PD 481 [2002] (in Hebrew); HCJ 8276/05 Adalah v Minister of Defence 62(1) PD 1 [2006] (in Hebrew); HCJ 2605/05 Academic Center of Law and Business v Minister of Finance (19 November 2009), Nevo Legal Database (by subscription) (in Hebrew); HCJ 6298/07 Ressler v Knesset (21 February 2012), Nevo Legal Database (by sub scription) (in Hebrew); HCJ 10662/04 Hassan v National Insurance Institute (28 February 2012), Nevo Legal Database (by subscription) (in Hebrew); HCJ 8300/02 Nasser v Government of Israel (22 May 2012), Nevo Legal Database (by subscription) (in Hebrew). The Zemach case is somewhat unique, in that it included a peti tion of a detained soldier to invalidate the statute that authorized the period of his detention. By the time the case was heard, however, the petitioner’s personal interest had become moot, and the case turned into an abstract challenge.
128 Ori Aronson court adjudication, attorneys, public interest groups, and political activists (including Parliament Members) have recognised the newly available avenue for raising legal claims, and have used it profusely. The Court, in turn, rejects most of these petitions, but not for want of jurisdiction. In contrast to the diffuse avenue of judicial review, direct petitions to the Supreme Court for judicial review are of unique magnitude. The stakes in each case are immea surably higher, since the result of the Court’s decision as to the validity or invalidity of the challenged statute is final, supreme, and binds all courts and other branches of government. The concentrated system of judicial review is premised on a one-shot, winor-lose, model for challenging the constitutionality of legislation, and so each case is of critical significance to those implicated by the law under review. In concentrated judicial review the potential counter-majoritarian impact of the Court’s decision is at its extreme, and at the same time so are the agency and visibility of the petitioners who bring the constitutional claims. The incentive to pursue constitutional claims by direct petition to the Supreme Court is therefore significant, as are the risks embedded in this strategy. D. Synthesis: The Double Hybridity of Judicial Review in Israel In this formative period of its constitutional life, Israel has two concurrent jurisdictional structures for the exercise of judicial review. It is therefore best described, at least for mally, as a hybrid system, which combines elements of diffuse and concentrated judicial review: constitutional challenges can be pressed and decided either in the context of trial court litigation or in a direct petition to the Supreme Court, and multiple outcomes can potentially ensue in terms of the breadth of decisions’ application and the consequent institutional possibilities for litigants, judges, and legislators. However, as we have seen, the concentrated elements of the Israeli system of judicial review are much more pronounced than its diffuse elements. While inferior courts are in principle authorised to strike down legislation, they seem to have internalised a norm discouraging them from doing so. At the same time, the systems of public and profes sional discourse are all geared towards directly and immediately engaging the Supreme Court in questions of constitutional validity. And indeed, although constitutional chal lenges are regularly brought in both avenues, it is the Supreme Court, sitting as a High Court of Justice, that has been effectively exercising the power of judicial review in almost all such cases. But Israel’s High Court of Justice is a relatively unique institution in the comparative variety of concentrated structures of judicial review. Because it convenes as a mere ses sion of the ‘regular’ Supreme Court, the lion’s share of whose docket is dedicated to appellate jurisdiction, its justices are appointed by the same appointment procedures as all other judges in the civilian court system,25 and are mostly picked from within the cadre of lower court judges.26 Their professional expertise is primarily legalistic, and the politics of their selection are tamed by the significant professional and bi-partisan ele 25 With a slight difference: the nomination of a Supreme Court justice requires the support of a special majority of seven out of a nine-member appointment commission, while the nomination of all other judges requires a simple majority of five. See the Courts Law (Consolidated Vesrion), 5744-1984, s 7(c)(2). 26 Of the 15 currently serving Supreme Court justices, 12 have served previously as judges in inferior courts (10 were elevated directly from a lower bench).
Diffuse Judicial Review 129
ments that are involved in the appointment process.27 Israel’s High Court of Justice, therefore, is not a specialised constitutional court. It was not established in order to distinguish constitutional adjudication from the regular business of the judiciary, and its personal makeup does not involve distinctly political elements, or an exceptional com petence in constitutional doctrine. Israel’s High Court of Justice, title aside, is primarily a court. At the same time, Israel’s High Court of Justice is not a regular ‘supreme’ Court. Its original jurisdiction, which combines broadly-worded legislative authorization, very lib eral doctrines of standing and justiciability, and the absence of an effective discretionary mechanism for sorting petitions, renders it to a great extent a trial court of sorts: enter taining some 2,000 petitions a year,28 many of which entail courtroom hearings, the Court offers a degree of intensity and accessibility that is usually lacking from the more solemn and aloof conditions of other high courts, in both diffuse and concentrated sys tems. This is the second kind of hybridity apparent in the Israeli constitutional system: its High Court is perhaps best described as a ‘supreme trial court’, in which constitu tional litigation takes place on a constant and recurrent basis. The state of double hybridity in Israeli constitutional adjudication – jurisdiction that is both diffuse and concentrated, a High Court that is both trial and supreme – has char acterised Israel’s legal system since its inception, and has intensified during the past two decades, when judicial review was thrown into the fray as an added element in the dual function of the Supreme Court. The institutional forces of path dependency and a con sistently gridlocked political system have worked to leave this curious jurisdictional structure in place. Calls for reform, however, have been raised. I turn to those now. IV. THINKING ABOUT REFORM: THE STATE OF THE DISCOURSE
A. In Search of Institutional Possibilities Some two decades after the landmark case of United Mizrahi Bank, it seems that the legitimacy of some form of judicial review is now fairly settled in Israel. There is a legal, political, and intellectual recognition of the idea of a hierarchy of norms as an organiz ing principle of Israeli law, and an acceptance of an essential role of a court-like tribunal in enforcing the constitutional order. Therefore, while some participants in the discourse still question the validity of the introduction of judicial review in Israel by way of a judicial interpretive act, much of the focus has now turned to weighing the various institutional and jurisdictional alterna tives for the exercise of judicial review. As examples below will show, these are invoked and considered both as possibilities of reform within the current evolving constitutional structure, and as parts of comprehensive projects of drafting a new Constitution, or a new Bill of Rights, for the State of Israel – if and when one is finally adopted. While such activities have been quite prolific, involving prominent political, academic, and public 27 The appointment of all civilian judges and justices in Israel is controlled by a single nine-member com mission, comprised of three Supreme Court justices, two Cabinet Ministers, two Knesset Members (usually one representing the opposition parties), and two representatives of the Bar. See Basic Law: The Judiciary, s 4(b). 28 On top of some 10,000 other cases – appeals and various motions.
130 Ori Aronson interest leaders, the bottom line so far is of very limited consequence; for the past two decades, the law of judicial review has been consistently controlled and steadily devel oped through the Supreme Court’s case law based on the existing jurisdictional state of double hybridity, with little actual impact drawn from the surrounding (often critical) discourse. Still it is worth reflecting on the character of the ongoing discourse on the institutional modalities of judicial review in Israel. They reflect prevailing understandings of the political stakes implicated by judicial review, as well as the possibilities of reform, if and when it comes. B. A Concentration Bias Most of the normative contributions to the discussion on the institutional arrangement of judicial review in Israel, as well as most of the proposals for new constitutional texts, have tended toward the concentrated end of the jurisdictional spectrum. This general tendency towards concentration of judicial review is noteworthy, because it joins in general agreement some of the harshest detractors of the Supreme Court’s current ver sion of judicial review, along with some of its greatest champions. Two main institu tional variations on the concentrated theme are offered by these groups: the first variation, proffered by many of the critics of the current hybrid form of judicial review, calls for the establishment of a separate constitutional court, in the fashion of some European tribunals, whose makeup and appointment procedures would be to varying degrees controlled by, and reflective of, the shifting realities of partisan politics.29 These proposals, not surprisingly, have originated mostly from legislators concerned with the shifting balance of power between the Knesset (Israeli Parliament) and the pro fessional judiciary in the era of judicial review. But they reflect more than mere institu tional self-aggrandisement; they seek to express in jurisdictional terms the understanding that constitutional adjudication, and above all rights-based judicial review, is a practice infused with politically-laden value judgements, in a way that is arguably qualitatively different than all other fields of legal ordering.30 The calls for distinguishing judicial review into a separate forum therefore attempt to make the political stakes of the prac tice more visible and transparent, and at the same time to insulate ‘regular’ adjudication from being ‘contaminated’ by the political. The second variation, endorsed most notably by the central figure in the introduction of judicial review to the Israeli constitutional system, former Court President Aharon Barak, calls for concentrating the power of judicial review in the hands of the Supreme Court, in essence maintaining the concentrated elements of the current system while doing away with its diffuse component. In his writings on the topic, Barak has endorsed the concentrated model due to his concern for the fledgling status of judicial review in Israel – a practice that is still in the process of substantiation and legitimation. In these circumstances, Barak posited, the power of judicial review is best restricted for now to See, eg Draft Bill Basic Law: Constitutional Court, 2009, P/4/18. cf M Tushnet, ‘Popular Constitutionalism as Political Law’ (2006) 81 Chicago-Kent Law Review 991, 993 (stressing ‘The inevitable fuzziness of the distinction between the legal and political components of constitu tional law’). See also Comella (n 7) 45–50 (exploring the ‘autonomy of constitutional discourse’ and its institu tional implications). 29 30
Diffuse Judicial Review 131
the visible and familiar purview of the Supreme Court, which enjoys the high esteem of the Israeli constituency.31 Barak, however, rejects the idea of a separate constitutional court: he has derided such calls as attempts at politicizing constitutional adjudication, which ought to remain a legal enterprise above all;32 and has sought to distinguish the Israeli case from those of Italy, Germany, or South Africa, in which the regular judiciary was tainted by the totalitarian and discriminatory commitments of previous regimes.33 Other jurists who were supportive of the 1995 ‘Constitutional Revolution’ have also argued in favour of the concentrated model of judicial review for Israel, in various formations along the constitutional court–supreme court axis. Generally, they have cautioned against the constitutional uncertainty that would result in a diffuse system from contradictory decisions of lower courts until their resolution by the Supreme Court sitting in appeal, and have warned against the wasted costs of multiple constitutional litigations that could be saved in the one-shot structure of the concentrated model. Further, they have sounded the concern that lower court judges would be over-hesitant in exercising the power of judicial review, due to the political sensitivity of such deci sions (a prediction that seems to have materialised); and, conversely, the risk of runaway constitutional activism by judges on outlier tribunals such as Israel’s multiple religious courts, who do not necessarily share the ideological commitments of the main organs of the judiciary (a warning that thus far has not proven accurate). Finally, supporters of the centralised model have stressed the political and social gravity associated with the prac tice of judicial review – characteristics that are presumably best contained by the stature and expertise of Supreme Court justices.34 Embracing these kinds of arguments, three noteworthy projects of constitutional drafting by public institutions from the past decade have also opted for the concentrated version of judicial review. Two institutions – the Knesset’s Constitution, Law, and Justice Committee, led by mostly anti-judicial review parliamentarians, and the Israel Democracy Institute, known as a mostly pro-Court think tank – have produced two separate drafts of complete new constitutions for the State of Israel. Both drafts explic itly endorse judicial review, set within a concentrated structure in which the Supreme Court is the only forum authorised to strike down laws.35 Another significant piece of constitutional legislation was proposed by the Ministry of Justice in 2012. The Ministry’s Bill Memorandum proposing the enactment of Basic Law: The Legislation36 sought to entrench the constitutional structure that has evolved in the past two decades, including an explicit embrace of the authority of judicial review – but again, only in its highly con 31 See A Barak, ‘Judicial Review of the Constitutionality of the Law: Centralized or De-Centralized’ (2005) 8 Law and Government 13, 21–22 (in Hebrew); A Barak, The Judge in a Democracy (Princeton, Princeton University Press, 2004) 333–34. 32 See, eg A Barak, ‘The Supreme Court as a Constitutional Court’ (2003) 6 Law and Government 315, 318–23 (in Hebrew). 33 See Barak, The Judge in a Democracy (n 31) 334. 34 For such arguments see, eg, Y Zamir, ‘Judicial Review of the Constitutionality of Statutes’ (1993) 1 Law and Government 395, 409–10 (in Hebrew); Z Segal, ‘The Way to Judicial Review of the Constitutionality of Laws – The Authority to Declare the Unconstitutionality of Legislation – to Whom?’ (1997) 28 Mishpatim 239, 247–56 (in Hebrew); C Klein, ‘A Constitutional Tribunal – Not as Bad as It Sounds!’ (2003) 19 Bar-Ilan Legal Studies 497, 506–13 (in Hebrew). 35 See ‘Constitution by Broad Agreement’, Proposal of the Knesset’s Constitution, Law, and Justice Committee, 2006, ch VI, s 13: www.tinyurl.com/2w8yvbs (in Hebrew); Constitution by Consensus (Jerusalem, The Israel Democracy Institute, 2007) s 163: www.tinyurl.com/na7ccp7. 36 Bill Memorandum of Basic Law: The Legislation, 2012, s 24, published by the Ministry of Justice: www. tazkirim.gov.il/Tazkirim_Attachments/41283_x_AttachFile.doc (in Hebrew).
132 Ori Aronson centrated form as the power of a nine-justice panel of the Supreme Court. The multiple authors of these three documents are not partners in a joint partisan endeavour and are actually known for their conflicting constitutional visions. Nevertheless, they all converged on the question of the jurisdictional model of judicial review, opting for the concentrated alternative. So far, none of these drafts have come to legislative fruition. The examination of the discourse reveals, therefore, a peculiar coalition between indi viduals and groups that originally hold distinctly contrasting views on the legitimate role of the judiciary in enforcing the constitution. A preference for a concentrated system of judicial review seems to have brought together those who would like to limit judicial power and tighten political supervision and control over the practice of judicial review and the identity of the officials who exercise it; those who want to further entrench and embolden the power of judicial review within the judiciary by empowering its salient institutional pinnacle; and those who seek to form a comprehensive constitutional deal that would be palatable to constituents and parties from multiple sides of the political divide. Out of competing political interests and conflicting juridical visions emerged a surprising normative consensus – a view of institutional hegemony that would concen trate the power of judicial review in the hands of a small group of High Court justices; the remaining disagreement concerns only the degree of institutional independence this centralised forum would enjoy from the rest of the judiciary and from direct political influence. This result can be seen as a fortunate coalescence of mutual interests and com patible theories, a rare glimmer of hope in Israel’s fractured constitutional landscape. In the following part, however, I will attempt to destabilise this understanding by offering some insights into the redeeming potential the diffuse model of judicial review may hold for Israel’s constitutional future, after all. V. THE DEMOCRATIC PROMISE OF DIFFUSE JUDICIAL REVIEW
A. The Democratizing Potential of Trial Court Adjudication To sum things up: Israel’s constitutional system upholds, at its core, a diffuse system of judicial review. But this underlying structure is subject to the countervailing institutional force of a hybrid reality, in which the influential High Court of Justice functions as a ‘supreme trial court’, exercising original jurisdiction over constitutional matters and incentivizing parties to petition it directly. At the same time, the diffuse logic of the sys tem is subject to an intellectual and political thrust toward the formal concentration of judicial review, in one pattern or another. The apparent result is a retraction of inferior courts from any measure of activism in the field of judicial review, and a de facto con centration of much of Israel’s constitutional attention and constitutional stakes in its Supreme Court.37
37 Indeed, even the few noteworthy jurists that have declined to join the bandwagon, embracing instead a critical stance towards the concentrated discourse, have mostly done so with a continuing focus on the consti tutional role of the Supreme Court as a High Court of Appeals. See Y Dotan, ‘Does Israel Need a Constitutional Court?’ (2000) 5 Law and Government 117 (in Hebrew); G Sapir, Constitutional Revolution in Israel: Past, Present, and Future (Tel-Aviv, Yedi’ot Aharonot, 2010) 142–47 (in Hebrew).
Diffuse Judicial Review 133
I would like to suggest that in the long run, the judicial, political, and intellectual cap ture with a Supreme Court centered version of judicial review is a cause for concern. The reason for the concern is that a concentrated perception of judicial review overlooks, and therefore neglects, the independent potential for advancing democratic values that exists in constitutional adjudication by lower courts. The shift that is called for is pri marily conceptual: once we stop thinking about trial courts as mere feeders of constitu tional questions to a high, centralised forum (either by way of appeal in the diffuse system or by way of abstention or certification in the concentrated system), and once we start investigating the unique institutional realities that characterise trial court litigation and that cannot be replicated in high court adjudication, we might come up with novel insights as to how judicial power can be employed in the furtherance of democratic values. As I have argued extensively elsewhere,38 the regularities of trial court adjudication in systems based on the common law tradition (such as Israel) exhibit certain institutional characteristics that may render it uniquely supportive of deliberative, participatory, and pluralist elements in the processes of creating constitutional meanings. Thus, trial courts present a unique structural combination of professional govern mental discretion and constant, direct interaction with real-life constituents. The combi nation of official discretion and face-to-face interaction that characterises street level bureaucracies39 in the administrative state in general is inherent to trial adjudication as well: a social forum in which state power meets individual agency on a recurrent basis, and in which legal order is formed in practice through the discursive dynamics that take place between the judge (and other state officials) and the court’s constituents, as well as the multiple unreviewable heuristics employed by the judge in managing daily court room dynamics. These institutional realities can be understood as fostering a popularly participatory potential for the processes of democratic deliberation; a potential that becomes less and less available as one climbs up the hierarchical ladder of the judiciary, where fewer cases are heard, where the discussion is governed by seasoned (elitist?) attorneys and judges, and where the main focus is with settling constitutional doctrine in general, rather than arguing about it in context. Furthermore, the trial court level is institutionally proliferated and dispersed through out the country. It comprises multiple units of adjudication, which deal with vast amounts of cases, often without the guiding hand or supervising eye of the appellate levels. This sheer numerosity element in trial court adjudication, which is coupled with a sensitivity to localised variations, can serve to generate a pluralistic assemblage of voices, visions, and values regarding possible constitutional meanings – a pluralism that a strongly centralised system of judicial review might stifle. The multiple trial courts employ diverse judges that espouse diverging theories, methodologies, and ideologies of constitutional interpretation. These judges try scores of cases, which involve diverse parties, varying narratives and claims, and invoke distinct solutions, often tailored to specific constituencies and differing circumstances. All these products of trial court liti gation should be perceived as resources for further deliberation – in and out of courts – on the content of the constitutional order. 38 See O Aronson, ‘Inferiorizing Judicial Review: Popular Constitutionalism in Trial Courts’ (2010) 43 University of Michigan Journal of Law Reform 971, 994–1005. 39 See M Lipsky, Street-Level Bureaucracy: Dilemmas of the Individual in Public Services (New York, Russell Sage Foundation, 1980) 3–12.
134 Ori Aronson The institutional idiosyncrasies of trial level dynamics seem, therefore, to hold the potential for providing a unique environment for the flourishing of several democratic values: public and communal deliberation on matters of social import; civic participa tion in the processes of designing the legal order; and pluralism in the expression and accommodation of multiple visions of individual and social welfare. This potential is, arguably, especially significant in the processes of generating constitutional meaning, because it is there that the constituency gets to engage in political self-definition and in the expression and evolution of its moral commitments. Trial courts, by virtue of their direct and recurrent interaction between state and individual and their structural localism and multiplicity, are revealed as forums for a democratic mode of creating constitutional meanings, forums that are already up and running and are popularly accessible. To the extent that constitution-making is understood as a continuous process that ought to regularly underlie the political life of a community, and that this process ought to maintain a participatory, deliberative, and pluralist character, then trial courts can provide a useful institutional role in facilitating this process. They may do so as mere collectors of cases and arguments and feeders of the variety of constitutional informa tion to a centralised constitutional court; but they are more likely to fully fulfill their potential if they take full part in the process of constitutional adjudication, including the exercise of judicial review. Trials that could lead to the striking down of legislation demand a unique awareness of the stakes involved; they also provide the constituency with unique information as to the outcomes of judicial review in diverse cases of limited scope. Ultimate resolution of the constitutional question can therefore benefit from the experience collected from lower court decisions and their aftermaths. The implication is that the diffuse model of judicial review, which relies on trial court contribution as an essential element in the interpretation and enforcement of the consti tution, holds a democratic promise that is lacking from its concentrated alternative. How should this finding play out in the unique context of Israel’s constitutional law? To that I turn next. B. A Diffuse Model for Israel Israel’s constitutional condition is noteworthy in (at least) two respects. First, although it has a set of constitutional statutes and an established practice of judicial review, Israel never experienced a significant ‘constitutional moment’ that would engage the polity in an exercise of formulating a normative vision for generations to come.40 The Basic Laws that comprise its present constitutional framework have been enacted intermittently over the course of the state’s existence through the Knesset’s regular legislation proce dures, with conspicuously little deliberative rigour, and sometimes in the furtherance of
40 The nearest such episode – the very establishment of the State of Israel in 1948 – produced a value-laden Declaration of Independence (Declaration of the Establishment of the State of Israel, 1 LSI 7 (1948)) which to this day serves as a source of moral reflection and interpretive inspiration in Israeli constitutional discourse, but has been decidedly excluded from the country’s formally binding constitutional texts. See HCJ 10/48 Zeev v Acting District Commissioner of the Urban Area of Tel Aviv (Gubernik) 1(1) PD 85, 89 [1948] (in Hebrew). English translation available at: elyon1.court.gov.il/files_eng/48/100/000/Z01/48000100.z01.htm.
Diffuse Judicial Review 135
short-term considerations of political expediency.41 However, for its constitutional order to thrive – for its moral promise as a liberal democracy to remain viable and sustainable – Israel must seek avenues through which to engage its polity with the content and mean ing of its constitution. As efforts to draft a wholly new Constitution and to form a popu larly engaging process for deliberating and ratifying it continue, we are in need of forums in which constituents could formulate, argue, and weigh constitutional ideas already now. Second, Israel is a highly divided political community. Its population is profoundly split along national, religious, ethnic, and cultural lines. It includes a large ArabPalestinian minority, a significant ultra-Orthodox Jewish community, and a growing migrant worker class, as well as a myriad other cultural groups (within both the Arab and the Jewish communities) that are divided by countries of origin, degrees of religios ity or secularity, and other cultural and ideological markers. These divisions, often relat ing to matters of general welfare, have repeatedly brought political tensions to a boil, with sometimes violent repercussions. These recurring events have exemplified the need for institutional structures that would accommodate diverse understandings of constitu tional principles, and through such pluralistic treatment enable interaction, exchange, and deliberation among groups and individuals. While, clearly, not all constitutional visions can be upheld simultaneously, accessible avenues of voice and participation can defuse at least some of the tensions, and, more importantly, generate information, ideas, and perspectives that would be instructive in the process of settling on one constitutional direction or another. Both these concerns – the lack of popular engagement in Israel’s constitutional devel opment, and the divided nature of its constituency – may find some (not complete, to be sure) resolution, through the activation of Israel’s trial courts as central institutional agents in the process of designing the constitutional order. Surely, legal discourse is not the best possible medium for engaging in mutually respectful negotiation over the moral character of a political community; courts, as Robert Cover has noted, are jurispathic institutional entities, in that they employ violence in order to suppress certain legal visions in the preference of others.42 But courts are still forums that regularly consider the meaning and validity of public values43 and in which multiple voices are heard, with structural assurances for such hearing that rarely exist in other institutions of public discourse. At the same time, the multiple trial courts accommodate significantly more voices than higher levels of adjudication. If one accepts, therefore, that vibrant constitutional litigation in lower courts offers a uniquely democratic quality to the constitutional process, then the general jurisdictional trajectory of the Israeli system should arguably be heading towards the diffuse model of judicial review, rather than away from it, as it currently seems to be going. Such a trajectory shift is, of course, a matter of degree, and requires a process of institu tional, political, and intellectual acculturation. An immediate and complete elimination of the concentrated elements of judicial review in Israel – revoking the constitutional jurisdic tion of the High Court of Justice – will likely not be possible. In a system that has been 41 Basic Law: Human Dignity and Liberty – the centrepiece of Israel’s ‘Constitutional Revolution’ – was enacted by a vote of 32 against 21 Knesset Members, out of a 120-member chamber (DK (1992) 3973); the enactment garnered practically no special attention by the press at the time. 42 RM Cover, ‘Foreword: Nomos and Narrative’ (1983) 97 Harvard Law Review 4, 40. 43 See O Fiss, The Law as it Could Be (New York, NYU Press, 2003) 50–55.
136 Ori Aronson practically overwhelmed by the centralizing force of the Supreme Court’s constitutional dominance, a sudden move to a completely diffuse model would likely leave many claim ants and many judges at a loss, resulting in a weakening of the enforcement of Israel’s fledgling constitutional rights-guarantees. More measured moves are therefore required. The Supreme Court itself could, for example, choose to decline original HCJ jurisdic tion over at least some (less urgent?) constitutional petitions, sending the claimants to the ordinary processes of litigation instead.44 The Knesset could engage in jurisdictional experimentalism of its own, delimiting the comparative purviews of supreme and lower courts over different classes of constitutional claims. And trial court judges would be encouraged (for example through promotion rewards) to take greater risks and venture more actively into the realm of constitutional interpretation, cognizant that the effects of their decisions are, after all, limited in scope and application. Whatever the mechan ism might be, the purpose is to involve, rather than deter, trial courts and their multiple participants and stakeholders in Israel’s emerging constitutional discourse. Israel lacks a formal organ or a structured procedure for constitution-making. In face of this void it has turned to its available, functioning institutions of legal and moral reflection – Knesset, Supreme Court, academia. But all these institutions share an essen tial elitism and a detachment from the real-life experiences of actual constituents and the practical realities of constitutional rights-violations. The trial courts, in which scores of individuals of all walks of life get to regularly debate the meaning of legal and moral norms, are too much of a trove of narratives, arguments, and solutions to be overlooked in Israel’s continuing process of establishing a constitutional order. The diffuse system of judicial review is, therefore, more than a mere mechanism for expressing and enforc ing a constitutional regime. It is an institutional basis for a democratic mode of creating and recreating that regime. It would be unfortunate to give it up. VI. CONCLUSION
The choice between the diffuse and the concentrated models of judicial review (and, within the concentrated family of systems, between the Supreme Court version and the Constitutional Court version) is often understood to reflect a divide between two com peting theories of constitutional law. One theory perceives constitutional law as, first and foremost, a field of law, to which regular adjudicatory method is applicable, along with its assumptions of professionalism, objectivism, and interpretivism. This theory aligns with the diffuse model of judicial review, since it leaves constitutional jurisdiction in the hands of the professional judiciary. A contrasting theory posits constitutional law as a unique field of normative ordering, which is political in its essence because it calls for recurrent value judgements concerning the moral vision of the political community. This theory fits more easily with the concentrated model of judicial review, because it expresses in jurisdictional terms the special institutional treatment awarded to constitu tional questions.45 44 As I mentioned above, the Court has done this on a few occasions, but it has intimated no principled explanation as to how it had been making its choice in such matters. See n 17. 45 The concentrated model also accommodates, by virtue of the individuation of constitutional adjudication, more room for divergent institutional design – such as including non-lawyers in the judicial panel, as well as employing openly political methods of appointment, decision-making, and reasoning. See Comella (n 7) 36–50.
Diffuse Judicial Review 137
Israel’s double-hybrid system of judicial review has, in a sense, avoided the choice between the two theories. This is not surprising: in the absence of a written constitution or even an agreed-upon process for drafting and enacting such a constitution, Israeli constitutional law has been the result of the ongoing interaction between the clearly juridical and the clearly political. It is in the metaphorical zone that lies between these two spheres of social action that Israel’s constitutional law has emerged; and the ‘supreme trial court’ embodied in the High Court of Justice’s constitutional jurisdiction is one of the institutional expressions of this interim state of affairs. The lack of a foundational constitutional theory has led, therefore, to a lack of insti tutional choice. What we have instead is a complex combination of diffuse and concen trated elements. But this is not necessarily a bad thing. The jurisdictional ambiguity of Israel’s system of judicial review has maintained and reflected (without deliberate inten tion, but nonetheless with effective force) a blurring of the legal/political distinction in constitutional adjudication. Thus, Israel has a High Court that is essentially a court of appeals, which also upholds a set of unique procedures for reviewing the variety of polit ical transgressions. It shifts between both functions fluidly, often with little consequence to either substance or procedure. It does both ‘legal law’ (appellate review) and ‘political law’ (constitutional review) continuously and contemporaneously. Indeed, the Israeli experience may serve to undermine the prevalent distinction between what ‘courts’ do and what other organs of government do. Keeping judicial review within the courts, as the diffuse model urges, will not turn the interpretation and application of the constitution into an objective endeavour, free from moral, ideo logical, or theoretical commitments or from class, culture, and other group influences. Similarly, placing the Constitution away from the regular judiciary, as the concentrated model urges, will not relieve ‘legal’ adjudication from the political stakes that are involved in any exercise of judicial power, constitutional or other. Classifying constitu tional law as more or less ‘political’, therefore, offers no real guidance as to the optimal jurisdictional arrangement for the exercise of judicial review; Israel has effectively defied this assumption. We are returned, therefore, to the fundamental questions of institutional design for a thriving constitutional democracy, in which adjudication is understood to always tread the political. Within this context I have presented an argument for fostering the diffuse option of judicial review for the future of Israeli constitutional law. While a concen trated model might be more easily attainable – as the prevailing political/intellectual consensus seems to instruct – I have tried to show that it misses the already available institutional resource for democratizing our ongoing processes of constitution-making: the recurrent, accessible, and relatively low-stake event of the trial. The diffuse system of judicial review treats this very event as the nuclear unit in the production of constitu tional meanings. Israel’s constitutional history is devoid of a ‘constitutional moment’ of serious partici patory magnitude, and its ‘Constitutional Revolution’ is consistently criticised for hav ing lacked a democratically robust deliberative dynamic. In these circumstances, a turn to the trial courts as full institutional partners in the constitutional endeavour seems plausible. Rather than retracting toward a concentrated design, where the Israeli system seems to be heading, reimagining the diffuse possibilities of judicial review may hold a key for making the yet unborn, though already vital, Israeli Constitution an actuality for Israelis to engage with, negotiate over, and thus promote.
10 Constitutional Adjudication in Israel: Some Comments VÍCTOR FERRERES COMELLA
I
T IS A great pleasure to offer some comments on the chapters written by Professors Yoav Dotan,1 Joshua Segev2 and Ori Aronson.3 The chapters are similar in spirit: they use comparative law to shed light on the basic features of the Israeli system of judicial review. Their goal is to better understand the logic of this system, in order to identify the conditions that need to be preserved for it to retain its democratic legitimacy (Dotan), to evaluate the Supreme Court’s discourse on the justification of judicial review (Segev), and to explore the advantages and disadvantages of various reforms that have been proposed (Aronson). For those of us who are not experts on Israeli constitutional law, these chapters supply the necessary information to start a fruitful conversation – and they do so very effectively. The comparative law analysis they engage in, moreover, is illuminating. They set up a useful scaffolding for an external observer like me to dare make some comments. The papers address problems that, though interrelated, are different. So I will comment on them separately. It is important to stress, however, that they all focus on the institution of judicial review of legislation. This is an important restriction. Judicial review may be applied to many different types of legal rules, issued by different institutions within the governmental structure. Actually, judicial review can also extend to private actions (such as contracts). There is a certain tendency to believe, however, that judicial review is particularly problematic when the legal rule, whose validity is at stake, is a piece of legislation that has been enacted by a popularly elected parliament. The objection against judges having the power to annul the decisions of the political branches seems to acquire special force when the decision under review is a parliamentary enactment. Statutes appear to have a special dignity because of their deeper democratic source. The various systems of judicial review that we find in the contemporary world are more or less sensitive to this background intuition. The so-called centralised model of constitutional review can be understood to be strongly tied to the idea that legislation is special, because of its democratic pedigree. Under that model, statutes can only be checked for their validity by a special tribunal that is separate and different from ordinary Chapter 7 in this volume. Chapter 8 in this volume. 3 Chapter 9 in this volume. 1 2
140 Víctor Ferreres Comella courts.4 If we allow ourselves to play with words, we might say that, within the set of legal norms that are located under the constitution, statutes form an ‘aristocracy’ in the context of the centralised model. As a result of their nobility, statutes cannot be judged by ordinary courts. Only a special tribunal can be asked to intervene. The centralised model is ‘aristocratic’ in this peculiar sense. In contrast, the decentralised model of judicial review puts all legal rules in the same basket: their conformity with the constitution is to be assessed by ordinary courts. Statutes are afforded no special treatment. The decentralised model is ‘democratic’ in the sense that it creates no privilege for statutes: all legal norms are treated the same. Maybe this has the effect of emphasizing that the really privileged norm is the constitution: no aristocracy is allowed within the ordinary legal system, we might say, in order to make the King shine with all his majesty. Even in countries that have adopted a decentralised system, however, there is a certain consensus that judicial review of legislation is particularly problematic from a democratic perspective. Maybe the institutional arrangement is not built around this intuition, but the practice of courts – and the attitudes of politicians and citizens – is usually sensitive to it. Courts are expected to be more ‘deferential’ when reviewing a statute than when reviewing an administrative regulation, for example. And the debates about the power of courts in a democracy get more passionate when the fate of legislation issued by a parliamentary assembly is at stake. So it is not surprising that our three panelists have also focused their attention on judicial review of legislation. This is the practice that raises most concerns. I. YOAV DOTAN ON COURTS AND POLITICAL ACCOUNTABILITY
Let me start with Yoav Dotan’s chapter, which explores the different ways in which judicial review of legislation can be checked by the political branches. He claims that a balance needs to be struck: on the one hand, we wish to protect human rights against ‘majoritarian tyranny’. Judicial review is set up to serve this goal. On the other hand, we do not want courts to act in an abusive manner. In order to protect ourselves against ‘judicial tyranny’, we have to establish mechanisms to constrain judges. Judicial review, of course, does not take place in a vacuum. Courts have to measure the validity of laws against the constitution. It is the norms included in that document that may be employed by judges to determine the validity of laws. Dotan believes, however, that this constraint is insufficient to allay fears about judges having too much power. The constitutional text is framed in very abstract terms that give interpreters much room for manoeuvre. This is particularly true of the Bill of Rights. So we need to look elsewhere to find ways to limit courts. Dotan asserts that accountability is increased if the political branches are involved in the process of selecting judges. The length of judicial tenure is also relevant in this regard: tenure for a limited period allows the political branches to have more opportunities to contribute their inputs. With this background in mind, Dotan proceeds to distinguish three basic models: the United States model, the continental European model, and the common law model. Each of them follows its own distinctive strategy to achieve a certain level of political 4 I have explored the rationale of the centralized model of judicial review in V Ferreres Comella, Constitutional Courts and Democratic Values: A European Perspective (New Haven, Yale University Press, 2009). Some of the points I will offer in this chapter derive from ideas I have developed in that book.
Constitutional Adjudication in Israel 141
accountability. In the United States, the appointing process is the key piece. The constitution is very hard to amend, and federal and state judges have tenure for life (or for very long periods).5 What permits the political branches to inject some doses of democratic accountability into the system is the fact that federal and state judges are appointed through procedures that are highly political in nature. The continental European model, in contrast, attacks the problem through a dualist structure. A constitutional tribunal is established with the exclusive power to review the constitutionality of legislation. While judges on ordinary courts are appointed through relatively non-political procedures, the members of the constitutional tribunal are instead chosen by the political branches. The tenure of constitutional judges is limited to a certain period of time, moreover, which helps enhance their democratic accountability. The common law model, in turn, is similar to the United States model in that judicial review is placed in the hands of regular courts. What is different, however, is that judges are selected through procedures that are basically non-political. Because of this feature, judges are only granted a limited power of constitutional review. Their power is limited in the following sense: Parliament can easily override, through legislative means, a judicial determination that a statute is incompatible with the relevant body of constitutional norms. This tripartite classification is useful, for it allows us to better understand the logic of the Israeli system. Dotan claims that Israel follows the third model, as a result of the English common law traditions that were so influential in the formation of the State of Israel. Indeed, judicial review is performed by ordinary courts – not by a separate constitutional tribunal – and judges enjoy lifetime tenure, after being appointed through a process that is non-political in nature. This being so, accountability can only be maintained if the Knesset (Israeli Parliament) retains its power to freely amend or repeal the Basic Laws that courts are expected to protect against offensive legislation. Dotan’s main conclusion, therefore, is that the Israeli system might face a serious problem of democratic deficit if the Knesset lost its current power to easily override judicial pronouncements through the repeal or amendment of the Basic Laws. I think Dotan’s argumentative strategy is attractive. It is certainly useful to distinguish three different models to answer the accountability problem, in order to see what conditions need to be preserved for the Israeli system, in particular to exhibit a sufficient measure of democratic accountability. I have three comments, however. First, there are some important variations within each model, and it is important to be aware of them when mapping out the different institutional possibilities a country may choose from. It is true, for example, as Dotan explains, that the separation between a constitutional tribunal and regular courts makes it possible, under the continental European model, for a political procedure to be followed to appoint the members of the constitutional tribunal, while a more bureaucratic or professional method is employed to pick ordinary judges. But this possibility is rejected in some countries (such as Luxembourg, for example), which have preferred to use non-political methods to select all judges, including those that serve on the Constitutional Court. Similarly, it is true that the distinction between constitutional courts and regular courts makes it possible to establish two different legal regimes with respect to judicial tenure. But some European countries have chosen to establish a unified regime: tenure until retirement age applies to 5 It is important to note, however, that in a significant number of states in the United States judges serve for rather limited periods. Life tenure or something close to it is not the prevailing rule. See S Levinson, Framed: America’s 51 Constitutions and the Crisis of Governance (Oxford, Oxford University Press, 2012) 258.
142 Víctor Ferreres Comella all judges. This is the case in Luxembourg and Belgium and, more importantly, in the country that is usually associated with the birth of the continental European model of judicial review: Austria. Second, if, under the common law model, accountability is achieved through the availability of legislative overrides, it is important to study the actual workings of the political system. As Dotan acknowledges, even if Parliament is formally authorised to modify the Basic Laws, it may be too difficult for it to do so in practice, for political reasons. We should certainly include the political factor into the equation. When we try to measure how hard it is to amend a given constitution, we cannot stop the analysis once we know what kind of parliamentary majority is required. We have to ask ourselves how tough it is to get that majority – given the number of political parties there are, for example, and how much internal discipline they exhibit. It may well happen that getting a simple majority to change the Basic Law in one country is actually more difficult than getting a super-majority in another country. Changing a Basic Law in Israel, in particular, may turn out to be harder than reforming the ‘technically more rigid’ Austrian Constitution, for example, which requires a two-thirds parliamentary super-majority for its amendment, but which can be easily reformed in practice as a result of the frequent coalition agreements reached by the two largest political parties.6 My third point relates to the set of criteria we would like a system of judicial review to satisfy. Dotan thinks, quite rightly, that we should avoid two tyrannies: the tyranny of parliamentary majorities, and the tyranny of courts. Put differently, we wish to ensure both that human rights are protected, and that the institutions in charge of such protection do not abuse their powers. But we need to be clear on two things. First: why do we believe courts are likely to improve the overall level of rights protection if we empower them to scrutinise laws for their constitutionality? What institutional theory do we rely upon, to predict that judicial review will be better than parliamentary sovereignty, when it comes to rights? Second, once we have constructed such a theory, what space should we create for the political branches to activate checks on courts? Arguably, the goal of protecting rights should be given priority over the goal of ensuring democratic accountability. If the institution of judicial review of legislation is instrumentally sound in terms of rights protection (and this is a big ‘if’, of course), then the political checks we design should not undermine the independence that courts need to enjoy in order to perform their functions in the right way. I cannot develop my own thoughts on this problem here, but it seems to me that we cannot detach our theory of ‘accountability’ from our foundational theory of judicial review. The answers to the questions why, and under what conditions, courts are well equipped to interpret and enforce human rights clauses, will tell us some interesting things about what forms of political accountability we should accept. II. JOSHUA SEGEV ON THE JUSTIFICATION OF JUDICIAL REVIEW
Yoav Dotan’s chapter fits nicely with Joshua Segev’s, in that the latter takes us to the foundational questions about the justification of judicial review and thus serves as a necessary complement to the former. 6 On the easy amendability of the Austrian Constitution, given the system of political parties, see M Stelzer, The Constitution of the Republic of Austria: A Contextual Analysis (Oxford, Hart Publishing, 2011).
Constitutional Adjudication in Israel 143
Segev’s contribution is twofold. First, it usefully characterises the different kinds of theories of judicial review one can construct. Second, it portrays the evolution of the Supreme Court of Israel’s discourse on the justification of judicial review. Segev begins by distinguishing three concepts of ‘legitimacy’: legal, social and moral. When we ask whether judicial review of legislation is a legitimate institution, we may be asking different questions. We may be interested in the legal aspect of legitimacy – whether the right interpretation of the relevant legal sources supports the proposition that courts have the authority to test legislation under the constitution. We may instead want to know whether citizens believe the institution is acceptable and worthy of respect. This would be the problem of social legitimacy. Finally, the moral question would be whether the institution is justified under a sound theory of political morality – quite apart from what the law provides, or what the public expects. Segev is right that we need to draw these distinctions in order not to be confused as to the nature of the different debates that the institution of judicial review has generated at various levels. This does not mean, of course, that there are no connections between the three dimensions of legitimacy. I do not take Segev to deny this. A legal argument in favour of judicial review may have to rely, in part, on some normative theory of judicial review. The text of the constitution, for example, may be open to rival interpretations as to the existence and scope of the power of judicial review, and the choice between them may be guided by ideas drawn from the moral debate. Also, whether judicial review is acceptable from a normative point of view may depend on certain conditions that may or may not be met in practice. If, for example, a certain level of public support is thought to be one of the conditions that need to be fulfilled for judicial review to be legitimate from a normative point of view, then we are forced to learn something about social legitimacy before we conclude that the establishment of judicial review in a given polity is ultimately justified. Segev is also insightful when he asserts that the objections that judicial review faces may be of a ‘comparative nature’ (when the objections assert that alternative arrangements are better) or of a ‘non-comparative nature’ (when the objections hold that judicial review breaches certain principles). Consequently, as he explains, theories of judicial review may be of two kinds: ‘optimal’ theories defend judicial review on the comparative ground that it is superior to alternative constitutional arrangements, while ‘permissible’ theories merely claim that judicial review passes certain thresholds (derived from moral principles), even if it is no better than alternative institutions. This classification helps us clarify the nature of the dispute. If we get into the merits of the dispute, however, it would be fruitful to investigate whether a merely ‘permissible’ theory of judicial review is ultimately tenable. Let me very briefly explain the kind of concern I have in mind. Suppose we face a choice between two institutional alternatives in a particular country: a regime of parliamentary sovereignty (with no judicial review of legislation), and a regime of constitutional democracy (with judicial review of legislation). Suppose someone prefers the first alternative, on the grounds that judicial review is not democratic.7 A ‘permissible’ theory is then advanced to respond to the objection. The theory argues that the kinds of rights courts are expected to safeguard are part and parcel of genuine democracy, and that the role of judges is actually constrained through 7 Jeremy Waldron is one of the most well-known champions of this view. See J Waldron, ‘The Core of the Case Against Judicial Review’ (2006) 115 Yale Law Journal 1346.
144 Víctor Ferreres Comella various mechanisms.8 Suppose the objector replies that courts are fallible, and that, as a matter of fact, rights will be better protected if Parliament has the last word, with no judicial review. If the ‘permissible’ theorist were to concede this latter point, but insisted that judicial review is nevertheless democratic, the opponent would not be impressed. Only an ‘optimal’ theory – one that would take judges to better protect rights than an unconstrained parliament – would seem to provide a satisfactory justification of the institution of judicial review. However democratically accountable courts may be, a popularly elected parliament seems to have stronger democratic credentials. If judicial review does not protect rights better than an unconstrained parliament, there is then a strong democratic case in favour of parliamentary sovereignty.9 This connects with another point that Segev makes, when he asserts that the justification of judicial review is ‘content-independent’: it cannot depend on the specific outcome that is reached by courts in a given controversy. I completely agree with this. The case for or against the institution of judicial review cannot be a function of the performance of parliaments and courts with regard to a particular controversy. This does not mean, however, that the justification of judicial review can be completely detached from the content of the decisions by parliaments and courts when viewed as a whole, over a relatively long period of time. There is no escaping the instrumental question, it seems to me, as to which institutional arrangement is better equipped to protect rights. The case for judicial review would be defeated by the democratic argument, if courts were not instrumentally better. Of course, it is very hard to make these overall assessments. We need to rely on an institutional theory that connects, ex ante, certain institutional features to likely outcomes, and that then tests those connections on the ground, ex post, as experience is gathered. And, obviously, we need a substantive moral theory to discriminate between ‘good’ and ‘bad’ results. But unless we have some faith in the instrumental advantages of judicial review, the democratic objection cannot be properly answered. Segev devotes the second part of his chapter to reconstruct the evolution of the Supreme Court of Israel’s discourse on the justification of constitutional review of legislation. Interestingly, he explains that the Court was first ‘pragmatic’ – it abstained from offering an explicit affirmation and justification of the power of legislative review. From 1995 onwards, the Court developed a very detailed and theoretical defence of the institution – one which Segev finds unpersuasive in many respects. Court President Aharon Barak’s opinion in the United Mizrahi Bank case10 was the crucial event here. After this ‘radical’ period, as Segev calls it, the Court has entered a ‘conservative’ phase, where it appears to be sceptical of grand theories. This is a fascinating account of the Court’s struggle with the foundations of constitutional review. Segev’s criticisms of the Court’s theory during the ‘radical’ period are interesting, though underdeveloped in the chapter. I would like to hear more, for example, about his criticism of the ‘rule of law’ justification of judicial review – which I take to be similar in nature to Marshall J’s reasoning in 8 This is, in a nutshell, Ronald Dworkin’s position in R Dworkin, Freedom’s Law: The Moral Reading of the American Constitution (Cambridge, Harvard University Press, 1997). 9 Things get more complex, however, if we take into account the democratic virtues that judicial review may exhibit (in terms of the participation and deliberation by citizens who are not powerful enough to have their interests and views taken seriously in electoral politics), quite apart from the substantive outcomes produced by courts. On the intrinsic democratic advantages of courts, see L Sager, Justice in Plainclothes: A Theory of American Constitutional Practice (New Haven, Yale University Press, 2006). 10 CA 6821/93 United Mizrahi Bank Ltd v Migdal Cooperative Village 149(4) PD 221 [1995] (in Hebrew).
Constitutional Adjudication in Israel 145
Marbury v Madison.11 The idea that, if you have a constitution that is part of the law, and if that constitution is superior to ordinary laws, then judges have the power of constitutional review when deciding cases, unless otherwise explicitly provided in the constitution, is a pretty plausible legal argument. Whether such conditions obtain in Israel is part of the controversy, as I understand it. Maybe Barak did not have the necessary legal pieces on the table – whereas Marshall did have them – to construct a plausible argument in favour of judicial review along those lines. III. ORI ARONSON ON ISRAELI HYBRIDITY
The third paper, by Ori Aronson, focuses on the particular way judicial review of legislation has been organised in Israel. As is true in other countries (in Latin America, for example),12 Israel combines features of both the centralised (‘concentrated’) and the decentralised (‘diffuse’) models of constitutional review of legislation. There have been proposals to introduce a more centralised system in Israel, but Aronson claims, quite interestingly, that important things would be lost if such proposals were implemented. The Israeli system is ‘hybrid’, Aronson explains, in two senses. First, the Supreme Court acts in two kinds of jurisdictional capacities. As a ‘High Court of Appeals’, it is the tribunal of last resort for cases litigated within the civilian adjudicatory system. This allows the Court to lay down precedents that fix the meaning of civil, criminal and administrative law for all judges to follow. In addition, as a ‘High Court of Justice’, it has original jurisdiction for writs of mandamus brought against the state, as well as particular governmental organs and officeholders. In this capacity, the Court has considered petitions by both private and public actors, challenging various types of governmental actions, both in the abstract and in the context of specific violations. The Court looks very much like a constitutional tribunal when it rules upon such challenges. But it is not really such a tribunal, of course, to the extent that it is also in charge of interpreting ordinary law – as an ordinary Supreme Court. The second sense in which the Israeli system is hybrid is that lower courts are formally empowered to check legislation for its constitutionality, but judges are reluctant in practice to exercise this authority. To a large extent, this reluctance springs from the criticisms that were visited on the few judges who decided to set aside statutes on constitutional grounds in some cases. So, in practice, most of the legislative checking is done by the Supreme Court. Various proposals have been advanced, Aronson explains, to reform this hybrid system. Most of them have exhibited a preference for the concentrated model. Some voices have suggested the establishment of a separate Constitutional Court, following the continental European model. Others prefer a system where the power of legislative review is concentrated in the Supreme Court. Aharon Barak, for example, favours this latter version of the centralised model, in order to better protect judicial independence and reduce the risk of politicization. Barak celebrates the fact that, in Israel, judges that serve on the Supreme Court are appointed in a non-political way. If the members of a new constitutional tribunal were to be selected by the political branches, as is the case in many Marbury v Madison 5 US 137 (1803). See, JO Frosini and L Pegoraro, ‘Constitutional Courts in Latin America: A Testing Ground for New Parameters of Classification?’ (2008) 3 Journal of Comparative Law 39, 39–63. 11 12
146 Víctor Ferreres Comella European countries, adjudication would be too politicised. This worry is to be taken seriously, it seems to me. The Spanish experience, for example, suggests that there is indeed a risk of politicization of the Constitutional Court if political parties are allowed to be the key players throughout the appointment process. The risk is especially serious in Spain as a result of the fact that the judges of the Constitutional Court do not serve for life or until retirement age, but for short periods of time (nine years). Most constitutional judges cannot help thinking what opportunities will be open to them once they step down from the Court – depending, of course, on how they have ‘behaved’. Ori Aronson claims, quite importantly, that the centralizing proposals that have been articulated in Israel neglect a significant strength of the diffuse model: constitutional adjudication by lower courts has an enormous potential in terms of advancing democratic values. He gives us several reasons to believe that legislative review at the trial level better serves democratic deliberation and participation. Those reasons can be summarised as follows. First, trial courts are not worried, as the highest courts are, about fixing general doctrines: they reason on the law in the context of cases. Second, judges in lower courts are less seasoned – less elitist. There is also more ideological diversity among them. Third, trial courts decide many more cases than the highest courts. This exposes them to a wider variety of situations, and makes them sensitive to the rich plurality of opinions that emerge in a highly divided Israeli society. These are extremely interesting points. I would like to obtain more details about the judicial system in Israel, to assess how many democratic benefits can be reaped if trial courts are encouraged to exercise the power of legislative review. Just to play the devil’s advocate (in this case, the advocate of the continental European model of judicial review), let me suggest a few points that may be made in favour of some type of centralization. First of all, it is certainly important for citizens to have access to courts, to allow them to advance legal claims in the context of particular controversies. There is a deep democratic value to the fact that the doors of trial courts are open to citizens. Because the judicial system has the form of a pyramid, letting lower courts intervene in constitutional matters means that larger numbers of constitutional cases can get litigated. When a statute is challenged on constitutional grounds, however, it may be reasonable to send the issue to a special forum – to a body whose members can devote their full time and energy to the task of legislative review. Trial courts may be overwhelmed with too many ordinary cases. Centralizing legislative review in a constitutional tribunal, or in the Supreme Court, may be better. Note that lower judges can still perform a crucial function within a centralised system: they can trigger the jurisdiction of the central court through a ‘constitutional question’ mechanism. When they do so, they can offer their reasons why they think the statute that is applicable to the case is at odds with (or may be at odds with) the constitution. But the final decision is made by a tribunal that has more time to think deeply about the problem that the statute poses. In this regard, of course, a Constitutional Court that does not have to worry about laying down precedents on ordinary legal matters may be in a better position than a Supreme Court that has a dual function to perform. Aronson worries, quite rightly, about ideological diversity. But it is not so difficult, it seems to me, to construct a system that ensures that the members of the Constitutional Court – or of the Supreme Court – mirror the diversity of beliefs that emerges within the judiciary as a whole. Actually, a constitutional tribunal may have an advantage in this
Constitutional Adjudication in Israel 147
regard: the viewpoints and professional experiences of its members may be more varied than those of regular judges. As far as seniority is concerned, I do not think we should be worried either, if we opt for a centralised system. In general, we should prefer statutes to be reviewed by senior jurists, rather than by less experienced ones. One of the reasons for the creation of constitutional courts in Europe has to do with the extreme youth of trial court judges. Legislative review is too grave a matter to be placed in such inexperienced hands. Of course, we need to be concerned if the senior judges at the very top of the hierarchy are very old. If the judges on the Supreme Court are too old, we have a good reason not to centralise legislative review there. But, in such circumstances, the diffuse system that Aronson favours does not fare much better. For, even if the trial court judges in charge of constitutional review are quite young, the doctrines they are required to apply when engaging in that task are those that are generated by the Supreme Court. Under the diffuse model, it is ultimately the judges on the Supreme Court that drive the machinery of constitutional adjudication. A constitutional tribunal, in contrast, has a potential advantage in this regard. A better balance can be achieved, by way of the separation between the Supreme Court and the constitutional tribunal. While the judges on the Supreme Court are usually at the end of their judicial careers, the members of the constitutional tribunal tend to be younger. It is not necessary for them to be picked from the top of the judiciary. Nor is it necessary for them to be judges: they may be professors, lawyers or governmental officials, all of whom need not be at the very last stages of their respective careers. The members of the Constitutional Court are thus ‘senior’ when compared to trial court judges, but rather ‘junior’ when compared with the judges on the Supreme Court. So when it comes to striking the right balance with respect to the age of judges, there is something to be said in favour of constitutional tribunals. I offer these remarks, not as criticisms of Aronson’s interesting proposal in favour of a diffuse system in Israel, but merely as a counterpoint, to emphasise the potential strengths of the centralization of legislative review in a special Constitutional Court. It goes without saying that every system has its own problems. In the case of the ‘European model’, one of the most serious problems concerns the potential tensions between the constitutional tribunal and the regular Supreme Court. Before Israel decides to erect such a special body, it is important to make sure that a smooth relationship will develop at the top of the judicial pyramid. Unfortunately, there is no easy recipe to guarantee institutional harmony at that level.
11 The Use of Foreign Law in Israeli Constitutional Adjudication IDDO PORAT*
I. INTRODUCTION
R
ECENT YEARS HAVE seen a fierce debate in the United States over the use of foreign law in American constitutional law decisions.1 In Israel, however, the use of foreign law in constitutional decisions is a longstanding practice which seems to raise much less concern among judges and academics.2 Nevertheless, such practice has its flaws and deserves consideration. This chapter will attempt to explain the reasons for the prevalence of the use of foreign law in Israeli constitutional decisions, canvas the different parameters that shape it, and assess several objections that have been raised against it. The conclusion would be that Israeli constitutional culture is traditionally receptive to the use of foreign law, and there are good reasons for it to keep using it. However, regarded as part of a global trend in which constitutional courts compete over leadership and innovation in rights protection, and taking into account Israel’s shaky ground for its ‘Constitutional Revolution’, the use of foreign law may raise some valid concerns. The chapter will begin in Part II by presenting evidence for the extensive use of foreign law in Israeli law. It will then put forward seven explanations for this practice in Part III. The first four pertain to the nature and history of Israel’s constitutional law. They are: (1) non-textualism and the fact that Israel has no written constitution; (2) the effects of * I would like to thank Moshe Cohen-Eliya, Binyamin Blum and Michael Ramsey for their helpful comments and suggestions. 1 See, eg M Tushnet, ‘When is Knowing Less Better than Knowing More? Unpacking the Controversy Over Supreme Court Reference to Non-US Law’ (2006) 90 Minnesota Law Review 1275; VC Jackson, ‘Constitutional Comparisons: Convergence, Resistance, Engagement’ (2005) 119 Harvard Law Review 109; R Posner, ‘No Thanks, We Already Have Our Own Laws’ (July–August 2004) Legal Affairs 40; J Waldron, ‘Foreign Law and the Modern Ius Gentium’ (2005) 119 Harvard Law Review 129; RP Alford, ‘Four Mistakes in the Debate on “Outsourcing Authority”’ (2006) 69 Alberta Law Review 653; O Benvenuto, ‘Re-evaluating the Debate Surrounding the Supreme Court Use of Foreign Precedent’ 74 Fordham Law Review 2695; D Farber, ‘The Supreme Court, the Law of Nations and Citations of Foreign Law: The Lessons of History’ (2007) 95 Commonwealth Law Reports 1335. 2 For a recent exception, see B Blum, ‘Doctrines Without Borders: The “New” Israeli Exclusionary Rule and the Dangers of Legal Transplantation’ (2008) 60 Stanford Law Review 2131 (criticizing the use of foreign law by the Israeli Supreme Court in the context of evidence law and defendants’ rights). Other examples include H Sandberg, ‘Cultural Colonialism – the Americanization of Legal Education in Israel’ (2009) 27 Hamishpat 52 (in Hebrew) (criticizing the extensive reliance on American law in Israeli legal education).
152 Iddo Porat the Israeli ‘Constitutional Revolution’ on the use of foreign law; (3) the recent adoption of a European-based mode of constitutional adjudication, and (4) the anti-formalism of Israeli constitutional law. The other three explanations pertain to general characteristics of Israeli law, which also affect the use of foreign law in constitutional law. They are: (1) the fact that Israel is a young legal system; (2) Israel’s geopolitical isolation and fas cination with the West, and (3) the strong professional and academic ties that Israeli lawyers have with other legal systems. As regards the appropriateness of the use of foreign law in Israeli decision-making, Part IV of the chapter will document several general objections to the use of foreign law in constitutional law, and assess their applicability to the Israeli system. It will start with a substantive objection based on local sovereignty and on original intent, and will move on to discuss methodological objections according to which citing foreign law is undisciplined, and amounts to cherry picking, and comparison is often inaccurate and distorted due to the multiplicity of factors that need to be integrated into the comparison. While these objections are not without merit, they do not present strong enough reasons to object to the use of foreign law. The final objection may raise more concerns, and it is based on the pitfalls of global constitutionalism. II. DOCUMENTING THE PREVALENCE OF THE USES OF FOREIGN LAW IN ISRAEL
Two preliminary remarks are in order. First, throughout this chapter, and unless otherwise noted, I will use the term ‘foreign constitutional law’ to indicate both internal constitutional law of other countries and international law. Although, as Vicky Jackson has shown, there are important differences between foreign constitutional law and inter national law,3 both represent foreign legal materials pertaining to human rights and constitutional rights, and may reasonably be coupled together for the purpose of this study. Moreover, the border lines between constitutional law and international law have become blurred over time – a process termed as the ‘internationalization of constitutional law’, and the ‘constitutionalization of international law’,4 and this is another reason for coupling the two together. I should note however, that I do not include in my review international law that is binding on the State, but only the use of international law as suggestive authority. My second remark is that whenever I speak of the use of foreign law in Israel, this always applies to the jurisprudence of the Israeli Supreme Court, because Israeli constitutional jurisprudence takes place almost exclusively within the bounds of Israeli Supreme Court adjudication.5
3 See, eg V Jackson, ‘Constitutional Law and Transnational Comparisons: The Youngstown Decision and American Exceptionalism’ (2006) 30 Harvard Journal of Law and Public Policy 191, 204–14 (documenting five differences between citing foreign law and citing international law). 4 See J Klabbers, A Peters and G Ulfstein, The Constitutionalization of International Law (Oxford, Oxford University Press, 2009) and compare M Tushnet, ‘The Inevitable Globalization of Constitutional Law’ (2009) 49 Virginia Journal of International Law 985. 5 The Israeli Supreme Court, sitting as High Court of Justice, used to handle all petitions against the State as a first and last instance. Since the enactment of the Administrative Affairs Courts Law, 5760-2000, some petitions are brought to the District Courts, sitting as Administrative Courts and may be appealed to the Supreme Court.
The Use of Foreign Law 153
Taking these remarks into consideration, the first thing to note is that the use of foreign law in Israeli constitutional law is extensive. Several indications attest to it, beginning with a study conducted in the mid-1990s which documented citation practices in the jurisprudence of the Israeli Supreme Court.6 This study applies to all cases and not only to constitutional ones, but it is the only study that is comprehensive for that period, and it is reasonable to assume that some of the general trends apply to constitutional cases as well. The first finding of the study is that an average of 21 per cent of the total number of citations in all7 Supreme Court decisions (including citations of cases, statutes, and academic or literary sources) published between the formation of Israel in 1948 and 1994 were foreign citations (I shall refer to this as the ‘citation ratio’).8 Generally speaking, in the earlier years of Israeli law there were more foreign citations, as there were fewer local sources to draw on, and as there still existed formal and cultural ties with English law as a result of the British Mandate over Palestine before the creation of Israel.9 The bulk of foreign citations from the early years of Israeli law were therefore from English law.10 As the ties with British law weakened, and the formal tie abolished, and as there took place a natural process of building up a local reservoir of precedents, the citation ratio levelled down to around 10 per cent towards the last decade of the study.11 The study has further found that this percentage applied equally to citations from academic or literary sources and citations from law cases: 10 per cent of all law review and book citations come from foreign sources, and 10 per cent of all case citations come from foreign law sources. Although considerably smaller than in the early years this is still a substantial citation ratio taking into consideration comparative data to be discussed below.12 The figure of 10 per cent may be somewhat misleading, however, for two reasons. First, it does not reflect the increase in terms of the number of foreign law citations per case (the ‘per case number’). One should note that in the later years of the study the average per case number of total citations (both local and foreign) rose dramatically, in accordance with the rise in the average length of Supreme Court cases. In 1955, for example, the per case number of all citations (both foreign and local) was 5, whereas in 1993 it rose to 15.7.13 This means that in 1993, the per case number of foreign citations was 1.8, which means that, on average, every Supreme Court decision had 1.8 foreign citations in it. In addition, both the 10 per cent citation ratio and the 1.8 average per case number are misleading since they refer to all published Supreme Court cases some of which carry 6 See M Gross, R Haris and Y Schachar, ‘References Patterns of the Supreme Court in Israel-Quantitative Analysis’ (1996) 27 Mishpatim 119 (in Hebrew). 7 The study built a data base comprised of 7,146 court decisions which is 40%, and a representative sample, of the total 18,000 cases that were published in the official PD publication of Supreme Court decisions during these years. See ibid 126. 8 See ibid 152 (last row, column 3, of the table there). 9 Formal ties to English law existed in s 46 of the King’s Order in Council 1922, which remained in effect until 1980, and determined that in cases in which analogy, statute or case law provide no clear rule of decision, the courts shall decide based upon ‘the substance of the common law, and the doctrines of equity in force in England’. The provision was later replaced by the Foundations of Law Law, 5740-1980, which in s 2 replaced the common law with ‘the principles of freedom, justice, equity and peace of Israel’s heritage’. 10 At the highest point, in 1952, English law comprised of 37% of all citations – foreign and local. Gross, Harris and Schachar (n 6) (row 5 in the table there). 11 See ibid (average of the years 1984–94, in the table there). 12 See ibid 141. 13 See ibid 140.
154 Iddo Porat little precedential weight.14 If we turn our gaze to the leading and precedential cases, the results in terms of the use of foreign law are much more dramatic. A study that looked for citations practices in the 100 most cited cases in Israeli law between 1948 and 2000 found that the per case number on that list was 7.8, that is 8.7 times more than in the regular cases.15 That study does not calculate the citation ratio in these cases, but it does indicate that it is higher than in the regular cases.16 A third type of indicator for the use of foreign law is the percentage of cases citing any foreign law out of the total number of cases in a given year (the ‘cases ratio’). Two rudimentary studies, that I conducted, indicated a very high cases ratio in Israeli constitutional law. I conducted the first on the 100 most cited cases and it showed that the cases ratio was 75 per cent: 75 out of the 100 most cited cases cited foreign law. Among the 50 cases of the 100 most cited cases that were constitutional cases, the ratio was 100 per cent.17 In another rudimentary study that I conducted I looked for citations of foreign law in the cases appearing in my syllabus for the first year course of constitutional law, as a very rough estimate of leading constitutional cases. The results here were also very distinct. The cases ratio was 61 per cent: out of the 77 cases on my syllabus, 47 cited foreign law.18 These findings were supported by a more extensive study, still in its early stages, which for the first time checked cases ratios in all Israeli Supreme Court constitutional decisions. In a study conducted by Professor Suzie Navot it was found that between the years 1985 and 1994 the foreign law cases ratio in all Israeli Supreme Court constitutional decisions was 31 per cent.19 The same study also separated between ‘institutional’ constitutional decisions and ‘human rights’ constitutional decisions and found out that 70 per cent of the cases citing foreign law were human rights cases and only 30 per cent were institutional cases. Combining these facts with the fact that only 40 per cent of the constitutional cases were human rights cases, we arrive at the conclusion that within the human rights cases, the cases ratio is a high 53 per cent.20 A fourth type of indicator attests to the breakdown of foreign law sources according to different countries. The 1994 study, which checked all types of cases, and not only constitutional ones, found that in the last 12 years of the study (1982–94) the two main sources of foreign law citations were English and Commonwealth cases and American cases comprising each of almost 50 per cent of all foreign law citations in all Supreme Court cases. Continental and international sources were very few.21 In the Navot study 14 The published cases are comprised of the more important ones, however, and even among those many are of limited precedential weight, as that number amounts to several hundred cases a year. 15 C Goldschmit, M Gross and Y Shachar, ‘100 Leading Precedents of the Supreme Court – A Quantitative Analysis’ (2004) 7 Law and Government 243, 267 (in Hebrew). 16 The indication for that comes from the fact that citations of Israeli cases in the 100 most cited cases increases only 6.6 times compared to regular cases, while citations of foreign law increases 8.7 times compared to regular cases, ibid 267. 17 Study results available with author. 18 Study results available with author. I thank Avichai Shalom, for excellent research assistance in producing this study and the study in the previous footnote. 19 S Navot, ‘The Use of Foreign Precedents by Constitutional Judges Israeli Report (draft)’ table 13, p 7 (on file with author). 20 ibid 4.The calculation is as follows: 40% of all constitutional cases are human rights cases; 31% of all constitutional cases cite foreign law and within those 70% are human rights cases – so 31%*0.7=21.2% of all constitutional cases are human rights cases that cite foreign law. So we divide 40% by 21.2% and we get 53%. 21 But see E Salzberger and F Oz-Salzberger, ‘The German Heritage of the Israeli Supreme Court’ (1998) 21 Tel-Aviv University Law Review 259 (in Hebrew) (arguing that the influence of continental law, and especially of German law, on Israeli law is profound but hidden, due to anti-German sentiments after the Holocaust).
The Use of Foreign Law 155
which surveyed only constitutional law between the years 1985 and 1994 the results were very different. Her study found that 65 per cent of all foreign law citations in constitutional cases (both institutional and human rights) were from American cases, 16 per cent from Canadian, 8 per cent from English and 7 per cent from German cases.22 Data regarding the use of foreign law in more recent years is not available yet, but an example from a recent leading case could give us some indication. Adalah v Minister of Interior23 involved the controversial issue of whether residents of the Palestinian Authority who marry Israeli citizens may move to Israel and become residents and eventually citizens of Israel. Indeed, such a case lends itself to comparative perspectives regarding immigration and residency policies, however the extent of foreign law citations in that case is very impressive. The citation ratio of foreign law in this case is 24 per cent: out of a total of 243 cases cited in the Court’s decision 60 are foreign law cases. The breakdown between different countries is also interesting and seems to correspond to the Navot study finding regarding the lower share of English citations. The wide range of countries cited is also remarkable: 30 cases from the United States, 9 from Canada, 8 from the European Court of Human Rights (ECtHR), 6 from England, 3 from Germany, 3 from South Africa, 1 from Ireland and 1 from Australia. How extensive is the use of foreign law in Israel in comparison to other countries? Comparative perspectives are tricky, but it seems that Israel can be situated among the countries that tend to cite foreign law extensively. The US Supreme Court is obviously on the lower side of the scale in terms of foreign law citations. It cites much less than the Israeli Supreme Court, or any other court in the Western world for that matter.24 In Australia, a study that checked for foreign citations in State Supreme Courts, found that, setting English citations aside (Australia being subject to the English Privy Council, and hence tied formally to English law), 3.65 per cent of the total number of citations in 1995 came from foreign sources, and in 2005 – 2.14 per cent.25 By comparison, the use of foreign sources in Israeli constitutional law is indeed extensive. Frequent use of foreign law can also be found in several other countries. Thus, in the first decade of the Canadian Supreme Court (after the Canadian Charter in 1984) the citation ratio of foreign and international citations amounted to nearly 10 per cent of the total number of citations – the same as the citation ratio in all (not only constitutional) Israeli Supreme Court cases in roughly the same years.26 The cases ratio (ratio of cases citing any foreign law) reached a high of 32 per cent in 1995,27 which is roughly the same as the average in Israeli constitutional jurisprudence between 1984 and 1995. The per case ratio reached a high of 1.61 in the Canadian Supreme Court in 1990 (compared
Navot (n 19) 8. HCJ 7052/03 Adalah v Minister of Interior 61(2) PD 202 [2006] (in Hebrew). 24 As an illustration of this point, when Steven Calabresi and Stephanie Zimdahl reviewed the major US Supreme Court decisions that used foreign law between the years 1940 and 2005, they mentioned only 14 cases. See SG Calabresi and SD Zimdahl, ‘The Supreme Court and Foreign Sources of Law: Two Hundred Years of Practice and the Juvenile Death Penalty Decision’ (2005) 47 William and Mary Law Review 743, 838–83. 25 R Smyth, ‘Citations of Foreign Decisions in Australian State Supreme Courts Over the Course of the Twentieth Century: An Empirical Analysis’ (SelectedWorks, 2008) 29: www.works.bepress.com/russell_smyth/1. 26 P McCormick, ‘The Supreme Court of Canada and American Citations 1945–94: A Statistical Overview’ (1997) 8 Supreme Court Law Review 527, 533. 27 DM London, ‘Cites Unseen: The Canadian Supreme Court and Comparative Constitutionalism’ (Philadelphia, Annual Conference of the American Political Science Association, 31 August 2006) 17, available at: www.allacademic.com/meta/p152935_index.html. 22 23
156 Iddo Porat to 1.8 in all Israeli Supreme Court cases between 1982–94).28 However, studies surveying later years of Canadian jurisprudence show a dramatic decline in the use of foreign law suggesting that the extensive use of foreign law was due to the fact that these were formative years in which the Court lacked any jurisprudence of its own. As for the citation ratio in Canada it dropped from 10 per cent down to 6 per cent in 2000, and to 3 per cent in 2005.29 The cases ratio declined from 32 per cent in 1995 to 14 per cent in 2005,30 and the per case number reached a level of just 0.34 foreign citations per case in 2005.31 In South Africa numbers are even higher, owing partly to the fact that the South African Constitution requires reference to foreign and international law. Between the years 1995 and 2009 the cases ratio was a high 50 per cent. Israel can therefore be situated in a category somewhat similar to that of the Canadian early Charter jurisprudence but well above later Canadian jurisprudence and somewhat below South African constitutional jurisprudence. The purpose of the following parts will be to explain the reasons for this phenomenon and describe the factors that shape it. I will begin with a discussion of the special features of Israeli constitutional law, and then proceed with an analysis of several general factors that may provide an explanation for this extensive use of foreign law.
III. EXPLAINING THE PREVALENCE OF THE USE OF FOREIGN LAW IN ISRAELI CONSTITUTIONAL JURISPRUDENCE
A. Israeli Constitutional Law and the Use of Foreign Law Four major features of Israeli constitutional law shape the use of foreign law in Israeli jurisprudence and can explain its pervasiveness: (1) its non-textual nature; (2) the ‘Constitutional Revolution’ of 1992; (3) the adoption of a European based and universalistic model of constitutional adjudication, and (4) its non-formalistic nature, especially since the 1980s. i. Non-Textualism before 1992 An historical overview is necessary in order to understand Israeli constitutional law in general and its non-textual nature in particular.32 The history of Israel’s constitutional law is usually divided into two stages: from the establishment of the State in 1948 until 1992, and from 1992 onwards. As will be shown, both stages are marked by non-textualism. Israel was established in 1948 after a period of 30 years of British colonial rule over Mandated Palestine. Although in its defining document – the Declaration of
ibid 16. ibid 18. 30 ibid 17. 31 ibid 16. 32 For a general review of the constitutional history of early Israel, sympathetic to the review here, see R Gavison, ‘The Constitutional Revolution, Description of Reality or a Self-Fulfilling Prophecy?’ (1997) 28 Mishpatim 21, 75–95 (in Hebrew). See also G Sapir, Constitutional Revolution in Israel (Tel-Aviv, Ybook, 2010) 31–41 (in Hebrew). 28 29
The Use of Foreign Law 157
Independence33 – Israel’s early leadership vowed to adopt a Constitution, this commitment was made without any public debate since the UN Resolution that provided the legitimacy to the State of Israel, demanded a Constitution as a condition for gaining international recognition of the State.34 However, once the international community recognised the State of Israel even without a formal Constitution, the need for a Constitution was again questioned, and the Israeli Parliament (the Knesset) decided, in a decision called the ‘Harari Decision’, not to adopt a Constitution.35 Instead, the form of government that was adopted was a Westminster-like form of Parliament –sovereignty without a formal Constitution. However, the Harari Decision not to adopt a Constitution did not close the door on a Constitution but took the form of a compromise: the Constitution would be drafted in stages, chapter by chapter, each chapter to be called a Basic Law. These Basic Laws, promulgated by the Knesset in ordinary acts of legislation, would then be compiled into a formal Constitution. This compromise decision, devised to avoid a Constitution at that stage but also to remain faithful to the internal and international commitment to have one, intentionally left vague both the status of these Basic Laws during the interim period, and the deadline by which they should be compiled into a Constitution. The political and legal community consequently interpreted it as a decision to defer the question of a Constitution to a later stage. The 9 Basic Laws that were enacted intermittently until 1992 (including Basic Law: The Knesset, Basic Law: The Government and Basic Law: The Military) were therefore treated as regular laws, albeit important in view of the fact that they included some of the basic rules of government. No Basic Laws containing a Bill of Rights were enacted. To conclude, the entire first stage of Israeli constitutional law, from its formation in 1948 until 1992, was completely devoid of any textual components, and Israel was categorised as a system without a formal Constitution. The Court did develop an impressive set of civil rights, including the rights to freedom of speech, equality, freedom of consciousness, freedom of religion and from religion, freedom of occupation, liberty from arrest, due process of law and the right to a fair trial, but all these rights were made judicially, without a textual basis, and also could not be used to defy a formal manifestation of legislative will (ie no judicial review over primary legislation).36 The most salient feature of Israeli constitutional law is therefore its non-textual nature and the fact that Israel has no formal Constitution.37 This non-textual nature substantially facilitates the use of foreign law in Israeli constitutional law in two major respects. First, it sets aside one of the main objections to the use of foreign law in constitutional interpretation – loyalty to the text and original intent. As will be discussed more extensively in Part 33 Declaration of the Establishment of the State of Israel, 1 LSI 7 (1948). The Declaration refers to ‘the establishment of the elected, regular authorities of the State in accordance with the Constitution, which shall be adopted by the Elected Constituent Assembly not later than the 1 October 1948’, ibid. 34 GA Res 181 (II) (A+B), UN Doc A/RES/181(II) (29 November 1947). 35 The Resolution states: ‘The first Knesset directs the Constitutional, Legislative and Judicial Committee to prepare a draft Constitution for the State. The Constitution shall be composed of separate chapters so that each chapter will constitute a Basic Law by itself. Each chapter will be submitted to the Knesset as the Committee completes its work, and all the chapters together shall be the State’s constitution’ DK 5 (1950) 1743 (in Hebrew). 36 See generally, D Kretzmer, ‘Fifty Years of Supreme Court Jurisprudence in Human Rights’ (1999) 5 Law and Government 297, 298–300 (in Hebrew). 37 As described in Part IIIAii, the non-textualism continues to be a salient feature of Israeli constitutional law also after 1992.
158 Iddo Porat III, constitutional interpretation methods based on textualism or on original intent are antithetical to some uses of foreign law, since the way other nations interpret and apply their own Constitutions is generally considered irrelevant to the question of the textual content of one’s own Constitution or of the original intent of its drafters. Therefore, a constitutional jurisprudence which is not based on text, but only on case law, is not prone to this particular objection. Indeed, as Jamal Greene has well shown, even constitutional systems with a formal Constitution do not usually adopt an originalist interpretation, unless, as in the American case, their Constitution has achieved a canonical or a sacred text status, and unless the drafters of the Constitution have achieved the status of cultural giants and bigger-than-life figures.38 This is all the more so when it comes to systems that have no constitutional text whatsoever, as in the Israeli case before 1992. This is not to say that non-textualism necessarily means extensive use of comparative law, just as textualism does not necessarily deny the use of foreign law. It only means that it makes such use easier by lifting the obstacle of the text. Secondly, the entire project of building up a corpus of constitutional rights from scratch, as it were, by judicial means, lends itself to comparative use and to drawing on the experience of other countries. Foreign law, especially when originating in countries with a developed and longstanding jurisprudence of constitutional rights, is a natural source of reference for such a project. Accordingly, some of the early decisions that built the impressive corpus of judge-made constitutional rights used foreign law materials extensively. A striking example is the leading and breakthrough case of Kol Ha’am.39 The Kol Ha’am case of 1953, which for the first time recognised the right to free speech in Israeli law, without any textual anchoring, was replete with foreign law citations. The decision struck down an administrative decision by the Minister of Interior to temporarily close down two newspapers because of ‘seditious’ op-eds run by their editors. Written by Agranat J – himself legally trained in the United States – the decision included citations from such classic free speech American cases as Abrams v United States (1919)40 (citing the famous paragraph regarding the ‘free market of opinions’), Whitney v California (1926)41 (citing Brandies J on the quality of free speech ‘to make men free to develop their faculties’), Schenck v United States (1918)42 (citing Holmes J on the special conditions of free speech in times of war) Gitlow v New York (1924),43 as well as from cases and books that were more recent to the case, such as Cantwell v Connecticut (1939),44 United States v Associated Press (1943),45 Zecharia Chaffee’s Free Speech in the United States (1948),46 and Dennis v United States (1951)47 – a case on which Agranat J based the particular balancing test that has become the standard free speech test in Israeli law ever since. The case also included citations from Milton, Blackstone, Mill, Lord Scrutton and Lord Sumner. All in all the case included citation of nine American cases, eight English cases, and only three Israeli cases. J Greene, ‘On the Origins of Originalism’ (2009) 88 Texas Law Review 1, 63–66. HCJ 73/53 Kol Ha’am v Minister of Interior 7(2) PD 871 [1953] (in Hebrew). 40 Abrams v US 40 250 US 616 (1919). 41 Whitney v California 274 US 357 (1926). 42 Schenck v US 249 US 47 (1918). 43 Gitlow v NY 268 US 652 (1924). 44 Cantwell v Connecticut 310 US 296 (1939). 45 United States v Associated Press, 52 F supp 362, 372 (1943) (Hand J). 46 Z Chaffee, Free Speech in the United States (Cambridge, Harvard University Press, 1948). 47 Dennis v US 341 US 494 (1951). 38 39
The Use of Foreign Law 159
Many later leading cases that established newly created rights or shaped and enlarged their contours followed suit. This is especially true with regard to precedential cases involving doctrines of free speech, such as a right of free speech against film censorship,48 the right to protest49 free speech versus racist incitement,50 censorship of pornography,51 but also with regard to other rights such as equal rights for women52 and gay rights.53 ii. Non-Textualism after 1992 and the ‘Constitutional Revolution’ The second stage of Israeli constitutional law begins in 1992. At that time two new Basic Laws were adopted by the Knesset (Israeli Parliament) – Basic Law: Freedom of Occupation54 (protecting the freedom of occupation), and Basic Law: Human Dignity and Liberty55 (protecting the rights to life, bodily integrity, dignity, property, freedom from arrest and extradition, the right to move in and out of the country, and the right to privacy). These two Basic Laws included, for the first time, civil rights protections that could be seen as the first stage of a full Bill of Rights. The Court interpreted the 1992 Basic Laws very broadly, and regarded them, in the leading 1995 United Mizrahi Bank56 case as a ‘Constitutional Revolution’. First, in United Mizrahi Bank, it considered them as conferring on the Court the authority to impose judicial review over primary legislation, despite the fact that there was no direct legal provision to that effect. Secondly, in a series of later cases it read into Basic Law: Human Dignity and Liberty almost an entire Bill of Rights, including rights which were not written into it, such as the right to equality, freedom of conscience, freedom of religion, the right to a family, the right to have access to the courts, the right to freedom of speech, the right for minimal living conditions, the right to education, and what may appear to be a right of a new type – a right against privatization. Thirdly, the Court used these two laws to read back the status of the previous 9 Basic Laws, giving them superiority over regular laws, so as to create a quasi-constitutional regime, based on the 11 existing Basic Laws.57 The second stage of Israeli constitutional law does, therefore, have a textual basis, in the form of the two new Basic Laws and the reinterpretation of the 10 older ones. However, at HCJ 243/62 Ulpanei Hasrata v Gery 16 PD 2407 [1962] (in Hebrew) (citing 5 American cases). HCJ 153/83 Levi v Police Commissioner for the Southern District 38(2) PD 393 [1984] (in Hebrew) (citing 7 American cases, 2 English cases, and 1 Irish case) (a list of citations, sorted by country, appears at the top of every court decision in the PD publication); HCJ 2481/93 Dayan v Wilk 48(2) PD 456 [1994] (in Hebrew) (including citations from 12 American cases, 2 English cases, 2 Canadian cases, 1 German case, 1 Indian case, and 1 Australian case). 50 HCJ 399/85 Kahana v Broadcasting Authority Management Board 41(3) PD 255 [1987] (in Hebrew) (citing 8 American cases, 1 Canadian case, and 1 case of the ECtHR). 51 HCJ 4804/94 Station Film v Council of Film Supervision 50(5) PD 661 [1996] (in Hebrew) (citing 10 American cases, 3 Canadian cases, 3 English cases, and 1 Irish case); HCJ 5432/02 SIN v Council for Cable and Satellite Broadcasting 58(3) PD 65 [2004] (in Hebrew) (including citations from 3 American cases, 2 Canadian cases, and 4 American law review articles). 52 HCJ 104/87 Nevo v National Labour Court 44(4) PD 749 [1990] (in Hebrew) (citing 2 English cases, and 2 cases of the ECtHR). 53 HCJ 721/94 El-Al v Daniloviz 48(5) PD 749 [1994] (in Hebrew) (citing 8 American cases, 7 Canadian cases, 2 cases of the ECtHR, and 1 English case). 54 Basic Law: Freedom of Occupation originally enacted in 1992, replaced in 1994. 55 This Basic Law protects, on top of the right to human dignity, also the rights to property, movement, and privacy. 56 CA 6821/93 United Mizrahi Bank Ltd v Migdal Cooperative Village 49(4) PD 221, 353 [1995] (in Hebrew). 57 See Sapir (n 32) 89–95. 48 49
160 Iddo Porat least in terms of a Bill of Rights, this basis is very incomplete and idiosyncratic. The gap between the limited set of rights given by the text, and a more complete set of rights, was breached by the Court itself, in the form of the very broad reading of these Basic Laws. In many respects, therefore, after 1992 Israeli constitutional law continued the tradition of building up constitutional rights by judicially made law, rather than by basing them on text. Consequently, non-textualism remains to a large extent a valid explanation for the use of foreign law. Indeed, the ‘Constitutional Revolution’ provided an additional reason for relying on foreign law, since it was seen as an opportunity to revamp and revisit the existing set of judicially made rights. The period following the United Mizrahi Bank case was characterised by enthusiastic and hectic judicial work, aimed at making use of the newly acknowledged constitutional framework in order to enlarge the scope of constitutional rights, and revisit those rights that were already established, based on the new Basic Laws. As was the case in the first (and more limited) wave of constitutional creativity, this project could find considerable support, both in content and in legitimacy, by referring to other Constitutions and other constitutional courts which have a larger and more established set of constitutional rights. Accordingly, in many of the decisions in which as part of the ‘Constitutional Revolution’ new rights were acknowledged through interpretation, or old rights were revisited in view of the new Basic Laws, one can find extensive recourse to foreign law. These decisions include the cases in which the right to human dignity was extended so as to also include the right to equality, the right to family, the right to education, the right to free speech and the right to minimal living conditions.58 In addition, one may regard the entire thrust for a ‘Constitutional Revolution’ as based on a comparative claim, namely that Israel is amongst the few Western democracies without a constitutional regime, and without judicial review, and that it should ‘catch up’ with other countries in terms of its constitutional law (this claim was raised in the United Mizrahi Bank case). Even the way the Court legitimised its bold move of recognizing judicial review without clear textual authorization, was made by allusion to comparative law. Court President Barak, referred to Marbury v Madison,59 where the American Court also used judicial interpretation to claim its authority for judicial review. iii. European Law A third factor that can explain the intensive use of foreign law in Israeli constitutional law is the adoption of a European based model of constitutional adjudication in Israeli constitutional law. The two Basic Laws of 1992 have a limitation clause framed after the Canadian Charter of Rights and Freedoms, which is based on a European model and a model that exists in international human rights Conventions. Section 8 of Basic Law: Human Dignity and Liberty reads:
58 HCJ 5578/02 Manor v Minister of Finance 59(1) PD 729 [2005] (in Hebrew) (citing two American cases, three South African cases and six Canadian cases). 59 Marbury v Madison 5 US 137 (1803).
The Use of Foreign Law 161 There shall be no violation of rights under this Basic Law except by a law befitting the values of the State of Israel, enacted for a proper purpose and to an extent no greater than is required.
The wording resembles the wording of the general limitation clause in the Canadian Charter.60 The latter part of the clause was interpreted as imposing a proportionality test.61 The adoption of a proportionality test and of the European-based limitation clause model resulted in an increased willingness to look into models of countries that also use proportionality and share this model, termed by Loraine Weinberg ‘the postwar constitutional paradigm’.62 German constitutional law, for example, has been cited in several key decisions developing the new tests for the application of the proportionality test, as well as the Canadian constitutional law.63 By adopting the proportionality test, the Israeli Supreme Court has entered the ever growing family of constitutional courts that use proportionality, which has allowed it to borrow more easily from those systems and create a dialogue with them.64 A similar phenomenon has been well documented with regard to other constitutional courts adopting the European model, and many times this came at the expense of citing American constitutional law, which has a substantially different doctrinal framework.65 There is no data regarding a decline in the use of American constitutional law in Israel, but some of the examples cited above may indicate that there is an increase in the use of foreign law from such countries as Canada, Germany and South Africa. Hand-in-hand with adopting the methodology of European constitutionalism, Israel may have also gotten closer to the substantive commitment of the European model that differs from the American one. As Jed Rubenfeld has argued, the European conception of rights is universalistic and elitist, while the American one puts a high premium on popular democracy and sovereignty.66 In addition, the European model is more expansive in terms of the concept of rights, and allows for a positive rather than only negative rights, and for rights to have effect in the private sphere and not only in the public one.67 These developments characterise Israeli constitutional law, and further promote and ease the use of foreign law in it. 60 S 1 of the Canadian Charter of Rights and Freedoms maintains that the Charter ‘guarantees the rights and freedoms set out in it subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society’. 61 See United Mizrahi Bank (n 56) 436. 62 LE Weinrib, ‘The Postwar Paradigm and American Exceptionalism’ in S Choudhry (ed), The Migration of Constitutional Ideas (Cambridge, Cambridge University Press, 2007) 84, 84. 63 See, eg United Mizrahi Bank (n 56) 436–37 (citing German and Canadian authorities in the interpretation of proportionality). 64 For the documentation of the spread of proportionality in different constitutional systems, and for the dialogue it allows between them see, eg see A Barak, Proportionality: Constitutional Rights and their Limitations (Cambridge, Cambridge University Press, 2012) 181–208; A Stone Sweet and J Mathews, ‘Proportionality Balancing and Global Constitutionalism’ (2008) 47 Columbia Journal of Transnational Law 72, 74–75, 111–59; M Cohen-Eliya and I Porat, ‘The Hibidden Foreign Law Debate in Heller: Proportionality Approach in American Constitutional Law’ (2009) 46 San Diego Law Review 367, 380–81. 65 M Waters, ‘Mediating Norms and Identity: The Role of Transnational Judicial Dialogue in Creating and Enforcing International Law’ (2005) 93 Georgetown Law Journal 487, 493 n 25 (describing the increase in citations of European countries, and countries such as Australia and Canada, at the expense of citations of American cases). See also, A Liptak, ‘US Court is Now Guiding Fewer Nations’, New York Times, NY, 18 September 2008. 66 J Rubenfeld, ‘Commentary, Unilateralism and Constitutionalism’ (2004) 79 New York University Law Review 1971, 1991–99. 67 See M Cohen-Eliya and I Porat, ‘The Culture of Justification’ (2010) American Journal of Comparative Law 463, 474–81.
162 Iddo Porat iv. The Non-Formalist Nature of Israeli Constitutional Law The fourth central feature of Israeli constitutional law that explains the extensive use of foreign law in Israel is its non-formalist and open-ended nature, especially since the 1980s. This feature is obviously related to the non-textual nature of Israeli constitutional law, but it is not necessarily the product of it, and is related also to the general move in Israeli law towards anti-formalism. Since the late 1970s or early 1980s the Court has been on a continuous move towards an open-ended mode of interpretation, higher involvement in public life, and greater supervision over administrative and legislative actions. In short, since the late 1970s the Court has been showing clear signs of judicial non-formalism and activism. Although it has not exerted judicial review over primary legislation made until the ‘Constitutional Revolution’ of 1992, it has dramatically upgraded its ability to supervise administrative decisions, and even internal decisions, of the Parliament by loosening up entry barriers such as standing and justifiability, and by adopting a ‘purposivist’ and non-formalist mode of interpretation. There are also several empirical pieces of evidence for the increase in anti-formalism in the Court’s jurisprudence; these include the on-average lengthening of court decisions, and the increase in the relative part of academic and literary citations at the expense of statutory citations. It is interesting to note that the length of court decisions was relatively high also in the earlier and formative years of Israeli jurisprudence, in which canonical cases such as Kol Ha’am paved the way for later cases, and set up the foundations for the Israeli legal system.68 Professor Mautner reviews several reasons that explain why the court became more anti-formalist in the 1980s. These reasons are: (1) the shift within Israel from a society that was more collectivist, socialist and State-centred to one that is more individualistic, capitalistic and more suspicious of the State. This shift has accorded the courts a greater role in Israeli society, and has shifted to them more decision power. This is manifested in the fact that the Court has been assuming a less formalistic role, and adopting a broader conception of its function; (2) the weakening of the Israeli political system, due to the fracturing of major parties to smaller ones, and due to unstable coalitions, which again channelled power to the Court, and encouraged non-formalism and a larger conception of its role; and (3) the loss of political hegemony of the liberal leftist party to the rightist party and to the religious parties, and the use of the Court by the liberal left to rebalance its loss of political power, given its liberal inclinations.69 Whatever the reason, the dramatic move of the Court towards activism and anti- formalism since the early 1980s can also explain its extensive use of foreign law in at least three ways. First, anti-formalism represents an expansion of the notion of legality and of interpretation, and therefore allows and even encourages the use of non-formal legal sources. The more the judiciary steers away from seeing its own function as applying mechanically pre-existing law, and moves towards the role of shaping the law according to considerations of policy and justice, the more it becomes relevant for the judge to look into a variety of sources that deal with similar policy or moral questions, regardless of their formal pedigree. 68 See generally, M Mautner, Law and Culture in Israel (Oxford, Oxford University Press, 2011) ch 4 (originally published as M Mautner, ‘The Decline of Formalism and the Rise of Values in Israeli Law’ (1993) 17 Tel-Aviv University Law Review 503 (in Hebrew)). 69 ibid.
The Use of Foreign Law 163
Secondly, anti-formalism in judicial writing has also resulted in a judicial style that is similar to an academic style. This is particularly characteristic of former Supreme Court Justice Aharon Barak’s style, himself a former law professor and Dean of the Faculty of Law at the Hebrew University. Academic style includes citations from a variety of sources, including comparative law and foreign law review articles. As in a law review article, an ‘academic-style’ court decision would not restrict itself to the solution of the particular legal dispute, but attempt to encapsulate the general parameters of a particular legal question, including history, statutes, case law, policy questions, academic literature and comparative law. A survey of comparative law is considered to be part of the ‘due diligence’ standard of Israeli academic writing. For example, in some Israeli law schools, an Israeli law student who writes a research paper on a general legal question is required to survey comparative law in her paper, and her grade would be affected if she did not do so. Thirdly, one of the important implications of anti-formalism in the field of constitutional law is the lowering of procedural barriers of entry to the Court. During the 1980s the doctrines of standing and of political question loosened up considerably.70 This resulted, among other things, in the ability of civil society organizations and human rights groups to petition directly to the Court. Among these organizations are: the Israeli Association of Civil Rights, Adala (the Legal Centre for Arab Minority Rights in Israel), the Israeli branches of the International Red Cross and of ‘Doctors without Borders’, as well as many other groups. With almost no formal requirement of standing and a loose requirement of political question, such groups can petition the Court not only as representing claimants that were particularly harmed by governmental action, but also on their own initiative addressing general matters pertaining to the rule of law. The role of human rights organizations in Supreme Court petitions and the number of petitions originated by such groups have therefore increased substantially. Such organizations specialise more than other petitioners in international and comparative law, as they consider themselves part of an international community of civil society and human rights organizations, and as they often try to import what they would consider to be more advanced notions of human rights into the local system. These petitions contribute therefore to the more frequent use of foreign law and international law in legal briefs and consequently in Supreme Court decisions.71 B. Israeli Law Generally and the Use of Foreign Law i. A Young Legal System In its first stage, as a young legal system, Israel had no prior case law of its own to draw on. The next best thing was Mandatory case law, which was heavily influenced by English law. Early Court decisions cited mainly Mandatory as well as English cases, which were both formal sources of law at the time (the tie to English cases as a formal See D Barak-Erez, ‘The Justiciabltiy Revolution Revisited’ (2008) 40 The Lawyer 3, 3–4. Compare, Y Elbashan, ‘Aharon Barak – the Juridication of Protest’ in C Fassberg B Medina and E Zamir (eds), The Judicial Legacy of Aharon Barak (Jerusalem, Mishpatim, 2009) 139, 141–42 (in Hebrew) (arguing that Aharon Barak J was respectful towards human rights organizations, and raised their status in Supreme Court litigation to that of central players in the shaping of law, rather than marginal players). 70 71
164 Iddo Porat source of law was abolished only in 1984).72 The use of non-Israeli materials in the early years was therefore common and natural, and partly formalised. This was true of all legal decisions, including decisions on constitutional matters. ii. Geopolitical Isolation Since Israel is a democracy, surrounded mostly by hostile and non-democratic countries, it has always sought legitimacy and acceptance from the West – from Europe and North America in particular, and wished to be integrated in Western culture. The use of American and European legal materials is one way of such a cultural exchange and integration. Israel is also relatively isolated in terms of the extent of relevant legal materials that can be used to expand its legal horizon and imagination. The United States, for example, has a large repository of legal materials to draw on from the legal systems of the 50 States and of the federal system. This is also so with Europe, which, especially since the formation of the European Union, has very important legal interactions between the different countries and can also draw on the jurisprudence of the European Court of Justice, and the ECtHR. In Israel, however, there is no such regional or internal large repository of legal materials, so that in order to go beyond its relatively limited scope of legal materials it must draw generally on foreign materials. Some view the fascination with the West as part of a post-colonial syndrome, shared by other former British colonies, in which the newly created State attempts to mimic the former colonial ruler, and looks for it as a role model.73 Others find its reasons in the ever increasing importance of American culture, including its legal culture.74 What might be one of the distinct signs of the attempt at cultural assimilation or mimicry is the use of legal icons, legal epitaphs and legal quotations from foreign rather than only local sources. An Israeli justice wishing to spice up her free speech decision with famous quotations, for example, would turn to Holmes J, and Brandeis J, almost as naturally as she would turn to Agranat J. Leading American academics such as Ronald Dworkin, Lawrence Tribe, and Alexander Bickel also appear regularly in the repertoire of Israeli court decisions.75 One should note again that, given the young age of the Israeli judicial system, not many of its justices have attained the stature of canonical figures, legal icons, or larger than life figures, so that sometimes Israeli justices look for them outside the borders of Israel. Jewish giants of Hebrew law are apparently too remote in time, in professional affinity and in worldview to effectively fill this niche. iii. Professional and Academic Ties The third factor that shapes the use of foreign law in Israel is the close ties of Israeli legal professionals with foreign legal systems. Israeli lawyers, judges and law professors have See nn 8–10 above and accompanying text. According to Binyamin Blum, ‘Israel looks to England and other nations for parental approval of its jurisprudence. In this regard Israel can be viewed as experiencing a broader postcolonial syndrome, residual of the Mandate’, Blum (n 2) 2161. 74 See Sandberg (n 2) 54 (text accompanying nn 18–20 above). 75 See, eg recently, CA 751/10 John Doe v Dayan-Urbach (8 February 2012), Nevo Legal Database (by subscription) para 64 of Court Vice President Rivlin’s opinion (in Hebrew) (quoting Holmes, and citing Tribe, in a free speech opinion). 72 73
The Use of Foreign Law 165
always had close ties with other countries, and many of them acquired their legal training abroad. This is definitely the case with the first generation of law professionals, in the first two decades since independence, most of whom were born and trained abroad in a large variety of countries from Russia and Poland in the East to Germany, Italy and England in Central and Western Europe and up to the United States.76 These lawyers, judges and academics possessed, therefore, a wealth of knowledge on a wide array of legal systems and had easy access to them. These assets were naturally integrated into their legal products – in court decisions, legal briefs and legislation. Most of the lawyers, judges and professors of the following generations were already born in Israel and had their basic legal training in that country. However, higher legal education, especially for those wishing to follow an academic career, continued to be carried out predominantly abroad, especially in the US. According to a 2009 survey, 65 per cent of Israeli law professors received their graduate and postgraduate education in North America.77 Consequently, over time, the former close acquaintance with the laws of European countries has dwindled, and on the other hand US influence has grown considerably. Furthermore, in the Israeli legal profession academic prestige is highly dependent on publications abroad, especially in US law reviews. In consequence, academic research and writing in Israel is oriented towards such areas of law that would be palatable to an American legal audience, and a large number of Israeli academics write more on US law than on Israeli law. As regards judicial ties with foreign legal systems, several justices (especially those who followed an academic career prior to their appointment to the Court such as Barak and Zamir JJ) earned advanced degrees in the US and England, and have integrated their knowledge into their judicial products.78 In addition, many Supreme Court justices over the last two decades have had foreign law clerks in their chambers, a practice that is now almost standard in the Israeli Supreme Court. Many of these clerks are second or third year law students coming from top US law schools, such as Yale and Harvard, but others come also from Germany, Canada, and other countries. These law clerks provide important information and easy access to foreign law, and are one more indication of the demand for foreign law resources by Israeli Supreme Court justices. IV. NORMATIVE ASSESSMENT
The preceding review was descriptive in nature, and aimed at mapping the main factors that affected the use of foreign law in Israeli constitutional law. The next part will deal with its normative aspects. It will attempt to assess the use of foreign law in Israeli constitutional decisions by discussing the different points of criticism addressed against such use.79 The most comprehensive set of attacks on the use of foreign law can be found in the burgeoning discussion over the use of foreign law in US constitutional law. I will thus rely on this discussion in order to map the different types of criticism and then turn to their application to the Israeli case. 76 E Salzberger, ‘The Supreme Court of Israel at 50 Years – Between a Hedgehog and a Fox’ (2000) 16 Bar-Ilan Legal Studies 141 (2000) (in Hebrew); see also Salzberger and Oz-Salzberger (n 21). 77 See Sandberg (n 2) 52. 78 See ibid 53. 79 Arguments in favour of using foreign law are not reviewed separately but as part of the discussion on the critique of such use, as, in most countries, using foreign law citation is the norm.
166 Iddo Porat A. Substantive Objection – Originalism The one major substantive objection to the use of foreign law in constitutional adjudication is that it undermines the democratic choice of the local population, and subjects its own conceptions of rights and of their proper scope to the notions of other peoples and international bodies.80 In the United States, in particular, this objection is tied to the originalist movement, which contends that the Constitution should be interpreted according to the meaning assigned to it originally, when it was drafted, since this meaning only reflects the will of the people.81 Foreign law is immaterial to the task of expounding the will of the people who drafted the Constitution, and therefore should not be used in interpreting the Constitution.82 Originalism bases the legitimacy of constitutionalism and of judicial review on democracy – the Constitution reflects the higher will of the people, and the Court only imposes the people’s higher will on the people’s ‘regular will’ as expressed by regular laws. However, as noted earlier, originalism does not travel well outside the United States. The legal and constitutional cultures of most other nations are different and do not put such an emphasis on sovereignty and on original intent.83 Moreover, an originalist interpretation of the Israeli quasi-Constitution may show that the meaning intended by it at the time of the drafting was not so alien to the use of foreign law in its interpretation and application. Other recent Bills of Rights, that of South Africa for example, make the use of foreign sources a formal requirement, and other ones, such as the New Zealand Bill of Rights, allude to them in the Preamble and in the travaux perperatoires of their drafters.84 In the Israeli case the evidence is less clear, but the fact that the Israeli Knesset has chosen consciously to draw on an existing model from a foreign country – Canada – could be taken to mean that the Knesset expected the Court to draw on the Canadian experience, and maybe also on the experience of other countries with a similar model of constitutionalism, such as Germany. Israeli constitutional culture and the circumstances under which the Basic Laws were enacted do not support, therefore, a strong objection to the use of foreign law, based on original intent. However, the democratic problem seems to lurk in the background of some more nuanced problems that may arise because of the use of foreign law, and these will be discussed later on with regard to global constitutionalism.
80 See Scalia J’s position in the Scalia–Breyer debate set out in N Dorsen, ‘The Relevance of Foreign Legal Materials in United States Constitutional Cases: A Conversation between Justice Antonin Scalia and Justice Stephen Breyer’ (2005) 3 International Journal of Constitutional Law 519, 521–22, 524–25. 81 See generally, RH Bork, The Tempting of America (New York, The Free Press, 1990) 143–53; A Scalia, ‘Originalism: The Lesser Evil’ (1989) 57 University of Cincinnati Law Review 849, 863–64. 82 See, eg Roper v Simmons 543 US 551, 622–28 (2005) (Scalia J, dissenting). 83 See J Allan and G Huscroft, ‘Constitutional Rights Coming Home to Roost? Rights Internationalism in American Courts’ (2006) 43 San Diego Law Review 1, 32–33. 84 The Preamble to the New Zealand Bill of Rights Act 1990, states that one of the purposes of the Act is to affirm New Zealand’s commitment to the International Covenant on Civil and Political Rights (adopted 16 December 1966, entered into force 23 March 1976) 999 UNTS 171. Moreover, the White Paper to that Act made clear that the use of comparative law was contemplated by the Government that proposed the Bill of Rights. J Allan, G Huscroft and N Lynch, ‘The Citation of Overseas Authority in Rights Litigation in New Zealand: How Much Bark? How Much Bite?’ (2007) 11 Otago Law Review 433, 445: www.ssrn.com/abstract=1095541.
The Use of Foreign Law 167
B. Methodological Objections – Cherry Picking and Bad Comparisons There are two major methodological objections to the use of foreign law in constitutional adjudication. The first is that it is undisciplined. Since not all foreign law decisions can be taken into account in each decision, and as there is no rule regarding which foreign law a judge should use, the choice, so it is argued, is necessarily arbitrary and boils down to ‘cherry picking’, that is to choosing such foreign law material which is supportive of the judge’s preferred legal outcome. According to Scalia J, for example, a particular foreign doctrine is either authoritative, in which case its use is mandatory, or not authoritative, in which case its use is completely undisciplined and open to judicial manipulation.85 Cherry picking is manifested not only in decisions on which foreign law to turn to in a particular legal issue, but in the first place in the very selection of the cases in which one turns to foreign law. Some US justices, for example, choose to turn to foreign law to support their views on homosexual rights and on the death penalty, but not on abortions nor on criminal defendants’ rights.86 The second criticism is based on the impossibility of comparison and of migration of legal concepts. It claims that the use of foreign law is misleading because any comparison would have to take into account so many factors that it would be hard to draw any conclusions from it. Since any statutory provision or court decision is so deeply embedded in the particular legal culture it belongs to, as well as in the general culture, history, political climate and particular circumstances of the case, the use of it out of its specific context would teach us nothing. The same reasons would also make any attempt at legal transplantation almost impossible.87 Unlike objections based on originalism, objections based on cherry picking and bad comparisons, definitely have a hold in the context of Israeli constitutional law. To illustrate this point one can go back to the ‘father’ of all foreign law citations in Israeli constitutional law, the Kol Ha’am case.88 In Kol Ha’am, Agranat J cited the Dennis case in order to transplant the American free speech balancing test into Israeli constitutional law. As described earlier, this was no less than a revolutionary moment that helped establish and protect the right to free speech in the years to come. However, Agranat J completely disregarded the role of the balancing test in American free speech law, which was the inverse of the role it played in the Kol Ha’am case. As lamented by the dissenting justice in the Dennis case, the balancing test diluted the previous clear and present danger test, and amounted to a judicial admonition of the Congress’s sense of when hurting free speech would be reasonable.89 The crucial comparative difference is that the United
85 See Dorsen (n 80) 522 (Scalia J: ‘What’s going on here? Do you want it to be authoritative? I doubt whether anybody would say, “Yes, we want to be governed by the views of foreigners”. Well if you don’t want it to be authoritative, then what is the criterion for citing it? That it agrees with you? I don’t know any other criterion to bring forward’). 86 See Roper (n 82) 622–28 and the examples there in Scalia J’s dissent. 87 See P Legrand, ‘The Impossibility of “Legal Transplants” ’ (1997) 4 Maastricht Journal of European and Comparative Law 111 (arguing that the cultural barrier makes legal transplants virtually impossible); MD Ramsey, ‘International Materials and Domestic Rights: Reflections on Atkins and Lawrence’ (2004) 98 American Journal of International Law 69, 73–74 (arguing against the comparison drawn in the Lawrence case between American constitutional law and the European legal context, since the legal tests diverge substantially). 88 Kol Ha’am (n 39). 89 Dennis (n 47) 579–81 (Black J dissenting).
168 Iddo Porat States had a formal Constitution with a formal and absolute protection of free speech,90 so that a balancing test in effect diluted the absoluteness of the free speech precept that ‘Congress shall make no law abridging the freedom of speech’. In Israel, which has no formal Constitution and no textual protection of free speech, the balancing test was used to create the right to free speech by reading the balancing test into the law that authorised censorship. What was balanced was the administrative unlimited discretion to censor speech, rather than the right to free speech itself. This was an ingenious move, but one that disregarded the comparative difference between the two constitutional cultures. The second example is taken from the second wave of constitutional creativity. In the United Mizrahi Bank case of 1995 Court President Barak relied partly on the Marbury case to legitimise a judicially acknowledged right to judicial review.91 The comparative difficulty here is the fact that Marbury found this right in a full blown formal Constitution that was ratified in a long, demanding and comprehensive process, while the Court in United Mizrahi Bank relied on two Basic Laws that were ratified in a regular act of legislation, without much public attention, and with a low participation of Knesset Members. What can be derived from the former process of Constitution-making cannot be derived from the latter. Marbury itself, of course, is accused of cherry picking and bad comparison, by arguing that judicial review logically follows from the superiority of the Constitution and disregarding comparative counter examples in Europe at the time.92 The citation of Marbury amounts therefore to a ‘double’ bad comparison. There are many other examples of course including: the transplantation of the exclusionary rule while disregarding the difference between a bifurcated jury system and a unitary judiciary,93 turning to comparative law to establish a consensus around gay rights while diminishing the role of countries that do not follow that move,94 doing the same with regard to establishing a right to affirmative action,95 and other cases. However, while cherry picking, bad comparisons are real problems for Israeli constitutional law, they do not seem to be problems which apply only to Israeli law, or only to constitutional law, or even only to the citation of foreign law. Professor Tushnet, for example, argued that the cherry picking objection, ‘could be made . . . about nearly every approach to constitutional interpretation’, including about the selection of historical data for originalist-based interpretation, and judges often rely on hosts of non-authoritative sources, that similarly may raise the problem of cherry picking and bad comparisons, such as lower courts’ decisions, law review articles, public policy memoranda and the like. One may argue that the use of foreign law is more prone to the problems of cherry picking and to bad comparisons than the use of other non-authoritative materials. This is so since foreign law provides an especially large reservoir of examples, and thus amplifies the possibilities of manipulation, and that comparisons are especially hard to make 90 See the First Amendment to the United States Constitution providing in absolute terms that ‘Congress shall make no law . . . abridging the freedom of speech’, US Constitution, amend I. 91 United Mizrahi Bank (n 56) paras 75–76 of Court President Barak’s opinion. 92 See, eg M Troper, ‘Marshall, Kelsen, Barak and the Constitutionalist Fallacy’ (2005) 3 International Journal of Constitutional Law 24. 93 CA 5121/98 Yissacharov v Chief Military Prosecutor 61(1) PD 461 [2006] (in Hebrew). 94 See Dorner J’s opinion in Daniloviz (n 52) and the comparative examples she cites there, all of which support the expansion of gay rights. 95 See HCJ 453/94 Israel Women’s Network v Government of Israel 48(5) PD 510 [1994] Cheshin J’s opinion (in Hebrew).
The Use of Foreign Law 169
between different cultures. Constitutional law is sometimes believed to be especially imbedded in the culture of a nation because it represents its raison d’être, and therefore it may be harder to draw comparisons in this particular area of law. However, the difference between the dangers of foreign law manipulation and those of other types of manipulation appears to be a difference of degree rather than of kind, and the danger in using foreign law can sometimes be exaggerated. It seems, therefore, that we should look elsewhere if we wish to find a more principled objection to the use of foreign law. C. The Pitfalls of Global Constitutionalism After having shown that the originalist objection does not apply so much to Israel and that the methodological problems of the use of foreign law do not present a unique or particularly worrying problem, I would now like to point out several aspects in which the use of foreign law could be genuinely troublesome. i. Ratcheting Up, Race to the Top and Human Rights Organizations While cherry picking is not unique to the use of foreign law, it seems that a particular kind of cherry picking, related to the effects of globalism, is unique to the use of foreign law, and might present special problems. A New Zealand survey has found that the use of foreign law in interpreting the New Zealand Bill of Rights Act has been substantially biased in favour of expanding notions of rights, rather than diminishing them.96 Although New Zealand judges could find outside their own system examples of both diminished and enlarged scopes of rights, and of different ways of line drawing between rights and public interests, they tended to cite mostly those that expanded notions of rights and drew the line closer to rights, and tended to disregard those systems in which rights had a more diminished scope. The use of foreign law there had a ratcheting up effect on the scope of rights, at the expense of other considerations. David Law identified a similar phenomenon regarding the interrelations between different countries and different constitutional courts, and called it a ‘race to the top’.97 Using empirical data, Law found that countries and constitutional courts around the world look at each other’s jurisprudence and compete amongst themselves over who would be more ‘advanced’ or expansive in terms of rights protection. Law bases his finding on an economic explanation,98 but it is confirmed also by common wisdom. It would seldom be the case that a court would be proud of itself for being the least protective of rights, and more often one could find a court lamenting the fact that one’s own system ‘lags behind’ or does not ‘catch up’ with other systems in terms of rights protection. A ratcheting up effect may occur also as a result of the role human rights organizations play in initiating and participating in petitions. As argued earlier,99 human rights organizations are central players in the promotion of the use of foreign law in judicial Allen, Huscroft and Lynch (n 84) 10. DS Law, ‘Globalization and the Future of Constitutional Rights’ (2008) 102 Northwestern University Law Review 1277. 98 David Law argues that ‘As capital and skilled labor become increasingly mobile, countries will face a growing incentive to compete for both by offering bundles of human and economic rights that are attractive to investors and elite workers’, ibid 1282. 99 See n 71 above and accompanying text. 96 97
170 Iddo Porat opinions, and such organizations have an inherent bias in favour of expanding the notion of rights, and citing those foreign opinions that are the most expansive in terms of rights. There is no survey similar to the New Zealand one with regard to Israeli constitutional law, but a quick review of many of the cases mentioned above for selective use of foreign law would show that many of them were selective in one way only – by citing only those authorities that expand local notions of rights. This is true with regard to the issue of affirmative action,100 of gay rights,101 of the exclusionary rule,102 and of social and economic rights such as the right to minimal existence and the right to education. Binyamin Blum103 has shown nicely the way the Israeli Supreme Court made an effort to portray its ruling on evidentiary law as adopting the most ‘advanced’ standard in the global market – the exclusionary rule – participating, as it were, in the race to the top of rights protection in this field. This, despite the fact that the ruling itself did not diverge dramatically from previous precedent, and its portrayal as adopting the exclusionary rule subsequently caused much confusion in the lower courts’ jurisprudence. There are also counter examples, notably not choosing the American position regarding the legality of racism and other forms of incitement or of hurting public feelings, but here too the Israeli Supreme Court relies on a wide consensus among European countries, and cannot be regarded as diverging from a high standard of rights protection. ii. Writing to a Global Audience: Transjudicialism, the Global Community of Judges, and the Court as an Ambassador Constitutional judges see themselves more and more as part of a global community of judges104 and as participating in what Ann Marie Slaughter called transjudicial communication.105 Slaughter has documented the way constitutional judges around the world participate in international conferences, meet each other, and correspond with each other over the pages of their court decisions. Since rights jurisprudence is a global language, much of this communication is concerned with rights, and rights adjudication may become the lingua franca of transnational law. Today many courts make a conscious effort to transmit their decisions to other courts, and to expand their global influence, and this is aided by technological advancements that make such transmission easy and quick. The German Federal Constitutional Court, for example, issues English press releases and sends them to its counterpart courts in other countries.106 The Supreme Court of Canada sends automated messages to subscribers, giving them one week’s notice of impending decisions, along with brief summaries of the relevant history of a case.107 The Israeli Court is no exception. Its leading decisions are translated and put on the website Israel Women’s Network (n 95). Daniloviz (n 53). 102 Yissacharov (n 93). 103 Blum (n 2) 2169–71. 104 A Slaughter, ‘A Global Community of Courts’ (2003) 44 Harvard International Law Journal 191. 105 A Slaughter, ‘A Typology of Transjudicial Communication’ (1994) 29 University of Richmond Law Review 99; see also, A Slaughter, ‘Judicial Globalization’ (2000) 40 Virginia Journal of International Law 1103. 106 See www.bundesverfassungsgericht.de/links.html. This link opens a website in German, which lists the German Government institutions and links to other courts. 107 See Allan and Huscroft (n 95) 11. 100 101
The Use of Foreign Law 171
of the Court sometimes a short span of time after they are published in Hebrew. Israeli judges are sought after participants in international conferences, and global meetings of constitutional judges, because of the interest Israeli law promotes around the world, and they promote their decisions in those venues as well. The result of this is that constitutional judges are writing their decisions knowing that they would be read by a global community; they therefore inevitably write also to the global audience, and may try to ‘sell’ their decision to that audience, present it in a way that would be palatable to that audience, or even in a way that would distinguish it and portray it as innovative, or advanced. One may talk in this respect about judges getting ‘compensation’ in terms of international recognition, and participation in international conferences and events. This phenomenon exceeds the mere citation of foreign law, but should be regarded as part of it, since it addresses the more general phenomenon of the global influences on local constitutional law. Several Israeli decisions portray this transjudicial dialogue, or a possible attempt to address the palate of the international audience. Leading cases in terms of the protection of rights versus threats to State security, such as the precedential case banning torture,108 and to a lesser extent, the decision limiting ‘targeted killing’,109 were very well received abroad, and presented by Israeli justices as attesting to the way Israel is leading among democracies in rights protection. In another decision, pertaining to the holding of Lebanese Hizballa militants as bargaining chips for the release of an Israeli captive soldier, the Court changed its mind in a second hearing of the case, following strong internal, as well as international and academic criticism of the first decision.110 While obviously these decisions are led by other considerations as well, the effect they have in the global community of judges and lawyers cannot be overlooked. Another example that shows a more direct dialogue between the Israeli Court and the international community pertains to the ruling regarding the security fence erected between Israel and the Palestinian territories.111 The timing of the decision was very telling; it was given a short time before an expected decision of the International Court of Justice on the legality of the fence. It cannot be interpreted other than as partaking in a dialogue with that other court, showing, as it were, that Israel does stand to international standards of human rights protection, with regard to the fence. Other courts also act similarly. The Lithuanian Constitutional Court, for example, issued a ruling in 1998 banning the death penalty in the State, despite an overwhelming objection in the populace.112 This could be interpreted as a move calculated to please the audience of the European Union, and ease HCJ 5100/94 Public Committee against Torture in Israel v State of Israel 53(4) PD 817 [1999] (in Hebrew). HCJ 769/02 Public Committee against Torture in Israel v Government of Israel 62(1) PD 507 [2006] (in Hebrew). 110 CrimFH 7048/97 John Does v Ministry of Defence 54(1) PD 721 [2000] (in Hebrew) (Barak J changing his opinion in the previous round). 111 HCJ 2056/04 Beit Sourik Village Council v Government of Israel 58(5) PD 807 [2004] (in Hebrew) )trs in (2005) 38 Israeli Law Review 83(. 112 The Constitutional Court of the Republic of Lithuania, ruling ‘On the compliance of the death penalty provided for by the sanction of Article 105 of the Republic of Lithuania Criminal Code with the Constitution of the Republic of Lithuania’ Vilnius, 9 December 1998 (official website of the Lithuanian Constitutional Court, www.lrkt.lt/dokumentai/1998/n8a1209a.htm). The Court based its decision on Art 19 of the Constitution, protecting the right to life, and on Art 21, providing that it shall be prohibited to torture, injure, degrade, or maltreat a person, as well as to establish such punishments. A November 1998 survey showed that 78% of Lithuanians were against and 10% for the abolition of the death penalty. V Gaibidys, ‘Attitudes Towards the European Union in Lithuania: 1997–2002’ (IES Proceedings 1.1, November 2002) www.ies.ee/iesp/ gaidys.pdf. 108 109
172 Iddo Porat the acceptance of Lithuania into the EU, the Court acting as an ambassador of sorts, in selling Lithuania as a rights protective State. iii. The ‘Constitutional Revolution’ and a False Sense of Consensus One may view the ratcheting up and the race to the top effects as positive developments.113 After all, they drive courts towards greater protection of human rights. However, they also hold obvious problems of legitimacy. In addition, there is another danger in such use of foreign law. Judges, who rely extensively on foreign law in interpreting rights, may find in it a false sense of consensus and agreement, which does not exist in their local community. Such judges may be reassured by the fact that their rulings are well accepted abroad, or by the fact that other nations seem to adopt similar interpretations, and disregard strong internal opposition to their rulings. This is often aided by sympathetic reaction from academia, cultural elites, and civil society organizations, all of which are also players in the global arena, and find their legitimacy and ‘compensation’ in that arena as well as in the local one. In a small State, such as Israel, which has a very strong need for international acceptance and legitimacy because of its geopolitical situation, in which the legal and judicial community has strong international ties, in which the judiciary has adopted an anti-formalist and activist jurisprudence, this danger is enhanced. The danger of losing touch with the local community is exacerbated by the shaky grounds on which the Israeli ‘Constitutional Revolution’ was founded. The Court is moving forward in great leaps constantly adding new rights to the set of constitutional protected rights, basing this move on two Basic Laws, which, as described earlier, provide very little textual support for such a move, and which were not meant to create a revolution. The comparative constitutional perspective functions both as a fuel to continue the ‘Constitutional Revolution’ and as a sedative to the danger of losing legitimacy for it within Israeli society. V. CONCLUSION
Israel has many good reasons to use foreign law in its constitutional law. It has a long tradition of using foreign law, it has relied on foreign law to create an impressive judicially made set of constitutional rights, and its constitutional culture is receptive to the comparative perspective. The use of foreign law has therefore many advantages for Israeli constitutional law, and is not expected to go away. However, seen in a more general perspective of a constant move of ratcheting up the set of rights in Israel, without textual anchoring or strong public support, and as a way of participating in a similar global move in other constitutional courts, the use of foreign law may raise important concerns, and should be evaluated more sceptically.
113 See E Benvenisti, ‘Reclaiming Democracy: The Strategic Uses of Foreign and International Law by National Courts’ (2008) 102 American Journal of International Law 241.
12 The Israeli Case of a Transformative Constitutionalism MOSHE COHEN-ELIYA
I. INTRODUCTION
T
HE NOTION OF a transformative constitutionalism is an exciting one: it invites lawyers and judges to engage in the stimulating and ambitious task of effecting a fundamental change to the code of values and attitudes of the entire nation. Constitutions that require judges to perform this job are post-traumatic; they signal a clear departure from the horrors of the past, as in the case of the Constitutions of post-Nazi Germany and post-apartheid South Africa. By its very nature, a transformative constitutionalism presumes an active role for the court that is delegated with its implementation. The jurisprudences of the German Federal Constitutional Court and the South African Supreme Court illuminate what it means in doctrinal terms to implement such a Constitution. First, there is extensive rhetorical resort to fundamental values, which imply very broad notions of constitutional rights and, hence, greater latitude for judicial review. Second, constitutional rights and values are interpreted as positive rights, imposing an active duty on the government to implement them. Third, the constitutional provisions are broadly formulated, applying not only to the government but also to individuals (Drittwirkung). And finally, the comprehensive and transformative mission of the Constitution is fulfilled by a judiciallyactive court that extensively employs balancing or proportionality procedures, deemed as necessary due to the Constitution’s broad scope and the need to contend with conflicting rights, values, and interests. In the context of Israeli law, these characteristics seem to be the most essential features of the Israeli Supreme Court’s constitutional jurisprudence. Over the past three decades, the Court has made significant reference to fundamental values, expanded its conception of rights and the scope of judicial review, and developed the notion of positive rights. It has signalled its willingness to make individuals subject to the duty to respect constitutional rights and has designated proportionality as the reigning analyt ical doctrine in constitutional law. The idea of turning Israel into a more enlightened, liberal and humanistic society has been a recurring theme in the judgments, scholarship, and lectures of Aharon Barak, the highly influential former Court President of the Israeli Supreme Court. This ideal is the legacy of the Barak Court.
174 Moshe Cohen-Eliya However, in at least one respect, there is an important difference between Israeli constitutional law and that of post-war Germany and post-apartheid South Africa, which derives not from the jurisprudence but is rooted in the constitutional text itself and intentions of the framers. While the framers of the South African and German Constitutions clearly intended to signal an unequivocal departure from the past (ie to transform the values and attitudes of the citizens and decisionmakers), this was not the case in Israel. From both the text and stated intentions of the drafters of the Israeli set of Basic Laws it emerges that the Israeli constitutional legislation corresponds more closely with what Lessig has termed with regard to the American Constitution – ‘a preservative constitution’, that is to say, one that seeks to maintain existing practices, to ensure that things do not get worse.1 The Basic Laws were not created in Israel following a dramatic turning point in the nation’s history, certainly not on the scale of post-Nazism or postapartheid. Rather, these laws were enacted incrementally, delaying to the very end the enactment of Basic Laws on fundamental rights. Though the liberal Members of the Knesset (Israeli Parliament) had intended to improve human rights protection, the Basic Laws’ texts do not imply a radical break with the past. Rather, they make broad reference to the same core Jewish and democratic values of the State of Israel that were first laid out in the 1948 Israeli Declaration of Independence.2 The Israeli case is thus a unique instance of a transformative constitutionalism, in that the ambitious project is undertaken by the Supreme Court in the absence of a nationally defining moment and without legitimacy from the constitutional text. Rather than implementing the agreed-upon fundamental values, the Israeli Supreme Court justices are sometimes perceived as rolling up their sleeves and climbing into the political arena as participants in the cultural battle being waged. There have been clear ramifications: significant backlash from politicians, with an unprecedented number of proposed Bills seeking to limit the Court’s power and, even more worrying, a dramatic decline in public trust in the Supreme Court, which cannot be explained simply as a manifestation of Israelis’ general scepticism towards the authorities. This chapter seeks to describe and evaluate the Israeli transformative constitutionalism and proceeds as follows. After describing the concept of such a constitutionalism in Part II, I present in Part III the transformative features of Israeli constitutional jurisprudence, including value-rhetoric, positive rights, Drittwirkung, and proportionality. To this end, the transformative German and South African constitutional jurisprudences are contrasted with the non-transformative features of American constitutional jurisprudence, with Israel then situated between the two. I briefly consider the weaknesses of the judicially transformative project in Israel due to the heterogeneity of its society and the significant lack of consensus over values. I conclude in Part IV by proposing what I take to be a more promising alternative for Israel: a democracy-reinforcing approach in judicial review.
1 CR Sunstein, Designing Democracy: What Constitutions Do (New York, Oxford University Press, 2001) 67–69; see also L Lessig, Code and Other Laws of Cyberspace (New York, Basic Books, 1999) 214. 2 Declaration of the Establishment of the State of Israel, 1 LSI 7 (1948).
Transformative Constitutionalism 175
II. SOME PRELIMINARY NOTES ON TRANSFORMATIVE CONSTITUTIONALISM?
Although the notion transformative constitutionalism often connotes post-liberalism and is associated with a progressive agenda,3 this chapter refers to a more morally- neutral conception of transformation. This follows Lawrence Lessig’s definition of a transformative constitutionalism as laying out ‘certain aspirations that are emphatically understood as a challenge to longstanding practices; they are defined in opposition to those practices’.4 The idea of transformation was captured well in a South African Supreme Court post-apartheid decision: The South African Constitution . . . represents a decisive break from, and a ringing rejection of, that part of the past which is disgracefully racist, authoritarian, insular, and repressive and a vigorous identification of and commitment to a democratic, universalistic, caring and aspirationally egalitarian ethos, expressly articulated in the Constitution. The contrast between the past which it repudiates and the future to which it seeks to commit the nation is stark and dramatic.5
From a theoretical perspective, a constitutional transformation can also be understood as a shift from authoritarianism to State-neutrality. Such a transformation, however, is almost impossible to attain without certain active measures from the State, at least during the first stages of the process. For example, should post-war Germany – in light of the traumatic outcomes of Nazism – have been transformed into a neutral liberal democracy (as Krieger famously argued in The German Idea of Freedom),6 or should it have maintained its traditional organic and communitarian conception of the State, which assumed a meaningful and active role for the State in securing the common telos of the German people? The German Federal Constitutional Court opted for the latter approach, ruling that the German Basic Law is constructed on a Wertrangordnung (a hierarchal value-order) in whose implementation the State must take an active role. Underlying this choice was the understanding that adopting State neutrality in post-war Germany would have run the risk of maintaining the racist and totalitarian attitudes that were prevalent there.7 Only once the project of transformation had been generally 3 See generally, KE Klare, ‘Legal Culture and Transformative Constitutionalism’ (1998) 14 South African Journal on Human Rights 146. 4 Sunstein (n 1) 224. A related concept is that of transitional justice. See RG Teitel, ‘Transitional Jurisprudence: The Role of Law in Political Transformation’ (1997) 106 Yale Law Journal 209. Theories of transitional justice are concerned with maintaining social stability while facing up to the horrors of the past. They often confine their discussions to ‘bounded periods, spanning two regimes’ (RG Teitel, Transitional Justice (Oxford, Oxford University Press, 2000) 5) and to changes or to transitions toward more liberal regimes. For a critic, see A Gross, ‘The Constitution, Reconciliation, and Transitional Justice: Lessons from South Africa and Israel’ (2004) 40 Stanford Journal of International Law 47, 51–52. 5 S v Makwanyane 1995 (3) SA 391 (CC) para 261 (S Afr) (abolishing capital punishment). On the transformative character of the South African Constitution, see Klare (n 3). 6 L Krieger, The German Idea of Freedom: History of a Political Tradition (Boston, Beacon Press, 1957) 121, 470 (arguing that in light of the moral bankruptcy of nationalism, Germany should depart from its communitarian understanding of the rule of law and adhere instead to an American-style neutral Constitution, one that guarantees negative liberties). 7 DP Kommers, ‘German Constitutionalism: A Prolegomenon’ (1991) 40 Emory Law Journal 837, 852 (arguing that ‘the Basic Law marks a radical break with the past. In broadest outline, the gulf between Weimar and Bonn represents a major shift from constitutional democracy to constitutional democracy, emphasizing both the permanence and enforceability of certain limits on majority rule’); C Jabloner, ‘Hans Kelsen: Introduction’ in AJ Jacobson and B Schlink (eds), Weimar – A Jurisprudence of Crisis (Berkeley, University of California Press, 2000) 67, 73 (criticizing the arguments that the concepts of formal democracy and neutrality that guided the Weimar Republic were conceived of as one of the reasons for its constitutional collapse).
176 Moshe Cohen-Eliya achieved could Germany allow the transformative measures to be relaxed and shift towards a more neutral conception of the State.8 There are certain similarities between supporters of constitutional transformation (transformatists) and advocates of moral perfectionism (perfectionists). Like transformatists, perfectionists reject State-neutrality and ‘favor political institutions and state policies that do the best job of promoting good human lives’.9 Both have an ideal in mind, which they seek to perfect. They differ, however, in one critical aspect: while transformatists universally seek a radical departure from the past, this is not necessarily the case for perfectionists. For perfectionists can also be conservatives and thus strive to preserve traditional practices by way of active State intervention.10 III. THE TRANSFORMATIVE DOCTRINES
In this part, I present post-war Germany and post-apartheid South Africa as proto typical of transformative constitutionalism. Both experienced a clear departure from a traumatic past, and in both, a rich and sophisticated body of case law evolved following this transformation. Drawing from both the constitutional text and from their jurisprudence, I will identify some of the essential doctrinal features of transformative constitutionalism and then consider for each of the features whether Israeli constitutional jurisprudence can be deemed transformative. A. The Rhetoric of Values Transformative Constitutions are aspirational and tend to make numerous references to values, while indicating a sharp shift away from the past. Values have rhetorical power. They speak to the general public, and not only to sophisticated institutional players, which is necessary for effecting a change in the values of the entire nation. The Constitutions of Germany and South Africa abound with value rhetoric. The drafters of the post-war German Basic Law decided to begin the constitutional text with absolute protection of human dignity alongside several provisions that define the core values of the new Germany: substantive democracy, social justice, and federalism.11 Similarly, the Preamble to the South African Constitution proclaims the aspiration to ‘heal the divisions of the past and establish a society based on democratic values, social justice and fundamental human 8 See, eg the second abortion decision of the German Federal Constitutional Court (88 BVerfGE 203 (1993) (Ger)) (the Constitution does not require the criminalization of abortion, but imposes other limits on it), in which it retreated from its 1975 ruling that the State is under a constitutional obligation to criminalize abortion. See 39 BVerfGE 1 (1975) (Ger). 9 ‘Perfectionism in Moral and Political Philosophy’, Stanford Encyclopedia of Philosophy, 13 February 2007: www.lato.stanford.edu/entries/perfectionism-moral/#PerPol. Perfectionism is contrasted with anti- perfectionism – a strand of political moral philosophy that insists on neutrality – and is associated with Rawls, Ackerman, and Dworkin. See J Rawls, A Theory of Justice (Cambridge, Harvard University Press, 1971); R Dworkin, Taking Rights Seriously (Cambridge, Harvard University Press, 1977); B Ackerman, Social Justice in the Liberal State (New Heaven, Yale University Press, 1981). 10 Thus, perfectionism can couple both progressive liberals such as J Raz, The Morality of Freedom (Oxford, Oxford University Press, 1986), who hold sacred individual autonomy, with conservatives such as G Sher, Beyond Neutrality: Perfectionism and Politics (Cambridge, Cambridge University Press, 1997). 11 Grundgesetz für die Bundesrepublik Deutschland (Grundgesetz) (GG) (Basic Law), 23 May 1949, BGBl I, art 1 (Ger).
Transformative Constitutionalism 177
rights’.12 Moreover, Article 1 of the Constitution states that South Africa is a democratic State founded on the values of human dignity, equality, human rights and freedoms, and non-racialism and non-sexism.13 In both Germany and South Africa, then, the constitutional values are defined in contrast to the past. Israeli constitutional laws have not been codified into one formal document. Instead, in accordance with its 1950 ‘Harari Decision’,14 the Israeli Parliament has adopted 11 Basic Laws, each designed and intended to be a chapter in a future Israeli Constitution. These laws have established the democratic institutions of the State of Israel on the principle of separation of powers and make scarce reference to fundamental values. Two Basic Laws adopted in 1992, Basic Law: Human Dignity and Freedom and Basic Law: Freedom of Occupation, were the first to constitutionally protect fundamental rights, with frequent mention of values and fundamental principles in both. For example, section 1 of Basic Law: Human Dignity and Freedom reads as follows: Fundamental human rights in Israel are founded upon recognition of the value of the human being, the sanctity of human life, and the principle that all persons are free; these rights shall be upheld in the spirit of the principles set forth in the Declaration of the Establishment of the State of Israel.
Section 1A continues, [t]he purpose of this Basic Law is to protect human dignity and liberty, in order to establish in a Basic Law the values of the State of Israel as a Jewish and democratic state.
Additionally, section 8, which sets the terms for the restriction of the rights protected under the Law, provides that one of the necessary conditions for the constitutionality of a law limiting those rights is that it corresponds with ‘the values of the State of Israel’. Although this Israeli constitutional text is aspirational and value-oriented, it can hardly be considered transformative. The 1992 Basic Laws were enacted after continuous failure in Parliament to enact a full and comprehensive constitutional Bill of Rights. In order to overcome the fierce opposition of the religious political parties to such a document, Israeli liberal lawmakers made substantial concessions, in terms of both the number of constitutional rights protected and the scope of that protection.15 Another major concession was that whereas the original draft of the 1992 Basic Law defined Israel as a ‘democratic’ State, the final draft had the added feature of ‘Jewish’. This definition of Israel as a ‘Jewish and democratic State’ reverts back to the core definition of the State in the 1948 Declaration of Independence.16 So, although aimed at bolstering human rights protection, the Basic Law does so within the confines of an institutional structure taken from the past rather than adopting a radically transformed structure. If the 1992 Basic Laws were not intended to be transformative, what can account for the many invocations of principles, aspirations, and values in the texts? Given the deep rifts that exist in Israeli society across ethnic and religious lines, the use of abstract and open-textured ‘values’ serves to cover up the profound dissent over their supposedly shared contents. The effect of values-talk is the substantial delegation of power to the Sout African Constitution 1996, Preamble. ibid s 1. 14 DK 5 (1950) 1743. 15 G Sapir, The Constitutional Revolution: Past, Present and Future (Tel Aviv, Haifa University Press, 2010) (in Hebrew). 16 n 2. 12 13
178 Moshe Cohen-Eliya Supreme Court to define the meaning of those values. The Court has generally used this interpretive authority to emphasise the democratic values of Israel and narrowly construct the Jewish values. For example, when it nullified the Jewish Agency’s policy not to lease lands it administers to non-Jews, the Court was promoting Israel’s democratic values and narrowly construing its Jewish values: The Jewish State having been established, it treats all its citizens equally. The State of Israel is a Jewish state in which various minorities, including the Arab minority, live. Each of the minorities living in Israel enjoys complete equality of rights. It is true, members of the Jewish nation were granted a special key to enter (see the Law of Return, 5710-1950), but once a person has lawfully entered the home, he enjoys equal rights with all other household members.17
The Supreme Court has always been the flag-bearer of liberal, democratic, and humanistic values. But the enactment of the 1992 Basic Laws served to legitimise and accelerate the move towards humanism. Soon after their adoption, Aharon Barak published several articles in which he labelled this development nothing less than a ‘Constitutional Revolution’:18 first, because the Court could now strike down any statute violating human rights (despite the fact that this power was not explicitly granted in the Basic Law) and, second, because the Israeli Parliament had expressed the centrality of core humanistic values, such as human dignity, in Israeli constitutional law. The Supreme Court relied on its new authority to overrule some of its earlier decisions and to better protect fundamental rights in cases in which the rights infringement was not framed in a statute.19 Barak’s choice of the term ‘Revolution’ is no accident. Indeed, it is a powerful rhetorical tool that generated much public attention and was intended to accelerate the transformation of Israeli society into a more tolerant, liberal, and humanistic one. The need to strengthen democracy and humanism in Israel is a constant theme in Barak’s public talks, writings, and judicial decisions.20 His repeated reference to the 1992 Basic Laws as a revolution and to ‘values’ has created a critical mass of rhetoric aimed at making policymakers, legislators, judges, teachers, the media, and individuals internalise the fact that something dramatic has happened: that Israel has taken an important step toward a humanistic democracy that respects human rights. This rhetoric is catchy, exciting, and powerful – an inevitable trigger of media attention and a tool for reshaping public discourse. As noted in Part II, value-based judicial review does not necessarily imply transformative constitutionalism. Indeed, the invocation of ‘traditional’ values can serve precisely the opposite end, for it can operate as a mechanism for preserving the existing political structure. The US solution for the problem of misusing values was to generally refrain from referring in its constitutional text and jurisprudence to values and to suffice in protecting rights, which are relatively more concrete. What may account for this is the traditional American suspicion of the judiciary, particularly the idea of judges’ interpret17 HCJ 6698/95 Ka’adan v Israel Land Administration 54(1) PD 258 [2000] (in Hebrew). English translation available at: www.elyon1.court.gov.il/files_eng/95/980/066/a14/95066980.a14.pdf. 18 A Barak, ‘A Constitutional Revolution: Israel’s Basic Laws’ (1993) 4 Constitutional Forum 83; A Barak, ‘The Constitutional Revolution: Protected Basic Rights’ (1992) 1 Law and Government 9 (in Hebrew). This view is not shared by all. See, eg R Gavison, ‘The Constitutional Revolution: A Reality or a Self-Fulfilling Prophecy’ (1997) 28 Mishpatim 21, 32, 73, 95–97 (in Hebrew). 19 See, eg HCJ 2316/95 Ganimat v State of Israel 49(4) PD 589 [1995] (in Hebrew). 20 Gavison (n 18) 95–97.
Transformative Constitutionalism 179
ing vague ‘values’ subjectively and thereby undermining democracy.21 It is thus not sur prising that the transformative strand in American constitutionalism has met with very limited success.22 Consequently, basing judicial review on values is a gamble. In several Israeli cases, the reference to values in fact served to justify a severe limitation of human rights. For example, in the 1960 Yardor case, the Israeli Supreme Court relied on Israel’s ‘Jewish’ values as they appear in the Declaration of Independence to disqualify an Arab political party that opposed the definition of ‘Jewish’ as an essential characteristic of the State.23 Similarly, in the 2006 Adalah v Minister of Interior ruling, Cheshin J, writing for the narrow majority, justified a near absolute ban on family reunification in Israel between Israeli Arabs and their Palestinian spouses, on the basis of a ‘value-based balance’ that gives priority to the value of life (and security) over the rights to equality and family life.24 B. Positive Rights While in theory, the notion of a transformative Constitution does not exclude transitioning from an active State to a neutral one, the post-war Germany case is illustrative of how such a transition cannot succeed without the State and courts taking some active measures to uproot totalitarian and anti-democratic attitudes in society. It is quite clear from judicial legal doctrines in transformative societies that the courts assume such an exceptionally active role and their readiness to impose active duties on the State to ensure meaningful realization of the new constitutional values. In Germany, the Federal Constitutional Court has ruled that the Wertrangordnung (hierarchal value order) on which the Constitution is constructed is an objective one, with one of its central aspects being the requirement that the State take action to meaningfully realise constitutional values and rights (positive rights).25 The Court interpreted some seemingly negative rights to also include certain positive obligations. For example, it ruled that the constitutional freedom of choice of professional training imposes a duty on the State to subsidise professional training.26 Likewise, in its landmark abortions decisions, the 21 This approach is manifested in the persistence of legal originalism in American constitutional law. Legal originalism is the classical formulation of the importance of text for the legitimacy of judicial review. See generally, RH Bork, The Tempting of America (New York, The Free Press, 1990) 143–53; A Scalia, ‘Originalism: The Lesser Evil’ (1989) 57 University of Cincinatti Law Review 849, 863–64. 22 Clearly, there are some elements of transformative constitutionalism in the US, especially in the judgments of Marshall and Brennan JJ. The Warren Court can be regarded as the most transformative in the history of the US; see Lessig (n 1) 215 (referring to such a view expressed by many). Yet the persistence of originalism and the process-based theory of judicial review in US constitutional law are indicative of the significant controversy that this concept still stirs in the US relative to Germany and South Africa. 23 EA 1/65 Yardor v Chairman of the Central Elections Committee to the Sixth Knesset 19(3) PD 365 [1965] (in Hebrew). 24 HCJ 7052/03 Adalah v Minister of Interior Affairs 61(2) PD 202 [2006] para 107 of Cheshin J’s opinion (in Hebrew). English translation available at: www.elyon1.court.gov.il/files_eng/03/520/070/a47/03070520.a47. htm. (PSS (proportionality in the strict sense) is a value based test). For the same see also para 75 of Court President Barak’s opinion. 25 D Currie, ‘Positive and Negative Constitutional Rights’ (1986) 53 University of Chicago Law Review 864, 867–73. On the notion of positive and negative rights in American constitutional law, see S Bandes, ‘The Negative Constitution: A Critique’ (1990) 88 Michigan Law Review 2271; CR Sunstein, ‘Lochner’s Legacy’ (1987) 87 Columbia Law Review 873, 888–90. 26 33 BVerfGE 303 (1972) (Ger).
180 Moshe Cohen-Eliya Court ruled that the State has an active duty to protect a foetus’ constitutional right either by criminalizing abortions (1975)27 or by ensuring the economic wellbeing of women who seek not to have an abortion (1993).28 The South African Supreme Court often expresses pride in its protection of social and economic rights in its constitutional jurisprudence. In a series of judgments, the Court was willing to impose unprecedented obligations on the State to ensure social rights.29 For South Africans, the transformative mission is not only about antidiscrimination laws and equal political rights, but also about altering the economic institutions through redistribution and positive social rights. The rejection of the concept of positive rights in the American Constitution is in fact a rejection of transformative constitutionalism.30 In stark contrast to the German abortion decisions, the US Supreme Court has ruled that, although the Government is precluded from banning abortions under the Fourteenth Amendment due process rights31 it is under no obligation to subsidise abortions for poor women.32 Progressive liberal justices who have expressed certain transformative aspirations for the US Constitution, such as Brennan and Marshall, have made some efforts at developing the positive dimensions of constitutional rights, particularly in the framework of the equal protection clause.33 However, this has always been a controversial strand in US constitutionalism and far more limited in scope than, for example, in German and South African constitutional jurisprudence.34 Regarding this feature, Israeli constitutionalism can be situated between the two poles, with a gradual shift over the past two decades from strictly American in orientation towards the German or South African model. The unwritten Bill of Rights that the Israeli Supreme Court had developed from the early years of the State’s establishment was purely negative.35 This approach began to alter with the enactment of the 1992 Basic Laws. To the provision in section 2 of the Basic Law: Human Dignity and Freedom, ‘There shall be no violation of the life, body or dignity of any person as such’, section 4 adds an important layer of protection: ‘All persons are entitled to protection of their life,
39 BVerfGE 1 (1975) (n 8). 88 BVerfGE 203 (1993) (n 8). 29 See, eg Soobramoney v Minister of Healt Kwazulu Natal 1998 (1) SA 765 (CC) para 8 (S Afr) (the huge economic disparities ‘already existed when the Constitution was adopted and a commitment to address them, and to transform our society into one in which there will be human dignity, freedom and equality lies at the heart of our new constitutional order. For as long as these conditions continue to exist that aspiration will have a hollow ring’); Government of the Republic of South Africa v Grootboom 2001 (1) SA 46 (CC) (S Afr). But see DM Davis, ‘Socio-Economic Rights: The Promise and Limitation: The South Africa Experience’ in D BarakErez and A Gross (eds), Exploring Social Rights: Between Theory and Practice (Oxford, Hart Publishing, 2007) 193 (arguing that the South African Government adheres to capitalism and the Court’s protection of social rights is limited). For social rights and transformation on the global level, see K Rittich, ‘Social Rights and Social Policy: Transformations on the International Landscape’, ibid Exploring Social Rights 107. 30 Bowers v DeVito 686 F 2d 616, 618 (1982); DeShaney v Winnebago County 489 US 189, 195–96 (1989); Jackson v City of Joliet 465 US 1049 (1984). 31 Roe v Wade 410 US 113 (1973). Note that the constitutional rights in the American Constitution are drafted in the Constitution as negative and not positive rights. 32 Harris v MacRea 448 US 297, 316 (1980) (‘although government may not place obstacles in the path of a woman’s exercise of her freedom of choice, it need not remove those not of its own creation’). 33 C Jolls, ‘Accommodation Mandates’ (2000) 53 Stanford Law Review 223; FI Michelman, Brennan and Democracy (Princeton, Princeton University Press, 2005) 130. 34 Michelman ibid. 35 D Barak-Erez, ‘From an Unwritten to a Written Constitution: The Israeli Challenge in American Perspective’ (1995) 26 Columbia Human Rights Law Review 309. 27 28
Transformative Constitutionalism 181
body and dignity’.36 These articles served to induce the Israeli Supreme Court to construct a more positive conception of rights. This transformation has been particularly prominent in the jurisprudence of the Israeli Supreme Court in the context of the right to equality. In the Women’s Network case, the Supreme Court ruled that a positive requirement for affirmative action is innate to the constitutional right to equality.37 In Botzer, the Court declared that the requirement to accommodate a disabled pupil in an elementary school is ‘a right and not a grace’.38 Similarly, in Miller, in compelling the air force to open up its pilot training courses to women, the Court referred to Aharon Barak’s statement, ‘The protection of human rights costs money, and a society that respects human rights must be prepared to bear the financial burden’.39 The concept of positive rights was also introduced in the context of such classic polit ical rights as freedom of expression. In its 2006 Majority Camp decision,40 the Supreme Court ruled that the police cannot make permission to demonstrate contingent on the willingness of the demonstrators to bear the weighty cost of police protection. Writing the unanimous judgment, Barak J elaborated on the dual dimension of rights: The duty of the state to protect the constitutional right to freedom of expression and demonstration has two facets. First, the state is under obligation not to infringe a person’s freedom of expression and demonstration, by banning her ability to exercise her right. This is the ‘negative’ aspect of the right (status negatives). Second, the state is under obligation to defend freedom of expression and demonstration. This is the ‘positive’ aspect of the right (status positives).41
Moreover, despite the lack of explicit protection of social rights in the Basic Law: Human Dignity and Freedom, the Court has been willing to protect some core aspects of such rights through its interpretation of the constitutional right to human dignity. Such was the case in its 2005 decision in Commitment to Peace,42 pertaining to the constitutionality of drastic budget cuts in State income allowances. The Court drew on the Basic Law to find a State duty to maintain a system guaranteeing a safety net for the socially disadvantaged that protects them from existential shortages. The State, the Court ruled, must guarantee all people sufficient sustenance, a place of residence where they can realise privacy and family life, adequate sanitation conditions, and health services that allow them access to modern medicine.43 This noble rhetoric notwithstanding, however, the Court rejected the petition on the ground that the petitioners had failed to prove that the new State allowances were insufficient to meet their minimum basic needs. In later cases, pertaining to the right to education, the Court was more willing to impose
36 This provision was added to the Basic Law during the course of the deliberations in the parliamentary constitutional committee prior to the enactment of the Basic Laws in 1992, despite the objections of the liberal Minister of Justice, Dan Meridor, who feared that it would excessively burden the Government and police. See DK 124 (1992) 1531. 37 HCJ 2671/98 Israel Women’s Network v Minister of Labour and Social Affairs 52(3) PD 630 [1998] (in Hebrew). 38 HCJ 7081/93 Botzer v Municipal Council of Maccabim-Reut 50(1) PD 19 [1996] (in Hebrew). 39 HCJ 4541/94 Miller v Minister of Defence 49(4) PD 94, 113 [1994] (in Hebrew). 40 HCJ 2557/05 Majority Camp v Israel Police 62(1) PD 200 [2006] (in Hebrew). 41 ibid para 14 of Court President Barak’s opinion (my translation, MCE). 42 HCJ 366/03 Commitment to Peace and Social Justice Association v Minister of Finance 60(3) PD 464 [2005] (in Hebrew). 43 ibid para 14 of Court President Barak’s opinion.
182 Moshe Cohen-Eliya substantial costs on the State. In Marciano,44 for example, the Court invoked the right to education, which is not explicitly set forth in any of the Basic Laws, to require the State to allocate sufficient funds (estimated at no less than $40 million by the Education Ministry) for implementing an amendment to the Special Education Law 5748-1988. The Supreme Court has also shown a similar tendency in cases more loosely related to constitutional rights. For example, it ruled in Wasser that the existing system of protecting school pupils in the town of Sderot (a town located near the Gaza border constantly targeted by Hamas rockets) is inadequate.45 In Academic Center of Law and Business, the Court struck down legislation authorizing the privatization of prisons in Israel, forcing the Government to pay millions of dollars in compensation to the company that had won the franchise rights.46 A research study conducted by a libertarian think tank in Israel estimated the costs of implementing the Court’s rulings in recent years at no less than $4 billion.47 C. Drittwirkung Transformationists are well aware of the fact that in order to successfully effect a fundamental change, not only the State, but also individuals, must be bound by the new set of constitutional values. Accordingly, the post-war German Federal Constitutional Court ruled that constitutional values indirectly impact private law, thereby setting the Drittwirkung – ‘impact on third parties’ – doctrine.48 In its landmark Lüth case, the Court found that the call to boycott a film made by a director who had been a Nazi sympathiser does not constitute a violation of section 826 of the German Civil Code (BGB), which provides, [a] person who intentionally inflicts harm on another in a manner which offends against good morals is obliged to make compensation to the other for the harm.
The Court’s interpretation of the term ‘good morals’ took into account the boycotter’s constitutional right to free speech, which is closely related to the paramount constitutional value of human dignity.49 Likewise, South African private law has been dramatically transformed by the mandates of the new constitutional values.50 44 HCJFH 247/04 Minister of Finance v Marciano (10 May 2004), Nevo Legal Database (by subscription) (in Hebrew). 45 HCJ 8397/06 Wasser v Minister of Defence 62(2) PD 198 [2007] (in Hebrew). 46 HCJ 2605/05 Academic Center of Law and Business v Minister of Finance (19 November 2009), Nevo Legal Database (by subscription) (in Hebrew). English translation available at: www.elyon1.court.gov.il/files_ eng/05/050/026/n39/05026050.n39.htm. 47 B Arad, The Cost of Judicial Activism (Jerusalem, Jerusalem Institute for Market Studies, 2010) www. jims-israel.org/pdf/ppcourts.pdf. 48 S Gardbaum, ‘The “Horizontal Effect” of Constitutional Rights’ (2003) 102 Michigan Law Review 387, 403. Gardbaum himself offers a different interpretation of the American model (‘although private actors are not bound by individual constitutional rights in the United States, they are indirectly subject to (and may be adversely affected by) them because such rights govern the laws that private actors invoke and rely on against each other . . . The extent of this reach of individual rights into the private sphere defies the standard understanding of the United States as creating a rigid public-private distinction in constitutional law’) ibid 458. 49 7 BVerfGE 198 (1958) (Ger). 50 F Du Toit, ‘The “Constitutionalisation” of South African Private Law: Teaching Challenges and Solutions’ (2007) 5 Journal of Commonwealth Law and Legal Education 111, 111 (‘The South African Constitution has a determining effect on South African private law in that all private law legislation is testable against the fundamental rights enshrined in the Constitution’s Bill of Rights, and South African courts have an essential development
Transformative Constitutionalism 183
By contrast, most of the provisions of the US Constitution are not transformative in this respect, as they have been consistently interpreted to apply only to State action (ie to State and local government and not to private entities).51 The Thirteenth Amendment is the only provision to impose restrictions on private action, in prohibiting all types of slavery. This constitutional clause can truly be deemed transformative in the context of US history. In Israel, the German Drittwirkung model has flourished. This notwithstanding section 11 of the Basic Law: Human Dignity and Freedom which provides, [a]ll governmental authorities are bound to respect the rights under this Basic Law.
In an influential article published in 1993, Aharon Barak argued that, given the import ance of constitutional rights and the purpose of the Basic Law, this provision should not be interpreted as narrowing the scope of rights. Rather, it should be construed as emphasizing the application of the constitutional norms vis-à-vis the State, while not excluding its application to private action.52 Thus, Barak opted for the Drittwirkung approach: indirect application of constitutional rights through interpretation of open-textured private law provisions, such as good faith and public policy.53 The Israeli Supreme Court had addressed this issue only in dicta, in its 1991 Kestenbaum decision.54 This case related to the right of a deceased’s family to use Gregorian calendar dates and non-Hebrew letter notations on his tombstone in a Jewish cemetery. Barak declared in dicta, [i]t should seem clear and obvious that the basic principles of the legal system in general, and basic human rights in particular, are not limited to public law. Indeed, the true question is not whether the basic principles of public law apply within private law. The answer to this question is clear and simple: Yes. The true questions concern how these basic principles of public law flow into the confines of private law, and what are the conduits through which these principles are transferred to the behavior of an individual in his interactions with others.55
These signals from Barak’s judgments and academic writings have had a profound impact on lower court jurisprudence. In Na’amne,56 the Jerusalem Magistrate Court ruled that the enactment of the Basic Law: Human Dignity and Freedom had impacted on private law. The Court found that a racially-based denial of entry into a privatelyowned water park had violated tort law (‘a breach of a statutory duty’),57 whose interpretation was strongly influenced by constitutional rights. The Haifa Magistrate Court ruled in Matzkin that individuals cannot be required to disclose their sexual orientation task in respect of the common law principles of South African private law. In consequence, South African private law has undergone significant change under the influence of constitutional imperatives, which change will continue in future. I call this the “constitutionalisation” of South African private law’). 51 Lugar v Edmondson Oil Co 457 US 922 (1982); but see Shelley v Kraemer 334 US 1 (1948). 52 Interestingly, a similar provision in the Canadian Charter of Rights and Freedoms was interpreted to exclude the application of the constitutional rights on private action. See RWDSU v Dolphin Delivery Ltd [1986] 2 SCR 573 (Can). 53 A Barak, ‘Constitutional Human Rights and Private Law’ in D Friedmann and D Barak-Erez (eds), Human Rights in Private Law (Oxford, Hart Publishing, 2001) 13. 54 CA 294/91 Jerusalem Community Burial Society v Kestenbaum 46(2) PD 464 [1992] (in Hebrew). 55 ibid para 20. English translation available at: www.utexas.edu/law/academics/centers/transnational/ work_new/israeli/case.php?id=1391. 56 CC (Jm) 11258/93 Na’amne v Kibbutz Kalia (1 September 1996), Nevo Legal Database (by subscription) (in Hebrew). 57 Tort Ordinance (New Version), 5728-1968, s 63.
184 Moshe Cohen-Eliya when seeking to rent an apartment or office space.58 Finally, in its Samshiyan decision,59 the Tel-Aviv Magistrate Court held that intentional discrimination against a person with a disability regarding entry into a ‘singles club’ constituted a breach of good faith in contractual negotiations. The rise of the Drittwirkung doctrine in Israel is not surprising given the limited amount of antidiscrimination laws passed in the early 1990s. In the US, prior to the civil rights legislation in the 1960s, the Supreme Court ruled in Shelly v Kramer60 that, despite the fact that the Constitution applies only to State actions, an individual cannot expect the Court, which is an organ of the State, to issue judgments that violated the Constitution. Yet what is striking in Israel is that even after the enactment of antidiscrimination laws, courts have continued to apply the constitutional norms to individuals as a means of bypassing specific exemptions in those laws. For example, in Applebaum,61 the National Labour Court ruled that despite the explicit exemption from the Equal Opportunity in Employment Law’s requirement not to discriminate granted to entities employing less than six people, the constitutional duty not to discriminate (on the basis of pregnancy) applied indirectly to them. The Court used section 39 of the Contracts Law (General Part), 5733-1973, which lays down the general principle of good faith in contracts, to apply indirectly the constitutional norm of equality to private employers. D. Proportionality Transformation is essentially a political action, for it involves a restructuring of society and its institutions by way of an active judiciary that leaves, to use Barak’s term, no ‘black holes’.62 Since transformative Constitutions tend to be total in their scope, there is a structural need for a doctrine that allows the courts to be involved in policymaking yet, at the same time, maintain the ‘objectivity’ of the law.63 The three-pronged doctrine of proportionality, assists the courts in this respect: there must be a rational connection between the means and ends (rationality); the least restrictive means must be used when limiting constitutional rights (necessity); and a proper balance must be attained between the benefit from the restriction of the given right and the harm to that right (proportionality in its strict sense).64 These prongs seem objective and outcome-neutral,65 yet at the 58 CC (Ha) 4583/96 Association for the Protection of the Rights of Individuals v Matzkin [1996] (unpublished). 59 CC (TA) 15/97 Shamsiyan v Rosemary Garden Rest (12 January 1999), Nevo Legal Database (by subscription) (in Hebrew) (on file with author). 60 Shelley (n 51). 61 LA 1353/02 Applebaum v Holtzman 39 PDA 495 [2003] (in Hebrew). 62 A Barak, A Judge in a Democracy (Princeton, Princeton University Press, 2006) 194, 298 (arguing that there are no black holes where there is judicial review). 63 See M Cohen-Eliya and I Porat, ‘The Hidden Foreign Law Debate in Heller: Proportionality Approach in American Constitutional Law’ (2010) 46 San Diego Law Review 367, 405–09. 64 CA 6821/93 United Mizrahi Bank Ltd v Migdal Cooperative Village 49(4) PD 221, 404 [1995] (in Hebrew). 65 DM Beatty, The Ultimate Rule of Law (Oxford, Oxford University Press, 2004) 166–69 (praising such proportionality virtues as value-neutrality, objectivity, and focus on facts rather on values); J Rivers, ‘Proportionality and Variable Intensity of Review’ (2006) 65 CLJ 174 (highlighting the structured character of proportionality); A Barak, Proportionality: Constitutional Rights and Their Limitations (Cambridge, Cambridge University Press, 2012) 460–65 (pointing to the transparent, structured, and dialogical character of proportionality); P Craig, Administrative Law, 6th edn (Oxford, Oxford University Press, 2008) 637 (‘The proportionality test provides a structured form of inquiry. The three-part inquiry focuses the attention of both the agency being reviewed, and the court undertaking the review’); A Barak, ‘Proportionality and Principled
Transformative Constitutionalism 185
same time allow courts the broad latitude necessary for executing the ambitious task of social transformation. The American approach of categorization and line-drawing, that hostile to judicial balancing in constitutional law, is by no means sufficient to facilitate such transformation. It therefore should come as no surprise that the proportionality doctrine gained prominence in post-war Germany. In the 1950s, the Federal Constitutional Court resorted extensively to values and natural law theories. This was followed in the late 1950s and early 1960s by the emergence of the more sophisticated version of proportionality, with its three-pronged test. Ever since, the Court has systematically applied the three para meters in almost all of its judgments.66 While many other constitutional jurisdictions subsequently adopted the principle of proportionality as their central analytical framework, Germany has been unique in its clear emphasis of the third requirement: balancing.67 Commonwealth constitutional jurisdictions such as Canada and New Zealand, and, recently, Great Britain, place greater emphasis on necessity, the second prong.68 For Germans, balancing, an essentially political action, seems a legitimate and necessary element of the transformative Constitution.69 The Israeli Supreme Court, for its part, has followed the ‘German’ approach and has developed a relatively sophisticated form of proportionality, insisting on maintaining the internal logic of the structure of the analysis and almost religiously applying each and every subtest. This version of the doctrine places greater emphasis on balancing than does Commonwealth proportionality. In some of its major decisions, the Israeli Supreme Court has struck down laws and policies solely because they failed to properly balance between the competing rights and interests.70 In this respect, Israeli constitutionalism is closer to the German model, allowing the Court greater leeway in facilitating the political mission of transformation. Interestingly, the term ‘proportional’ has gradually infiltrated the public discourse. Politicians, journalists, academics, and others make frequent use of the term to justify or criticise government actions encumbering human rights.71 The infiltration of this Balancing’ (2010) 4 Law and Ethics of Human Rights 1 (from the abstract: ‘The advantages of proportionality stricto sensu with its three levels of abstraction are several. It stresses the need to always look for a justification of a limit on human rights; it structures the mind of the balancer; it is transparent; it creates a proper dialogue between the political branches and the judiciary, and it adds to the objectivity of judicial discretion’). 66 Cohen-Eliya and Porat (n 63) 388–96. 67 D Grimm, ‘Proportionality in Canadian and German Constitutional Law Jurisprudence’ (2007) 57 University of Toronto Law Journal 383, 393 (‘The most striking difference between [Canada and Germany] is the high relevance of the third step of the proportionality test in Germany and its more residual function in Canada’). 68 See, eg G Davidov, ‘Separating Minimal Impairment from Balancing: A Comment on R v Sharpe (BCCA)’ (2000) 5 Review of Constitutional Studies 195 (criticizing the Canadian constitutional jurisprudence that focuses on the necessity test). 69 Generally speaking, the European constitutional courts assume for themselves a political role in implementing constitutional values, therefore their application of a balancing, or a proportionality, approach enjoys a greater legitimacy. See AS Sweet, Governing with Judges: Constitutional Politics in Europe (Oxford, Oxford University Press, 2000) 31, 40 (arguing that European constitutional courts do not conduct judicial review in the antimajoritarian sense but rather as political organs). 70 A Barak, ‘Proportional Effect: The Israeli Experience’ (2007) 57 University of Toronto Law Journal 369 (describing the cases in which the Israeli Supreme Court applied proportionality in the strict sense). For an example see HCJ 2056/04 Beit Sourik Village Council v Government of Israel 58(5) PD 807 [2004] paras 59–85 of Court President Barak’s opinion (in Hebrew). English translation available at: www.elyon1.court.gov.il/ files_eng/04/560/020/A28/04020560.a28.htm. 71 The Israeli version of the word ‘proportional’ (Midati in Hebrew) is a judicially-invented word, that did not exist in the Hebrew language prior to the Supreme Court Rulings setting this doctrine.
186 Moshe Cohen-Eliya somewhat technical-sounding legal doctrine into the public discourse serves the transformatist mission, for it helps citizens better internalise the meaning of the trans formation: any action restricting human rights must be restrained and conscious of the new set of humanistic values. IV. AN ALTERNATIVE: CAROLENE PRODUCTS ABROAD?
The judiciary faces a far more demanding task when acting to realise transformative constitutionalism as opposed to a preservative one. As Lessig has noted, ‘Constitutional moments die, and when they do, the institutions charged with enforcing their commands, such as courts, face increasing political resistance’.72 The Israeli Supreme Court has faced political backlash for its judicial activism. It has been rigorously attacked for allegedly promoting a leftist agenda under the guise of legal doctrine, disregarding the constitutional text and circumventing the framers’ intentions, and serving the interests of the hegemonic elites that, it is claimed, no longer dominate the political process and have redirected their efforts to the Court.73 The upshot of this assault has been a barrage of proposed Bills in the Israeli Parliament in recent years seeking to limit the Supreme Court’s power in this respect.74 Yet more worrying has been the dramatic decline in public trust in the Supreme Court,75 which cannot be explained away as Israelis’ general tendency to be suspicious of the authorities. Preservative constitutionalism, however, would not be a better alternative for Israel. Given the tensions and contradictions inherent in Israel’s self-definition as both Jewish and democratic, such an approach would lead to nothing less than a social and legal stalemate in Israel. Due to the heterogeneity and dynamics of Israeli society, the very worst scenario would be a freezing of the current balance of powers by entrenching ‘values’ and thereby blocking any possibility of political change. A more promising alternative to either transformative or preservative constitutionalism is, I believe, democracy-reinforcing constitutionalism. A few months before Barak J retired from the Supreme Court, Grunis J, newly appointed to the Court, built on footnote four of Carolene Products76 and Ely’s process-based theory77 of judicial review to offer an alternative to Barak’s notion of a transformative judicial mission approach. Ruling on a petition filed by the Movement for Quality Government that challenged the exemption of ultra-Orthodox Jews from serving in the Israeli army,78 Grunis (following Ely) equated the role of the judge to that of a referee in sports,79 whose decisions should be procedurally fair and not outcome-oriented. The referee’s main objective is to enforce Lessig (n 1) 214. For an overview of the critique, see Sapir (n 15). 74 S Nave, ‘Law Proposals as Strategic Tool in the Discourse on the Independence of the Judiciary’ (2011) 33 Tel-Aviv University Law Review 631 (in Hebrew) (documenting the dramatic rise in the number of proposed Bills aimed at limiting the power of the judiciary. The majority of these Bills were not passed and rather served as a political ‘threat’ and a ‘signaling’ mechanism vis-à-vis the Israeli Supreme Court). 75 A Ratner, The Culture of Law, the Judiciary and the Legal System in the Mirror of the Israeli Society: 2000–09 (Jerusalem, Hebrew University, 2010) 6–10 (executive summary) (in Hebrew). 76 United States v Carolene Products Co 304 US 144 (1938). 77 JH Ely, Democracy and Distrust: A Theory of Judicial Review (Cambridge, Harvard University Press, 1980). 78 HCJ 6427/02 Movement for Quality Government v Knesset 61(1) PD 619 [2005] (in Hebrew). 79 ibid para 7 of Grunis J’s opinion, referring to Ely, above (n 77) 72–171. 72 73
Transformative Constitutionalism 187
the rules of the game. Likewise, the Supreme Court should intervene only where the democratic process has failed (for example, where political participation has been impaired or discrete and insular minorities are discriminated against). Grunis argued that since, in this instance, the majority of the legislature had decided to grant the ultraOrthodox minority this privilege, the democratic process was not flawed and the Court should not intervene.80 Barak regarded Grunis’s approach as a threat to his value-based transformative legacy and attacked Ely’s process-based model. Essentially, he contended in his judgment in the same case, that Ely’s theory is internally flawed and is based on a narrow and inappropriate conception of democracy; it is not accepted in the United States, where it emerged, and in other countries that have judicial review; and it is foreign to our constitutional structure and contradicts our constitutional judgments.81
Barak was correct in suggesting that Ely had failed to construct a purely procedural theory of judicial review. As many critics have justifiably argued, there is a certain level of indeterminacy as to what counts as a failure in the democratic process.82 Furthermore, Ely’s democracy-facilitating theory is an interpretive theory for the US Constitution. It would be quite difficult to show either that the drafters of the 1992 Israeli Basic Laws had this theory in mind or that they had grounded it in the text they formulated. All that said, the democracy-facilitating approach offers promise for constitutionalism in deeply-divided societies such as Israel, for it opens up channels through which citizens can decide for themselves whether a transformation is necessary and, if so, which path that transformation should take. This approach is built on the appropriate division of powers, where non-accountable and non-elected judges deal with facts and procedures while democratically-elected representatives deal with policies, values, and aspirations. It guarantees that the decision on value-transformation will be inclusive, by rigorously protecting political and participatory rights and by ensuring that marginalised ‘discrete and insular minorities’ are not stigmatised or demonised. As recently demonstrated by Eskridge, the democracy-reinforcing approach can help maintain social stability in pluralistic societies by lowering the stakes of politics.83 Prior to the enactment of the Basic Laws in 1992, the Israeli Supreme Court had struck down several statutes violating equality in general elections.84 These rulings did not ibid para 8. Court President Barak writing for the Court, ibid para 75 (my translation, MCE). Grunis was far from being consistent in his Carolene Products approach in subsequent cases. For example, in the Adalah v Minister of Interior case, pertaining to the near-absolute ban on family reunification, he was unwilling to rigorously scrutinize the law that discriminated against the Arab minority in Israel – a clear example of ‘a discrete and insular minority’ – and voted to deny the petition. See Grunis J’s opinion in Adalah v Minister of Interior (n 24). 82 Even one of the supporters of Ely’s process-based theory, MJ Klarman, ‘The Puzzling Resistance to Political Process Theory’ (1991) 77 Virginia Law Review 747 admits that ‘Ely’s critics have, in my view, been devastatingly successful in demonstrating that Ely’s “procedural” theory of prejudice is riven with substantive judgments’ (p 785). See also, LH Tribe, ‘The Puzzling Persistence of Process-Based Constitutional Theory’ (1980) 89 Yale Law Journal 1063, 1069–71; RA Posner, ‘Democracy and Distrust Revisited’ (1991) 77 Virginia Law Review 641; Bork (n 20) 194–99. 83 WJ Eskridge, ‘Pluralism and Distrust: How Courts Can Support Democracy by Lowering the Stakes of Politics’ (2005) 114 Yale Law Journal 1279; WJ Eskridge, ‘Foreword: The Marriage Cases – Reversing the Burden of Inertia in a Pluralist Constitutional Democracy’ (2009) 97 California Law Review 1785. 84 See, eg HCJ 98/69 Bergman v Minister of Finance 23(1) PD 693 [1969] (in Hebrew); HCJ 142/89 Laor Movement v Speaker of the Knesset 44(3) PD 529 [1990] (in Hebrew). 80 81
188 Moshe Cohen-Eliya undermine the public trust in the Court, but, rather, bolstered its legitimacy as a neutral arbiter between competing political camps. Today, things have changed considerably. All too often, the justices are seen as eagerly jumping into the muddy political fray as participants in the cultural war being waged. Transformative constitutionalism is thus a double-edged sword for courts: on the one hand, it assigns them a central role in bringing to fruition the ‘new’ values, yet on the other hand, it runs the risk of a loss of public trust in the judiciary and, consequently, diminished chances of the transformation succeeding. For every political action, there is a reaction. There is often public backlash against judicially transformative constitutionalism, which might ultimately lead to its failure and a rocking of the social boat. As a society with wide chasms and a constant security threat, Israel cannot afford for this to happen. My suggestion here is that judges should resist the temptation to get involved in the exciting project of transformation and instead divert their energies to facilitating the democratic channels by which the people decide whether and how to transform their society.
13 Proportionality in Israel and Beyond: Four Aspects MARGIT COHN*
I. INTRODUCTION
T
HE SPREAD OF proportionality (Verhältnismäßigkeit) across the Western world has, in the words of Stone-Sweet and Mathews, attained a ‘viral quality’,1 but it might be better depicted as yet another brand name that has gained global reputation. Originating in Prussian case law, the post-World War II German jurisprudence granted it a central place in its constitutional law.2 As one commentator of German law submits, the principle is now applied ‘as an independent and perhaps the most important and extensive umbrella ground for examining the validity of administrative actions’.3 This, then, is one of the few public law doctrines that have attained virtual universality.4 The principle is just as prominent in Israel: it is now part of its public law, * The author thanks the Israel Science Foundation for its (Grant No 160/09) funding that supported the research, and the editors and Neta Shapira and Avi Vaknin for their comments and meticulous work on the draft chapter. 1 AS Sweet and J Mathews, ‘Proportionality Balancing and Global Constitutionalism’ (2008) 47 Columbia Journal of Transnational Law 72, 74, 112. On the principle see, eg DM Beatty, The Ultimate Rule of Law (Oxford, Oxford University Press, 2004) ch 5; V Jackson, ‘Being Proportional about Proportionality’ (2004) 21 Constitutional Commentary 803; Sweet and Mathews, ibid; M Cohen-Eliya and I Porat, ‘American Balancing and German Proportionality: The Historical Origins’ (2010) 8 International Journal of Constitutional Law 263, 263–64; M Cohen-Eliya and I Porat, ‘Proportionality and the Culture of Justification’ (2011) 59 American Journal of Comparative Law 463; A Barak, Proportionality: Constitutional Rights and Their Limitations (Srigim-Li’on, Nevo, 2010) (in Hebrew) (hereinafter: Barak, Proportionality); G Webber, ‘Proportionality Balancing and the Cult of Constitutional Rights Scholarship’ (2010) 23 Canadian Journal of Law and Jurisprudence 179; A Barak, Proportionality: Constitutional Rights and Their Limitations (Cambridge, Cambridge University Press, 2012); a debate over proportionality, balancing and their impact on human rights, published in 7 and 8 International Journal of Constitutional Law (2009 and 2010); and two dedicated issues (4(1) and 4(2)) of Law and Ethics of Human Rights (2010). For a general analysis of the principle in the administrative law of European Member States, see J Schwarze, European Administrative Law (London, Office for Official Publications of the European Communities, 2006) 680–702. 2 For recent overviews in English of the ‘German genealogy’ see D Grimm, ‘Proportionality in Canadian and German Constitutional Law’ 57 University of Toronto Law Journal 383, 384–87; Sweet and Mathews (n 1) 97–111; I Zamir, ‘Israeli Administrative Law as Compared with German Administrative Law’ (1994) 2 Law and Government 109 (in Hebrew); Cohen-Eliya and Porat, ‘American Balancing’ (n 1) 271–75; Barak, Proportionality (n 1) 228–31. 3 MP Singh, German Administrative Law in Common Law Perspective, 2nd edn (Berlin, Springer, 2001) 160. 4 For a comparative analysis that reaches to South Korea, South America and Eastern Europe, in addition to most Western states, see Barak, Proportionality (n 1) 232–61; see also n 1 above. For proportionality in international law see, eg Y Shany and A Cohen, ‘A Development of Modest Proportions’ (2007) 5 Journal of
190 Proportionality in Israel and Beyond: Four Aspects has been analyzed in a multitude of judicial decisions and is treated extensively in the literature. Professor Aharon Barak’s recently published tome on proportionality is likely to become, now that it has been translated, the definitive work on the doctrine in its constitutional context, and enrich the already abundant literature.5 No analysis of constitutional law can marginalise the role of proportionality as a decision-making tool. Rather than dwell in detail on the substantive effect of the adoption of the principle in Israel, address the endemic weakness and cyclical nature of the doctrine, or normatively argue for or against the rise of proportionality and its potential for the enhancement of judicial participation in the public decision-making process, I have chosen to treat other conceptual and comparative issues. This chapter addresses four aspects of the concept of proportionality. My emphasis on the Israeli experience is supplemented by references to the status of the proportionality doctrine in Germany, two European courts (the European Court of Justice (ECJ) and the European Court of Human Rights (ECtHR)), Canada, Zimbabwe and the UK, all of which have adopted proportionality as a central review mechanism. First, I offer a distinction between proportionality as an overarching principle, typical to European and continental law, and proportionality as a tool for the application of constitutional protection of human rights, a specific context usually applied in common law systems. The second distinction offered in this chapter focuses on the difference between a concept, or doctrine, and a formula. Most systems that have adopted the proportionality principle/concept have also adopted the German three-pronged formula, although the transplant of the first does not necessarily require the transplant of the second. In this context, I touch upon the strategic elements of the choice of formulae. In part IV, I analyze the migration of the three-pronged formula, the most prevalent proportionality formula, and identify and provisionally explain different forms of discrete, undisclosed adoptions of foreign formulae. Finally, I touch upon the nature of proportionality as a reasoning device, challenging the universal depiction of proportionality analysis as one that requires balancing, and offering a version of the principle as consequence-oriented. II. PROPORTIONALITY: AN OVERARCHING PRINCIPLE OR A RIGHTS-PROTECTION DEVICE?
Proportionality can be an overarching principle, which operates as a guiding light of a legal system. More concretely, an overarching principle may operate as a rule of interpretation and a value of best practice. It is thereby an element of rule-making and International Criminal Justice 310, 311–12; Y Shany, The Principle of Proportionality under International Law, Policy Paper (Jerusalem, The Israel Democracy Institute, 2009) (in Hebrew); see also Court President Barak’s analysis in HCJ 2056/04 Beit Sourik Village Council v Government of Israel 58(5) PD 807, 836–38 [2004] (in Hebrew); Barak, Proportionality (n 1) 250–56. It would seem that only the United States remains reticent to adopt the doctrine as a distinct form of reasoning, but see studies of the United States which focus on proportionality decision-making; eg ET Sullivan and RS Frase, Proportionality Principles in American Law: Controlling Excessive Government Actions (Oxford, Oxford University Press, 2009). 5 The research in the Israeli context includes Zamir (n 2); D Dorner, ‘Proportionality’ in A Barak and C Berenson (eds), Berenson Book, vol 2, 2nd edn (Tel-Aviv, Nevo, 2000) 281 (in Hebrew); G Sapir, ‘Proportionality and Compromise’ in S Almog, D Beinisch and Y Rotem (eds), Dalia Dorner Book (SrigimLi’on, Nevo, 2009) 397 (in Hebrew); Barak, Proportionality (n 1); Cohen-Eliya and Porat, ‘American Balancing’ (n 1); Cohen-Eliya and Porat, ‘Proportionality and Justification’ (n 1); D Barak-Erez, Administrative Law (TelAviv, The Israel Bar Publishing House, 2010) 771–92 (in Hebrew).
Margit Cohn 191
rule-review in all fields of law. Beyond German law, European Union law offers a strong example. Now that the Treaty of Lisbon has been ratified, the principle of proportionality is proclaimed one of the central principles of European law, established not only in a new Article on these principles but also in a designated Protocol treating the principles of conferral, subsidiarity, and proportionality.6 The review powers granted by the treaties to the ECJ7 thus require judicial consideration of whether the necessity condition was met. But the Court’s recognition of the principle of proportionality has not been based solely on textual reasoning. The principle was recognised early on as one of the ‘general principles of law deriving from the rule of law’, which directs the Union and its Member States in the making and application of European law – long before the entry into force of the Treaty of Lisbon.8 Alternatively, proportionality may operate as a tool in rights-protection issues. The most typical expression can be found in many Bills of Rights, including the European Convention on Human Rights (ECHR). The Convention recognises and protects a series of human rights and creates an institutional structure for their protection. Several 6 Under Art 3b(1) of the Treaty of Lisbon amending the Treaty on European Union (Maastricht Treaty) and the EC Treaty, ‘The use of Union competences is governed by the principles of subsidiarity and proportionality’; Art 3b(4) proclaims that ‘the content and form of Union action shall not exceed what is necessary to achieve the objectives of the Treaties’; and the Protocol on the application of the principles of subsidiarity and proportionality, requiring national and European legislatures to consider both principles in the process of legislation. Proportionality was central even prior to the Lisbon Treaty. The European Union Founding Treaties (the EC Treaty (Treaty of Rome) and the Treaty on European Union (Maastricht Treaty)), were interpreted as subjecting Member States and Community institutions to the rules of necessity and proportionality. Provisions in the Treaty of Rome include Art 42 (empowering the Council to adopt measures in the field of social security ‘as are necessary to provide freedom of movement for workers’); Art 120 (empowering Member States to take, in case of a sudden crisis in the balance of payments, take precautionary necessary protective measures that ‘must cause the least possible disturbance in the functioning of the common market and must not be wider in scope than is strictly necessary to remedy the sudden difficulties’). Under Art 5, Member States were required to ‘take all general or particular measures which are appropriate for ensuring the carrying out of the obligations arising out of this Treaty or resulting from the acts of the institutions of the Community’, but this version is not considered as embodying the principle. Since 1993, upon entry into force of the Maastricht amendments, Art 5 of the Treaty of Rome requires that ‘Any action by the Community shall not go beyond what is necessary to achieve the objectives of this Treaty’. See, eg FG Jacobs, ‘Recent Developments in the Principle of Proportionality in European Community Law’ in E Ellis (ed), The Principle of Proportionality in the Laws of Europe (Oxford, Hart Publishing, 1999) 1, 2 (at the early stage of the Treaties, this could not have been more than a perfunctory acknowledgement of existing case law). Other provisions in the Maastricht Treaty include Art 13(3) (empowering the Council to ‘take the decisions necessary for defining and implementing the common foreign and security policy’); and Art 14(6) (empowering Member States to take ‘the necessary measures’ when joint actions are not deemed suitable to changing conditions and in the absence of a Council decision on the matter, ‘in cases of imperative needs’). 7 These include the power to review the legality of acts made by Community institutions, to decide upon one Member State’s challenge against another Member State, to decide upon Member State’s challenge to inaction of Community institutions, and to give preliminary rulings regarding the interpretation of all types of legal measures made by the Community (usually required when such a measure is challenged or relied upon in a domestic court). 8 See, eg Case 8/55 Federation Charbonniere de Belgique v High Authority of the European Coal and Steel Community [1954–56] ECR 292, 299 (‘in accordance with a generally-accepted rule of law . . . [a measure] must be in proportion to the scale of that action’); Case 4/73 Nold v Commission of the European Communities [1974] ECJ 491, 512–13; Case 107/63 Toepfer v Commission of the European Economic Community [1965] ECR 405, 427; Case 11/70 Internationale Handelsgesellschaft MBH v Einfuhr- und Vorratsstelle fur Getreide und Futtermittel [1970] ECR 1125. For literature, see Schwarze (n 1) 708–26; Gráinne de Búrca, ‘The Principle of Proportionality and its Application in EC Law’ (1993) 13 Year Book of European Law 105; T Tridimas, ‘Proportionality in Community Law: Searching for the Appropriate Standard of Scrutiny’ in Ellis (n 6) 65; Tridimas, The General Principles of EU Law, 2nd edn (Oxford, Oxford University Press, 2006) chs 3–5; P Craig and G De Búrca, EU Law: Text, Cases, and Materials, 3rd edn (Oxford, Oxford University Press, 2002) 178– 229; Sweet and Mathews (n 1) 139–45.
192 Proportionality in Israel and Beyond: Four Aspects Convention Articles require, among other conditions allowing interference with a right, that the interference be ‘necessary in a democratic society’ and be applied for the promotion of defined interests.9 Other provisions in the Convention subject state action to general necessity conditions, sometimes under a stricter requirement such as ‘absolutely necessary’.10 The ECtHR has interpreted all such provisions as embodying a proportionality requirement.11 Other limitation clauses, such as section 1 of the Canadian Charter of Rights and Freedoms, offer similar requirements,12 but I focus here on Israel. Israel’s development of the proportionality principle is a clear example of the extension of the principle beyond the protection of human rights. This is so, despite the fact that its formal origin is found in constitutional-type statutes designed to apply in human rights contexts only. In 1992, the Israeli legislature appended limitation clauses to two Basic Laws that defined and protected certain human rights. The Basic Law: Human Dignity and Liberty and the Basic Law: Freedom of Occupation contain a Limitation Clause, not dissimilar from such clauses in other human rights legal documents, that was interpreted as including a proportionality test: The rights under this Basic Law may only be violated by a law that befits the values of the State of Israel, is intended for a proper purpose, and to an extent that is not excessive, or under a law as stated by virtue of an express authorization therein.13
Some have argued that the Israeli Supreme Court had used proportionality reasoning in pre-1992 decisions.14 Some of these decisions relied on legislative arrangements which included a weighting requirement; in these cases, a simple balancing between the 9 The first is Art 8(2) (interference with the right to respect for private and family life must, inter alia, be ‘necessary in a democratic society in the interests of national security, public safety or the economic well‑being of the country, for the prevention of disorder or crime, for the protection of health or morals, or for the protection of the rights and freedoms of others’). Others follow a similar formula, matching the necessity requirement with different defined interests: Art 9(2) (freedom of thought, conscience and religion); Art 10(2) (freedom of expression); Art 11(2) (freedom of assembly and association); Protocol IV, Art 2(3) (freedom of movement). 10 Art 2 (right to life: use of force which is absolutely necessary in a list of defined cases); Art 5(1)(b) (right to liberty and security: arrest or detention of suspect when reasonably considered necessary to prevent his committing an offence or fleeing); Art 6(1) (right to a fair trial: exclusion of the press or the public, inter alia, to the extent strictly necessary in special circumstances where publicity would prejudice the interests of just ice); Art 15(1) (derogation in times of emergency may be made, inter alia, to the extent strictly required by the exigencies of the situation); Protocol I, Art 1 (protection of property: states not impaired from enforcing laws necessary to control the use of property in accordance with the general interest or to secure the payment of taxes or other contributions or penalties). 11 For early decisions relying on textual reasoning, see Handyside v United Kingdom (1979) 1 EHRR 737, 754 (‘necessary’ implied the existence of ‘a pressing social need’; it meant that every restriction imposed was ‘proportionate to the legitimate aim pursued’); Sunday Times v United Kingdom (1979) 2 EHRR 245, 275–78 (following Handyside). Later recognition of proportionality as a general principle can be found in Velaers J’s concurring opinion in Conka v Belgium (2002) 11 EHRR 555; it is applied in a wide range of cases by the court in its case law (particularly in its case law on para 2 of Arts 8–11, and Art 14) and may be regarded as part of the Art 5 requirement that persons are only to be deprived of their liberty ‘in accordance with the procedure prescribed by law’). See also Erkalo v Netherlands (1998) 28 EHRR 509, para 2 of Levits J’s dissenting opinion. 12 Under s 1, ‘The Canadian Charter of Rights and Freedoms guarantees the rights and freedoms set out in it subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society’. See, eg PW Hogg, Constitutional Law of Canada, vol II (Toronto, Thomson/Carswell, 2007) 38-18–38-44.2. 13 Basic Law: Human Dignity and Liberty, s 8. An identical provision appears in the Basic Law: Freedom of Occupation, s 4. 14 For the most prominent argument in this vein see Z Segal, ‘The Ground of Disproportionality in Administrative Law’ (1991) 39 The Lawyer 507 (in Hebrew); Z Segal, ‘Security Measures, Administrative Proportionality and Judicial Review’ 1 Law and Government 477 (in Hebrew).
Margit Cohn 193
impairment and the benefit was applied, rather than a developed proportionality doctrine.15 A few cases were later interpreted as relying on a proportionality-type reasoning, but at best, references to proportionality were discursive and no independent ground was established.16 The term ‘proportionality’ (midatiut) was first used in a decision delivered in March 1992, just one day after the Basic Law: Freedom of Occupation was passed by the Knesset (Israeli Parliament).17 By birth, then, the ground was inalienably connected to the protection of human rights, and has since been applied and reapplied in dozens of decisions.18 However, the Supreme Court has also applied the ground in several cases in which the human rights aspect was rather weak, and at best could be found in applicants’ arguments of alleged unfair treatment. I will discuss two examples. Ben-Atiyah (1995) was concerned with the Minister of Education’s policy to designate, by lot, one of the compulsory national matriculation subjects, thereby allowing high school pupils to receive their internal mark rather than sit the examination. This exemption was denied to pupils in several institutions, in which a high level of unethical practices in examinations was found in previous years. Applicants argued that the denial of exemption was unreasonable and discriminatory.19 The Supreme Court President Barak, with the other Justices on the Bench, found the denial of the exemption unreasonable, but used the decision to further develop the ground of proportionality. The analysis was not based on a human rights rationale; rather, it dwelt on the place of this ground in many legal systems and on its occurrence in earlier Israeli jurisprudence. The challenged administrative decision was quashed; disproportionately harmful to the pupils, it did not even pass the first proportionality subtest: no rational connection between the purpose (ensuring ethical examination practice) and the means (denying the exemption) could be found. The decision fared no better under the other two subtests.20 Similarly, and still in the field of education, Moustaki (2003) was concerned with Civil Service policy regarding accreditation of foreign academic institutions.21 Once accredited, degrees granted by these institutions were considered equivalent to Israeli degrees 15 See, eg HCJ 344/89 HSA International Commerce v Minister of Industry and Trade 44(1) PD 456 [1990] (in Hebrew); HCJ 2994/90 Poraz v Government of Israel 44(3) PD 317 [1990] (in Hebrew). 16 See, eg HCJ 935/89 Ganor v Attorney General 44(2) PD 485 [1990] (in Hebrew) (unreasonableness of the decision of GLA not to prosecute banks’ managers for the bank share scandal should be assessed by weighing the public interest in prosecution against the harm to the managers). 17 HCJ 5667/91 Jabbarin v IDF Commander of Judea and Samaria 46(1) PD 858 [1992] (in Hebrew). This decision, and a second, delivered about a year later, were both concerned with demolition of houses in the occupied territories; Basic Law: Dignity and Liberty was not cited either in HCJ 5510/92 Turkeman v Minister of Defence 48(1) PD 217 [1993] (in Hebrew). The third decision which used the term ‘proportionality’ cited the Basic Law: Freedom of Occupation, HCJ 987/94 Euronet Golden Lines (1992) Ltd v Minister of Communications 48(5) PD 412 [1994] (in Hebrew). 18 Some of the best-known decisions which have been translated into English and are available at: elyon1. court.gov.il/verdictssearch/englishverdictssearch.aspx are: CA 6821/93 United Mizrahi Bank Ltd v Migdal Cooperative Village 49 (4) PD 221 [1995] (proportionality in the context of the right to property); HCJ 5016/96 Horev v Minister of Transportation 51(4) PD 1 [1997] (freedom of movement and freedom of religion); HCJ 769/02 Public Committee Against Torture in Israel v Government of Israel 62(1) PD 507 [2006] (challenging the 2003 Citizenship and Entry into Israel (Temporary Provision) Law; Beit Sourik (n 4) (the ‘security fence’ decision); HCJ 2605/05 Academic Center of Law and Business v Minister of Finance (19 November 2009) (BOT privatization of prison facility). 19 HCJ 3477/95 Ben-Atiyah v Minister of Education, Culture and Sports 49(5) PD 1 [1996] (in Hebrew). 20 ibid 9–16. 21 HCJ 3379/03 Moustaki v Attorney General 58(3) PD 865 [2004] (in Hebrew).
194 Proportionality in Israel and Beyond: Four Aspects for salary purposes. Following criminal investigations of fraudulent grant of degrees by a Latvian university, the accreditation of this institution was cancelled. Hundreds of civil servant graduates, who were thereby denied earlier benefits, challenged the decision. The proportionality of the cancellation of the accreditation was but one of the flaws found by the Court. Applying the already well-established three subtests of proportionality, Justice Or found that even though the decision met the first subtest – the purpose to eradicate reliance on fraudulently achieved academic degrees was served by the blanket revocation of the accreditation – the act of revocation failed the second and third subtests, as it affected graduates who had not acted fraudulently.22 Here, too, the Court felt no need to embed its decision in a human rights context. Rather, proportionality was considered yet another important requirement that must be met in order to pass the test of good administration. In both cases, however, the Court was concerned with unequal allocation of public assets. Proportionality Israeli-style, then, has not achieved a continental-like status: it is not considered an overarching principle such as the rule of law, legality and democracy. However, its application in cases that have no direct link with a text that protects human rights may be a step towards such a development.
III. PROPORTIONALITY FORMULAE AND THEIR STRATEGIC VALUE
Analyses of legal doctrines often neglect a basic distinction between the doctrines themselves, which tend to be open-ended and malleable, and formulae that flesh out these concepts. A formula is distinct from the doctrine it serves; it operates as a template for its application.23 By the term ‘formula’, I refer to a criterion or a set of criteria, designed as an objective test, which is sufficiently established to be recognised as the basis for regular assessment of the applicability of a legal concept or doctrine. Formulae achieve their status through subsequent usage and consistent reliance. They need not be universally applied, but to retain the status of a formula, they should be generally recognised as the accepted mechanism for the application of the concept they embody. The Baker v Carr six criteria used for a decision over the justiciability of an application is one example,24 as are the levels of scrutiny tests formed in Carolene Products.25 It would seem that the proportionality doctrine is most afflicted by the insufficient attention to this distinction; most, if not all, analyses of proportionality assume without question that proportionality is identical to its ‘elements’, or its well-known stages or subtests.26 No student of constitutional law outside the United States can remain ignorant of these subtests. In its current form, the formula created by German courts comprises three subtests or limbs. First, the measure must be suitable for the achievement of the aim pursued. Secondly, no other milder means could have been employed to achieve that aim (a ‘necessity’ test). Finally, under a proportionality stricto sensu test, a type of cost ibid 907–11. For the distinction and further analysis, see M Cohn, ‘Form, Formula and Constitutional Ethos: The Political Question/Justiciability Doctrine in Three Common Law Systems’ (2011) 59 American Journal of Comparative Law 675. 24 Baker v Carr 369 US 186 (1962). 25 United States v Carolene Products Co 304 US 144 (1938). 26 Eg Sources cited at n 5 above. 22 23
Margit Cohn 195
benefit analysis is required; for the measure to be upheld, the benefit at large must outweigh the injury to the implicated individual.27 Israel is one of many systems that have followed this formula when it began to apply the clause ‘to an extent that is not excessive’.28 First adopting the three-pronged formula in United Mizrahi Bank, the court openly acknowledged the sources of the formula and cited German and Canadian sources.29 Since then, the Israeli Supreme Court applies almost verbatim the Canadian three-pronged test of proportionality introduced in Oakes, which in itself is very similar to the German formula. In its Hebrew version, [t]he first subtest is that the objective must be related to the means . . . The means used by the administrative body must rationally lead to the realization of the objective. This is the ‘appropriate means’ or ‘rational means’ test. According to the second subtest, the means used by the administrative body must injure the individual to the least extent possible. In the spectrum of means which can be used to achieve the objective, the least injurious means must be used. This is the ‘least injurious means’ test. The third test requires that the damage caused to the individual by the means used by the administrative body in order to achieve its objectives must be of proper proportion to the gain brought about by that means. That is the ‘proportionate means’ test (or proportionality ‘in the narrow sense’).30
Since its adoption, the three-pronged formula has become almost holy, an incantation that appears in most, if not all, discussions of the proportionality principle. It is found in countless decisions, in all studies of the principle, and is vigorously taught to students of all disciplines concerned with judicial review. However, this type of formula is not essential to the development of a robust proportionality doctrine. Proportionality is also a tenet of European Union law, and a second prototype of formula can be found here. The two main European fora have developed distinct proportionality formulae. The ECJ tends to apply two formulae, both similar to the three-pronged German test. The first follows the German contours,31 the second is a two-pronged test, sometimes described as one that emulates the first German subtest and combines the latter two. Under this formula, the Court considers whether the challenged measure was appropriate and necessary (the latter requires the Court to assess whether the measure did not affect the interest at stake beyond what was necessary to achieve the objective).32 27 See, eg H Maurer, Allgemaines Verwaltungsrecht (Munich, Beck, 2006) 250–51; Schwarze (n 1) 685–92; Singh (n 3) 165–67. 28 Similar tests exist inter alia in Canada, Zimbabwe, the United Kingdom (when applying the ECHR – but see part IV below), Japan, Chile and Colombia. 29 United Mizrahi Bank (n 18) 343, 345, 412, 436–73, also emphasizing that the Knesset was aware of the German and Canadian models and was especially influenced by the latter (ibid 299, 376). See also Ben-Atiyah (n 19) (citing Canadian sources). 30 Beit Sourik (n 4) para 41. See also cases at n 18; R v Oakes [1986] 1 SCR 103 (Can). 31 See, eg Case C-331/88 R v Minister for Agriculture, Fisheries and Food, ex p Fedesa [1990] ECR I-4023, para 13 (‘By virtue of [the principle of proportionality], the lawfulness of the prohibition of an economic activity is subject to the condition that the prohibitory measures are appropriate and necessary in order to achieve the objectives legitimately pursued by the legislation in question; when there is a choice between several appropriate measures recourse must be had to the least onerous, and the disadvantages caused must not be disproportionate to the aims pursued’). 32 See, eg Case 279-280/84 Rau v European Economic Community [1988] 2 CMLR 704, 750–51; Case C-426/93 Federal Republic of Germany v Council of the European Union [1995] ECR I-3723, para 42; Case C-84/94 United Kingdom and Ireland v Council of the European Union [1996] ECR I-5755, para 57; Case C-434/02 Arnold André v Landrat des Kreises Herford [2004] ECR I-11825, para 45. The very first decision that recognized and applied the principle of proportionality may be cited as applying this test, but could also be read as applying no more than a simple balancing test, with the German subtests relegated to the background (Internationale Handelsgesellschaft
196 Proportionality in Israel and Beyond: Four Aspects Rather than adopt the three or two-pronged approach, the ECtHR has employed a simple test. Often, the Court assesses the proportionality of a challenged measure by balancing the means employed and the aims pursued, supplemented by a reference to the margin of appreciation granted to Member States, which colours the final decision.33 As a further development, the Court has introduced two additional conditions aimed at protecting rights, the first requiring ‘a pressing social need’ (or a ‘compelling social interest’) to justify interference with a protected right,34 the second subjecting challenged action to ‘close scrutiny’, under which ‘very weighty reasons’ are needed to justify interference with a central human right.35 Why have these formulae emerged, and how can the amazing popularity of the German formula be explained? Does the complexity of the test make a difference? In a separate article, I have argued that the choice and development of a formula may be driven by judicial appreciation of the values of systematization and clarification, or simple human tendency to classify, but the use of formulae can also carry strategic bene fits.36 Reliance on a formula carries a ‘distancing effect’. Judges who rely on established seemingly objective tests find shelter from accusations of politicization, which are likely to arise whenever the case before them is sensitive or politically laden, as cases involving human rights may be. When they apply a formula to reject an application, they protect themselves from accusations of weakness, subjection to political pressure, or partisan support of the respondents’ substantive position. Likewise, when judges accept an application, seemingly objective tests operate as a shield from anticipated criticism which may originate from the challenged body or other actors in the social and political spheres that are unsatisfied by the outcome. In both cases, strong levels of criticism may directly endanger the social legitimacy of the judiciary. Thus, the more politically-laden the issue, the greater the utility of formalization. And the more detailed and elaborate the formula, the greater its distancing potential: a complex, highly structured formula requires judges to go through several hoops, which serve as a signal of judicial commitment to the application of complex, professional, even clinical processes that have nothing much to do with the political. How, then, to explain the near-universalization of the three-pronged test (excluding the ECtHR)? Its dominance may be credited to the formula’s superior character as a usable tool for application of an open-ended concept; however, under the strategic model suggested above, its complexity may be one of its main attractions. (n 8) 1134–35, paras 12, 14, 16). On the proportionality tests, see Schwarze (n 1) 854–60; Craig and De Búrca (n 8) 372–73; Tridimas, ‘Proportionality’ (n 8) 139 (different emphases on the prominence of the two and three-pronged tests); Sweet and Mathews (n 1) 145–48; M Fordham and T de la Mare, ‘Identifying Principles of Proportionality’ in J Jowell and J Cooper (eds), Understanding Human Rights Principles (Oxford, Hart Publishing, 2001) 27, 37–49. 33 See, eg Sunday Times (n 11) 277–78; for examples of recent cases see, eg Dogru v France (2009) 49 EHRR 8; A v United Kingdom (2009) 49 EHRR 29 (detention of suspected terrorists). In both latter cases, the Court did not overturn the decisions of the domestic courts. But see de Búrca (n 8) 113. 34 Handyside (n 11) 754 (‘pressing social need’ was implied by the term ‘necessary’ in the Convention); Sunday Times (n 11) 277–78 (‘close scrutiny’ of arguments regarding the necessity of the interference, to be justified only under a ‘pressing social need’); Lingens v Austria (1986) 8 EHRR 407 (conviction for defamation in press article found in breach of freedom of speech (Art 10); pressing social need not found); Abdulaziz v United Kingdom (1985) 7 EHRR 471, 501 (only ‘very weighty reasons’ could justify different treatment of sexes under Immigration Rules). 35 See the cases ibid, which were mainly concerned with freedom of speech and sexual discrimination. See also generally, Fordham and Thomas de la Mare (n 32) 49–60. 36 Cohn (n 23).
Margit Cohn 197
IV. THE MIGRATION OF THE PROPORTIONALITY THREE-PRONGED FORMULA
In this part I am concerned with the migration of the proportionality formula, as distinct from the adoption of proportionality as an operable doctrine. When the Israeli Supreme Court adopted the three-pronged formula, first in United Mizrhahi Bank, the court openly acknowledged the sources of the formula and cited German and Canadian sources.37 Since then, the court has remained faithful to the Canadian tests, although Oakes is no longer the central precedent; for example, it cannot be found at all in the Beit Sourik decision, arguably the Israeli Supreme Court’s most famous and detailed decision on proportionality, which recognised the force of the proportionality doctrine in both domestic and international law.38 But the transplant of a formula may be more discrete. The rest of this part is dedicated to counter examples elsewhere. Migration is not always openly celebrated; doctrines and formulae may be imported without open declaration. Alternatively, even when foreign sources are cited to support a new doctrine or formula, it may be in fact transformed without such transformation being acknowledged. Discrete transplants are especially interesting; they seem to be the outcome of the tension between universalization and exceptionalism, between the readiness, even need, to learn from other systems, the possible risk to domestic integrity inherent in this type of learning, and special loyalty to systems belonging to the legal family of the importing systems. The following analysis traces examples of such types of transplants. I begin with Canada. In Oakes, the 1986 decision in which the proportionality formula was adopted,39 the Canadian Supreme Court cited Big M Drug Mart, a Canadian precedent, as the source for this formula, but this precedent only provides dicta regarding the possible development of unreasonableness in the context of the Canadian Charter to ‘a form of proportionality test’.40 Some ECtHR jurisprudence is cited in Oakes, but only as part of a comparative survey of the protection of presumption of innocence, the right impaired in this case, and the ‘rational connection’ test relies on US cases regarding the presumption of innocence.41 Neither the German formula nor any reference to German jurisprudence can be found. The decision also ignores the proportionality tests developed by the European fora, discussed above.42 How can this be explained? In their biography of Dickson J, who delivered the decision, Professors Sharpe and Roach disclose that Dickson was aided by two law clerks, one of them, Joel Bakan, ‘fresh from graduate studies at Oxford’, who ‘immersed himself in the European human rights jurisprudence under the principle of proportionality’.43 Whether this is the missing link to the German formula remains a matter of speculation. What is clear, in any case, is that the proportionality formula emerged in Canada as a fresh, if not original test, attesting an emphasis on the force of domestic law. The link was recognised later on. See n 29 above. Beit Sourik (n 4) 836–40. 39 Oakes (n 30) para 70. 40 R v Big M Drug Mart Ltd [1985] 1 SCR 295, para 140 (Can). 41 Oakes (n 30) paras 50–55. 42 On this absence see Sweet and Mathews (n 1) 117. 43 RJ Sharpe and K Roach, Brian Dickson: A Judge’s Journey (Toronto, University of Toronto Press, 2003) 334. I thank Martin Friedland for directing me to this source. 37 38
198 Proportionality in Israel and Beyond: Four Aspects The Zimbabwe Supreme Court, in comparison, openly relied on Oakes in Nyambirai,44 but not on German law. Being part of the Commonwealth world, reliance on Canadian jurisprudence may have seemed more legitimate than foraging into the continent. Further, the slight transformation of the formula may be the result of domestic ‘pride’, an evidence of the ability of domestic judges to improve upon foreign solutions. The British adoption of the proportionality formula is even more complex, and exemplifies the tension between the force of the domestic, the influence of common law systems and the British commitment to European law. The rest of this part is dedicated to a review of this most tangled chapter in the tale of the adoption of a proportionality formula.45 To begin with, arguments that in the British Isles, ‘Verhältnismäßigkeit’ should be translated as ‘unreasonableness’ are unconvincing.46 As courts later admitted, proportionality was to be considered a more powerful ground for review than unreasonableness, despite potential overlap and possible similarity in outcomes.47 The traditional British reticence to adopt proportionality drew directly from the Diceyan heritage and its impact on the notion of judicial review. However, being central to European law, proportionality could not be ignored in areas in which European law was formally incorporated into domestic British law.48 On joining the European Community, the British Parliament enacted the European Communities Act 1972, under which European law would be ‘recognised and available in law’. This implied, in a rather convoluted way, that domestic law that squarely contradicted binding European rules could no longer be applied.49 Thus, applicants could rely on proportionality when the challenged measures were promulgated by European Community institutions and directly binding, or when the application included a challenge to the legality of a domestic measure due to its incompatibility with a binding European measure. Domestic courts applied European Community law when correctly relied upon; this was done, however, in conjunction with domestic grounds of review, including unreasonableness.50 Courts thus continued to emphasise the force of domestic law. Nyambirai v National Social Security Authority [1996] 1 LRC 64, 1995 (9) BCLR 1221 (ZS). The analysis of British law is based on M Cohn, ‘Legal Transplant Chronicles: The Evolution of Unreasonableness and Proportionality Review of the Administration in the United Kingdom’ (2010) 58 American Journal of Comparative Law 583, 616–22. 46 Such an argument, seemingly offered by Singh (n 3), may have been advanced only to explain the nature of the ground of proportionality to the (British) uninitiated reader. This chapter is based on Cohn ibid. 47 See, eg R v Chief Constable of Sussex, ex p International Trader’s Ferry [1997] 2 CMLR 164, 182 (CA) per Kennedy LJ (‘Proportionality requires the Court to judge the necessity of the action taken as well as whether it was within the range of courses of action that could reasonably be followed. Proportionality can therefore be a more exacting test in some circumstances’); R (Daly) v Secretary of State for the Home Department [2001] UKHL 26, [2001] 2 AC 532, 547 per Lord Steyn (‘the [proportionality] criteria are more precise and more sophisticated than the traditional grounds of review . . . there is an overlap between the traditional grounds of review and the approach of proportionality. Most cases would be decided in the same way whichever approach is adopted. But the intensity of review is somewhat greater under the proportionality approach’); R (Begum) v Headteacher and Governors of Denbigh High School [2006] UKHL 15, [2007] 1 AC 100 [30]. But see Barak on this issue, Barak, Proportionality (n 1) 455–65. 48 The United Kingdom follows a dualist approach: international conventions become part of domestic law only by further legislative action. On the dualist and monist approaches, see I Bronwlie, Principles of Public International Law (Oxford, Clarendon Press, 2008) 31–33. 49 European Communities Act 1972, s 2(1). This subjection was finally formally recognized in the courts only in 1992, in R v Secretary of State for Transport, ex p Factortame (No 2) [1990] UKHL 13, [1991] 1 AC 603. 50 See, eg R v Chief Constable of Sussex, ex p International Trader’s Ferry [1998] UKHL 40, [1999] 1 CMLR 1320 (proportionality test applied with regard to arguments pertaining to the European Treaty, in conjunction with unreasonableness). On the interface between unreasonableness and proportionality see below. 44 45
Margit Cohn 199
The European Convention of Human Rights51 presented a different challenge. The United Kingdom was one of the first states that signed and ratified the Convention in the early 1950s, but incorporating legislation was introduced only in 1998. Before the entry into force of the Human Rights Act (HRA) in October 2000, the Convention had no direct effect on British domestic law. The United Kingdom subjected itself to the review of the ECtHR in 1966; being answerable to the Court on the European level, applications to the Strasbourg Court were sometimes successful. While even failed applications served to raise political and public consciousness of the United Kingdom’s subjection to the European law of human rights, successful applications had, of course, a stronger impact.52 None of this, however, required direct application. Thus in GCHQ, Lord Diplock opined that further development of administrative law might lead to the adoption of new grounds of review, including proportionality, yet he found the existing grounds to be sufficient at the time.53 A gradual recognition of the value of the proportionality doctrine was subsequently evident only in judicial dicta and academic literature.54 The field was transformed upon the entry into force of the Human Rights Act. Novel review mechanisms were introduced, to be applied when breaches of the ECHR (more precisely, the parts of the Convention that were formally adopted by the Act) were found.55 Under section 2 of the Act, courts were required to take into account the jurisprudence of the ECtHR and other bodies operating in the context of the Convention – a statutory duty to at least consider European case law, which was more than an invitation to incorporate it when suitable. Thus, proportionality could no longer be resisted. The doctrine was indeed adopted by post-HRA courts, but the transplantation of the proportionality formula was far from straightforward. The multi-player nature of this transplantation process has been marginalised by many commentators, who focus on the impact of European law on the law of the United Kingdom. Yet, at least in rhetoric, courts were not ready to succumb easily to European law. In the case of proportionality, rather than drawing on ECtHR jurisprudence, the British courts were initially inspired by the Supreme Courts of Canada, Zimbabwe, and South Africa – all of them members or past members of the Commonwealth. The tale begins with De Freitas (1998), the first case in which British courts adopted the ground of proportionality.56 The case originated in Antigua and Barbuda, an independent state that retains the British Crown as its Head of State; therefore, the final judicial instance remains the Privy Council, which decided upon the constitutionality of a statute legislated in this territory. The Constitution of Antigua and Barbuda limits the freedom of 51 Convention for the Protection of Human Rights and Fundamental Freedoms (European Convention on Human Rights, as amended) (ECHR). 52 For a survey of Strasbourg decisions regarding the United Kingdom and the impact of the Convention on domestic law in 1997, see AW Bradley and KD Ewing, Constitutional and Administrative Law, 12th edn (Harlow, Pearson Longman, 1997) 470–76. 53 Council of Civil Service Unions v Minister for the Civil Service [1983] UKHL 6, [1985] 1 AC 374, 410. 54 For an analysis of the pre-Human Rights Act judicial treatment of the proportionality doctrine see Cohn (n 45) 617–19. 55 S 3 empowers courts to interpret domestic statutes in ways that contradict, to a certain extent, their textual meaning, to achieve compatibility with the Convention. Section 4 introduces the remedy of declaration of incompatibility, which replaces the continental – and American – remedy of statute invalidation. Both are prime examples of transformation, if not distortion, of the essence of constitutional judicial review recognized in the ECHR. I do not elaborate on this point, since this chapter is concerned with proportionality as one of the grounds of substantive review. 56 De Freitas v Permanent Secretary of Ministry of Agriculture, Fisheries, Lands and Housing [1998] UKPC 30, [1999] 1 AC 69 (adopting the test adopted in Nyambirai (n 44)).
200 Proportionality in Israel and Beyond: Four Aspects speech of civil servants only as ‘reasonably required for the proper performance of their functions’, a limitation similar to those found in other Commonwealth constitutions. Defining the ground of proportionality implied in the reasonable requirement clause, the Privy Council cited decisions from South Africa and Zimbabwe, mentioning the Canadian source, and relied on the three-pronged test adopted by the latter.57 The next step was Daly, the first case in which proportionality was discussed in the context of the Human Rights Act. Despite the reliance of the applicant on the ECHR and the requirement in the Human Rights Act to ‘take into account’ European jurisprudence, the House of Lords cited de Freitas as precedent for their adoption of proportionality.58 Noting that ‘the contours of the principle of proportionality are familiar’ – a blatant overstatement of the state of British law at the time – Lord Steyn proceeded to adopt the three-pronged test adopted in de Freitas.59 This emphasis on the status of the principle of proportionality as an existing part of domestic law came at the price of a full recognition of its European origin in the context at hand. One could justify de Freitas’ reliance on Commonwealth formulae, since, after all, the decision required the interpretation and application of a Commonwealth constitution. However, the absence of reliance on European doctrine in Daly is not selfexplanatory. In applying the Human Rights Act, judicial reference to the ECtHR ground of proportionality, and, possibly, the formula it tends to use, would have been appropriate, at least since section 2 of the Human Rights Act requires it. Furthermore, since the three-pronged test is not dissimilar to the German and Canadian ones, this affinity could have been recognised. This chronicle does not end here. As a final twist (to date), the application of the proportionality formula in British courts has since departed from the Commonwealth formulae. Some recent decisions have not cited de Freitas and Daly’s three pronged test; alongside such cases, others mention proportionality as a unitary concept, in a form closer to the ECtHR’s simple test, relying on ECtHR jurisprudence.60 Even more telling is the fact that in none of the cases applying British domestic law – Daly included – did the court actually consider each of the subtests separately, as do German, Canadian and Israeli courts, even the Privy Council in de Freitas. Rather, proportionality is decided by a general assessment consisting of balancing the interests involved. In this respect, the decisions resemble the ECtHR, rather than continental and Commonwealth decisions – but the European source is often marginalised. This is a clear example of a discrete transplant process: at the onset, British courts linked their transplant with Commonwealth countries only, marginalizing European De Freitas (n 56) 80. ibid 547. In Daly (n 47), the House of Lords cited three decisions of the ECJ, including Smith and Grady v United Kingdom (2000) 29 EHRR 493, but only in the context of the need to apply a stricter test than classic unreasonableness (545–56, 547, 549). 59 Daly (n 47) 547. 60 For cases applying a simple test, or merely deciding without detailed proportionality analysis, see, eg R (Alconbury Development Ltd) v Secretary of State for the Environment, Transport and the Regions [2001] UKHL 23, [2003] 2 AC 295; Kay v Lambeth London Borough Council [2006] UKHL 10, [2006] 2 AC 465; G (Adoption: Unmarried Couple) [2008] UKHL 38, [2009] 1 AC 173 (HL); Attorney General’s Reference (No 3 of 1999) [2009] UKHL 34, [2009] 3 WLR 142; R v Waya [2013] 1 AC 294. Decisions in which the three-pronged test is cited include A v Secretary of State for the Home Department [2004] UKHL 56, [2005] 2 AC 68; R (ProLife Alliance) v British Broadcasting Corporation [2003] UKHL 23, [2004] 1 AC 185, 253; Huang v Secretary of State for the Home Department [2007] UKHL 11, [2007] 2 AC 167; and AS (Somalia) v Secretary of State for the Home Department [2009] UKHL 32, [2009] 1 WLR 1385. 57 58
Margit Cohn 201
sources, and its actual reliance on the ECtHR-type formula remains virtually undisclosed. In this case, the courts have succeeded in presenting a process that is seemingly common law based, in which common law formulae seem to have been injected into the European measure. The more recent emphasis on ECtHR case-law has, again, been discrete, with no explicit departure from former reliance modes. V. PROPORTIONALITY REASONING: BALANCING AND CONSEQUENTIONALISM
The idea of balancing has become almost synonymous with the concept of proportionality. Here, I do not refer to balancing in its US meaning,61 but to the law of balancing, defined by Professor Alexy as follows: ‘The greater the degree of non-satisfaction of, or detriment to, one principle, the greater must be the importance of satisfying the other’.62 Balancing is used as the main metaphor to explain what judges do when they rule upon the proportionality of a contested act. This is evident in much of the literature. Sufficient to mention Stone-Sweet and Mathews’ recent article, which carries the title ‘Proportionality Balancing and Global Constitutionalism’; a recent conference held in the Ramat Gan Law College in Israel, under the title ‘Rights, Balancing and Proportionality’; and Aharon Barak’s extensive treatment of balancing in his recent book and elsewhere.63 Both Professors Alexy and Barak are careful to note that balancing is in fact linked, even ‘identical’ (in Alexy’s words) with the third proportionality subtest, the so-called proportionality stricto sensu test discussed below as part of the well-travelled threepronged proportionality formula.64 There is no denying that balancing, in the Alexian sense, is required at that stage, but identifying proportionality with balancing carries two difficulties. The usual formula used for the application of the proportionality doctrine includes several earlier steps, to be taken before the assessment of the benefits versus the impairment of a right. First, the purpose or objective of the measure should be identified and a decision should be made regarding whether a protected right was infringed. Then, the first two subtests should be applied: the requirement for a rational connection and the assessment of whether the least restrictive measure was chosen. None of these have much to do with balancing. The decision on the legislative purpose may require statutory interpretation or other modes of purposive reasoning; the attachment of the challenged measure to a distinct protected right has likewise no link to proportionality reasoning; similarly, the finding over the existence of a rational link between the purpose and the measure and the assessment of whether the least harmful measure had been chosen require other forms of reasoning. Of course, one may argue that all the stages preceding the third subtest are mere preliminary steps required for the exercise of the ultimate balancing act. But another type of 61 Reliance on ‘balancing’ in this context is different from the ‘balancing’ tests used in constitutional adjudication in the United States, which is often posited as the alternative to proportionality reasoning. For a presentation of the two as competing concepts see, eg Cohen-Eliya and Porat, ‘American Balancing’ (n 1); Barak, Proportionality (n 1) 600–17 (juxtaposing ‘categorization’ as an alternative to proportionality reasoning). 62 R Alexy, A Theory of Constitutional Rights (Oxford, Oxford University Press) 102, 401. 63 ibid 401; Barak, Proportionality (n 1) 426–54; A Barak, ‘Proportionality and Principled Balancing’ (2010) 4 Law and Ethics of Human Rights 1. 64 ibid 7.
202 Proportionality in Israel and Beyond: Four Aspects reasoning may stand behind proportionality decision-making (and, in fact, most other types of review). Should not the consequential element of proportionality reasoning receive further attention? The consequential element of proportionality is not completely absent from the debate: Professor Barak’s discussion of ‘principled balancing’, developed to fine-tune the third subtext, is considered a ‘consequential’ test, but in spirit, the thesis remains focused on the act of balancing.65 My tentative suggestion begins with some obvious observations. Under any classic vision of judicial decision-making, courts are to decide disputes between parties, who present a distinct set of legal arguments that support a distinct reality. For example, an application against a statute that permits indefinite detention in certain cases is based on the argument that the statute wrongfully impairs personal freedom; here, applicants push for a reality in which no such incarceration may occur. The respondent, in arguing that the statute passes all legal hurdles successfully (including, of course, proportionality), requires the court to assert the legality of a reality in which such incarceration is a viable possibility. Thus, one may depict the decision over the proportionality of a challenged act as one that is the product of a choice between two alternative realities. In this way, a judge may be viewed as making a consequence-oriented decision; in the example above, this means choosing between the existence of indefinite detention and its absence. The consideration of the consequences of the choice to be made between these two realities seems central to the application of proportionality reasoning. In fact, balancing seems no more than a tool for the assessment of the propriety of the chosen consequence. This type of reasoning is not limited to application of the proportionality principle; it is part of judicial decision-making in most cases involving judicial review. Its recognition in the context of proportionality is an important element of an all-rounded vision of judicial decision-making that embraces several reasoning methods that have little to do with balancing.66 VI. CONCLUSION
This chapter considered four aspects of the proportionality principle. First, I addressed a neglected distinction between proportionality as an overarching principle and proportionality as a device for decisions involved in the limitation of a constitutionally protected ibid 1, 10 and 12. Two comments are needed here. First, courts may deny redress for a variety of reasons that have nothing to do with a decision on the merits, or may decide without touching the substance of the arguments. For example, an application filed after the period allowed for challenging a decision or one that does not pass the standing test, will be summarily rejected, and applications may succeed due to the finding of improper procedure, as in the case of absence of hearing or consultation. In all of these cases, the court does not make any decision regarding the substantive arguments of the parties. However, the immediate outcome will still be the establishment of one of the two realities proposed by the applicants. My second comment refers to the fact that in some cases, a compromise is struck. This may happen when courts act as brokers for settlement outside the court, or inside the court, by granting the settlement the force of a judicial decision. In another type of case, courts may grant only a partial remedy or one that is otherwise different from the remedies requested by the applicants. In all such cases, the emerging reality will be different from the realities advanced by the parties. One may find that the court is involved in balancing, but this balancing can be viewed as outcome oriented as in the former cases. 65 66
Margit Cohn 203
human right. I showed that in Israel, proportionality essentially derives from its recognition in the human rights Basic Laws of 1992; it was applied also in cases involving unfair allocation of sources or unfair treatment, but to date, only in cases which touch, if loosely, on the idea of equality. Whether further development expands the remit of the principle is a matter for the future. Following a second distinction between a concept (or doctrine) and formulae developed to flesh out the concept, I studied the proportionality formula developed in Israel as one case of a cross-system preference for the three-pronged formula. I then noted the strategic benefits of formula-making and the distancing effect of reliance on complex, seemingly objective formula, which may explain the amazing popularity of the three-pronged formula. In part IV I compared the readiness of courts, when importing foreign formulae, to disclose the source of their transplant; I showed that different shades of candour exist, and these may reflect system identities and patterns of self-inclusion on the transnational level. I identified examples of discrete transplants, in which the true sources of the formula are undisclosed, and assessed the benefits of such types of transplants in settling tensions between universalization and exceptionalism. Finally, I suggested that excessive attention has been granted to balancing in the context of proportionality reasoning. While an important element, balancing is in fact supplemented by other modes of reasoning; ignorance of these other aspects, inter alia consequential reasoning, may lead to a misunderstanding of the complex task judges take when applying this ground of review.
14 Constitutional Law in an Age of Globalization VICKI C JACKSON*
I. INTRODUCTION
T
HE TITLE OF our panel was, ‘Constitutional Law in an Age of Globalization’. How does globalization relate to constitutional law? Three fine papers (now chapters of this book) grew out of this question. Before commenting on the excellent chapters, let me say a few words about globalization and about constitutional law. Globalization is a term that is frequently used and contains multiple meanings. The term may refer to the greater interaction among economies, persons, governments, and/ or legal systems, and it is in this sense that it was probably most germane to our conference. But it may also refer to the rapidity, ease, accessibility and coverage of communication and transmission and spread of information through global technologies; this aspect of globalization is also germane to our enquiry. Globalization may include as well the growth of transnational organizations that exercise power – governmental, or private; lawful or unlawful. Or it may refer to the processes and institutions which promote these phenomena, including cycles leading to cascades of greater degrees of global interconnection, as when global crime, global terrorism, or global environmental threats to the Earth call for responses beyond those of a single national state; thus, globalization may also refer to the pressures on national states as fundamental organizing units. In this sense, as well, the multiple processes of globalization bear on domestic constitutional law. In addition to having multiple meanings, globalization’s impact is not likely to be uniformly distributed across legal systems. For example, more powerful states and more vulnerable states may have very different relationships with international law, or to the views and practices of others. So one must be careful to avoid overgeneralizations and to recognize the complexity and, I will suggest, multi-directionality of the influences of globalization on constitutional law. On constitutional law: all states have ‘constitutions’ in the sense of identifiable organizing principles for constituting the government of the state and norms for how that power is exercised. But by ‘constitutional law’ one might mean something beyond the * The author retains the copyright to this chapter. With thanks to Jason Lee and Aatif Iqbal for able research assistance, and to Margit Cohn, Moshe Eliya-Cohen, and Iddo Porat for writing such interesting chapters.
206 Vicki C Jackson fact of a politically identifiable and describable constitution, such as a basic law that constrains government. Not all states have constitutions designed to limit the exercise of government power, and not all have constitutional law in one of three important senses: supremacy, entrenchment and judicial enforcement. On the first, most constitutions assert the supremacy of constitutional over ordinary law. Part of the ‘Constitutional Revolution’ in Israel in the early 1990s was the more clearly developed assertion, in important part through the Israeli Supreme Court, of this form of supremacy for the Basic Laws. Although most states have this feature, not all do and not all do consistently. For example, in states with what Mark Tushnet has called ‘weak’ constitutional review,1 ordinary law may be made supreme over constitutional law through constitutionally specified mechanisms. Moreover, increasingly, as a result of globalization, there are questions about the priorities of international law and domestic constitutional law, in situations where they are in conflict; significant scholarship suggests that constitutional law is not ‘supreme’ over conflicting international rules.2 Second, for many jurists, a constitution is a form of law that is entrenched more deeply than other forms of law, whether by legal rule or by convention. Yet not all states have deeply entrenched constitutions. So while in the United States it is common for scholars to assume that a constitution is necessarily more deeply entrenched than ordin ary law, this may not necessarily be so and plainly may not be so to the same degree. Whether a ‘Basic Law’ has to be adopted by a greater or different vote than regular legislation, and whether it is ‘entrenched’ against future change by special voting rules, seem to be somewhat unsettled questions in Israel. Yet entrenchment is not necessary for a constitution to emerge. British scholars have long discussed and sought to define the UK ‘constitution’, many years before the Human Rights Act 1998 (HRA) was passed.3 Twenty years ago, a debate was whether Israel had a constitution, a debate that continued into the 1990s. The debate in Israel now may have moved on to arguments about the content of the existing norms, about the role of the Supreme Court in articulating and enforcing those norms, and about prospects for change in those norms in the future. Third, although the enforcement role of courts is widely associated with constitutionalism, it has been questioned whether this is necessary and in particular, whether courts in a constitutional system must have the power to declare enacted laws unconstitutional. Plainly there are constitutional systems in Western democracies that have functioned without this latter power (as in the case of Switzerland with respect to national laws). In the United States there are scholars who argue that in order for a constitution to function as ‘law’ it must be in substantial part capable of enforcement in judicial actions with respect to questions arising under the constitution.4 Others disagree, arguing that ‘constitutional law’ is a special form of law that is based primarily on deep political judgements and thus need not be enforced primarily by courts but could be enforced concurrently, or primarily, by either legislatures or ‘the people’ through elections, refer1 See M Tushnet, Weak Courts, Strong Rights: Judicial Review and Social Welfare Rights in Comparative Constitutional Law (Princeton, Princeton University Press, 2008). 2 See, eg M Kumm, ‘The Cosmopolitan Turn in Constitutionalism’ in JL Dunoff and JP Trachtman (eds), Ruling The World? Constitutionalism, International Law, and Global Governance (Cambridge, Cambridge University Press, 2009). 3 See, eg SE Finer, V Bogdanor and B Rudden, Comparing Constitutions (New York, Oxford University Press, 1995) 40. 4 See, eg C Vázquez, ‘Treaties as Law of the Land: The Supremacy Clause and the Judicial Enforcement of Treaties’ (2008) 122 Harvard Law Review 599.
Constitutional Law in an Age of Globalization 207
enda and the like. 5 There is little doubt that the courts play a significant role in enforcing constitutional norms in the United States and, increasingly in Israel. But given prior Israeli experience, as well as British experience until the HRA, one can surely conceive of and identify states which have had functioning constitutions that are in important respects not subject to judicial enforcement. These factors, as well, provide important differentiations among the countries of the world with respect to what is described as a ‘constitution’ or as ‘constitutional law’. So my first point is simply on the need to be quite careful about what it is we mean when we refer to globalization and what it is we mean by constitutional law. I will comment briefly on each of the three excellent papers (subsequently chapters) that were prepared for the panel, and then develop an idea inspired in part by these chapters about two different ways in which ‘proportionality’ is used in constitutional law. II. TRANSFORMATIVE CONSTITUTIONS AND PROFESSOR MOSHE COHEN-ELIYA’S CHAPTER
Professor Cohen-Eliya offers a very helpful analytic for understanding transformative constitutionalism which, he says, is characterized by five features: first, an active role of the court; second a ‘value’ oriented discourse; third, the identification of ‘positive’ duties on government; fourth, the extension of constitutional ‘values’ to the realm of private relations; and fifth, the use of balancing or proportionality as the ‘reigning analytic’ designed to facilitate the advancement of constitutional ‘values’. I find this an illuminating account, but want to suggest that while these features are characteristic of some recent ‘transformative’ constitutions, they may not be true of all imaginable constitutional ‘transformations’, including the conservative or reactionary transformation that some might see occurring in the United States. I will also question whether the judicial challenges of a ‘transformative’ constitution are that different from the challenges of a ‘preservative’ one, and suggest that the degree of challenge depends on domestic politics rather than the particular ambitions of a particular constitutional instrument. First, the idea of a transformational constitution has a layer of ambiguity: is it the constitution that is transformative, that is, does the constitution have some form of agency in redirecting a society towards a new set of goals? Does the constitution simply embody change that has already occurred on the ground? Or is it somewhere in between with multi-directional influences running back and forth? How one answers this question in any given situation may bear on the role of ‘values’ in constitutional adjudication. I agree that resort to values can be a powerful tool for change. If existing legal rules are not operating satisfactorily, it is a classic legal move to articulate the purpose, or value, behind a legal rule as a basis – in legal reasoning – to change the rule itself. But this move can be deployed both for the kind of progressive transformational constitutional change we saw in Germany and South Africa, or for the kind of incrementalist ‘updating’ of law to meet new social conditions, as in changes in the ‘private law’ of tort with the recognition of duties unknown in prior decades. So while resort to ‘values’ may be typical of legal change, the direction and scope of the change is uncertain and whether 5 See, eg LD Kramer, The People Themselves: Popular Constitutionalism and Judicial Review (New York, Oxford University Press, 1994).
208 Vicki C Jackson the change is one in which law takes the lead in effecting social change or rather adjusts itself to forms of change that are already occurring is another matter. Second, on the idea of positive duties: both in Germany and South Africa the new constitution imposes affirmative duties on government. This may indeed be necessary for a transformational constitution; that is, if one wants to reject the status quo, requiring the government to act positively in that direction this might be thought necessary. On the other hand, can we conceive of a transformational constitution that empowers, but does not require, the government to act? Might we in that case call the constitution a democratically transformational instrument, to capture the element of democratically exercised discretion over the pace of change? Third, on the association of transformative constitutions with the imposition of constitutional values on private persons, I wonder whether this can be properly associated with a transformational constitution or is instead part of a more general phenomena in the world of law, consisting of influences between the worlds of public law and private law, and between domestic and international or foreign sources.6 Such seepages, or influences, of public law on private law, or of the values of public law on the development of private law, may reflect not only transformational aspirations (as, in the countries discussed in his chapter, they surely do), but also the growth of a kind of anti-formalism in both law and society, what Mayo Moran calls a ‘move away from the traditional account of legal authority’.7 Indeed, the United States might itself be regarded as home to some forms of ‘seepage’ between public ‘values’ and ‘private’ law.8 The ‘state action’ question in the United States has been a difficult one, and the constitutional discourse is undoubtedly, in general, hostile to the assertion that constitutional values influence the obligations of private persons inter se.9 Nonetheless, constitutional influence on private obligations even in the United States extends beyond the parameters of the Thirteenth Amendment and the rule in Shelly v Kramer.10 The entire edifice of constitutional protection of the press and other speakers from ordinary tort rules in libel and defamation has arisen in cases involving what could be conceptualized as ‘private law’ disputes. While in its origins the doctrine restricting common law defamation actions on behalf of freedom of expression involved public officials as libel plaintiffs,11 the expansion of the doctrine to ‘public figures’,12 and the extension of First Amendment protection in actions involving dignitary interests asserted by private persons against the behaviour of others,13 more clearly 6 See generally, M Moran, ‘Inimical to Constitutional Values: Complex Migration of Constitutional Rights’ in S Choudhry (ed), The Migration of Constitutional Ideas (Cambridge, Cambridge University Press, 2006) 233. 7 ibid 245. 8 Reflecting on whether the imposition of constitutional values on private relations is a hallmark of transformational constitutions leads me to wonder whether constitutions that impose duties on private persons vis-à-vis the polity have any connection with their status as transformational. There are constitutions that impose duties on private persons vis-à-vis the polity – duties of military service, duties to vote, or in Germany, a duty to resist change in its fundamental commitments. Is imposition of such duties ‘preservative’, ‘statist’or ‘transformative’? Could they be both part of a transformation and an effort to preserve that transformation, in the German case? 9 See, eg United States v Morrison 529 US 598 (2000). 10 Shelly v Kramer 334 US 1 (1948); see also Hurd v Hodge 354 US 24 (1947). 11 New York Times Co v Sullivan 376 US 254 (1964). 12 Curtis Publishing v Butts 388 US 130 (1967); Hustler Magazine v Falwell 485 US 46 (1988). 13 Snyder v Phelps 131 S Ct 1207 (2011) (upholding a lower court opinion overturning a jury verdict for common law tortuous injury for a family of a deceased soldier against persons who protested in hateful ways at the soldier’s funeral).
Constitutional Law in an Age of Globalization 209
have moved the constitutionalized discourse into the ‘private realm’ on the basis of an idea that application of common law rules by the courts would threaten constitutional rights.14 Finally, on the use of proportionality analysis as a hallmark of transformational constitutions: there is no doubt that as a matter of positive fact one can observe this doctrine being used in Germany, South Africa, Canada, Israel and many other leading courts. But, as Professor Cohen-Eliya notes in discussing the ‘gamble’ involved in reliance on values (and their use in the Adalah v Minister of Interior ruling),15 balancing may be associated with a distinctly statist, authoritarian or preservationist posture – as the Dennis case,16 or recent debates in the United States over the use of torture, might be thought to illustrate. As a normative matter, Professor Cohen-Eliya argues that instead of a value-oriented ‘transformative’ constitutional undertaking, the Israeli Court should re-conceive of its constitutional task as being to advance a ‘democracy-enhancing’ view of what constitutional law is intended to advance. As the paper recognizes, John Hart Ely’s theory has been subject to critique, even by those who are enthusiasts, in part because it itself depends on a set of value judgements that are relevant to defining democracy and its defects. In riven societies, where there are relatively stable majorities and minorities, it can be a particular challenge to sort out ‘process failures’ from the products of ‘ordinary democratic politics’ that should be respected. For example, in Professor Cohen-Eliya’s illustration (involving a challenge to exemption from military service), does it matter if a majority votes against its own seeming self-interest because it has made a judgement that the minority group needs and deserves the benefit or exemptions being conferred, or is instead doing so because of interest group bargains among coalitions, including that minority, that may entail adverse consequences in other areas for other minority groups? Second, there may be a tension between Ely’s theory and the rule of law in some instances. For example, if the action of the majority in favour of the minority were to contravene a clear written norm – if for example a legislature dominated by male members were to confer a benefit only on women and not on men – would that end the need for judicial enquiry of whether a legal norm requiring equality of the sexes had been met? While this example might argue for a more deferential standard of review, surely there is a rule of law basis for a court to determine whether the statute is consistent with the constitutional norm, notwithstanding who approved it. Let me close this discussion by emphasizing an important point on which I find Professor Cohen-Eliya quite persuasive. To the extent that there is concern about backlash and the perceived illegitimacy of the court, Ely’s approach can in some cir cumstances have real advantages. When a court makes a move that is understood as increasing democratic accountability and participation – as in the Baker v Carr line of cases in the United States17 – its legitimacy may in the long run be strengthened by the constituency that benefits from access to the polls. The law/politics connection is in this sense arguably a self-reinforcing one.
14 Query whether seepages and influences may run in both directions – that is, whether private law developments (at least in theory) hold the potential to influence public law understandings as well. 15 HCJ 7052/03 Adalah v Minister of Interior Affairs 61(2) PD 202 [2006]. 16 Dennis v United States 341 US 494 (1951). 17 Baker v Carr 369 US 186 (1962).
210 Vicki C Jackson III. PROFESSOR IDDO PORAT AND THE USE OF FOREIGN LAW IN ISRAELI CONSTITUTIONAL ADJUDICATION
Professor Porat’s chapter paints an interesting and complex picture of the role of foreign and international precedent in the Israeli cases, and offers a thoughtful analysis of seven factors that contribute to Israel’s use of such precedent. Due to limits of time and space, I comment only on three aspects of the positive analysis in the chapter: the ratio between the use of foreign law in rights cases as compared to others; the use of ‘proportionality’ analysis; and the role of geopolitical isolation. I will then briefly go on to consider the normative concerns Professor Porat identifies with ‘global constitutionalism’. As his chapter suggests, the ratio of foreign sources cited in Israeli Supreme Court cases is considerably higher than one would find in the US Supreme Court. In significant or highly contested cases it is even higher. According to Professor Susie Navot’s research, which Professor Porat discusses, in 31 per cent of the High Court’s constitutional cases between 1985 and 1994 there were citations to foreign law; and according to her study, which separately analyzed references in constitutional case law involving institutional issues and individual rights issues, in individual rights issues the percentage goes up much higher, to close to 70 per cent, while in ‘institutional’ constitutional cases the percentage is something around 30 per cent.18 This result is not surprising, and accords with theoretical work I and others have done comparing the use of foreign or international law in institutional as compared to rights cases. I have speculated, in other work, that on structural constitutional issues – especially of federalism – foreign law is less likely to be useful because of: (1) the highly distinctive and interconnected balance typically struck on federalism issues (and perhaps to a lesser extent on separation of powers schemes); (2) the very historically specific, compromised nature of those institutional arrangements, which can make analogies to other systems difficult to develop; and (3) the absence of international archetypes on which other constitutions are modelled and which courts both national and supra national may base their own judicial judgments.19 Second, on proportionality: I suggested above that proportionality may not be as endemically connected to transformational constitutionalism as Professor Cohen-Eliya suggested. Here, I want to suggest that proportionality analysis may not be as endemically connected to foreign law’s influence as many assume. The German Constitutional Court, for example, so far as I am aware does not typically cite foreign cases; I do not know its rate but would not be surprised if it was closer to the United States than to Israel. Yet its use of proportionality analysis is well known, indeed, it is one of the leading courts in developing proportionality analysis. While some of the enquiries of proportionality analysis lend themselves to considering foreign experience, so too does any constitutional approach open to considering the likely consequences of interpretive choices, which do not necessarily involve the structured form of enquiry or even the overt balancing called for by the proportionality doctrine; moreover, categorical interpretive approaches may find useful the discussions or views of other constitutional democracies on the meaning of categories such as ‘speech’, or ‘religion’, or ‘privacy’. Ch 11, text accompanying nn 20 and 21. See, eg VC Jackson, Constitutional Engagement in a Transnational Era (New York, Oxford University Press, 2010) 228–32. 18 19
Constitutional Law in an Age of Globalization 211
Third, on the significance of geopolitical isolation. This is a very interesting insight in this chapter. For it suggests that it is very difficult to be a constitutional democracy by oneself; that having other forms of the same basic governmental and constitutional structure as models is helpful to the distinctively domestic project of bringing it to success for one’s own people. In other words, Professor Porat is suggesting an interrelationship, an interdependence among the democracies of the world. So, questions: Why is this? Is it hard to be the only constitutional democracy, in a region, or in the world? Is this just a matter of institutional isomorphism – that is, the tendency of a dominant form of institutional organization to be imitated (even if it is not functional)? Is it because constitutional democracies in some more positive sense reinforce one another’s values, or progress? If so, by what mechanisms?20 Or is it because constitutional democracies will be more likely to come to the aid of other constitutional democracies to which they feel a sense of connection? On the normative questions, I agree with much of Professor Porat’s evaluation of the relative risks involved.21 He is completely correct to emphasize the risks of ‘getting it wrong’, because law – including constitutional law – is, as he says, ‘deeply embedded’ in a broader legal system and a particular socio-legal and political and cultural context. I raise only a small caveat to his discussion criticizing the Israeli Supreme Court in Kol Ha’am for referring to Dennis without taking into account important differences in the two contexts. That a reference disregards distinctions between two constitutional cultures may not entirely condemn it: courts do not always have the time and resources that academics have fully to explore the comparative setting, and must often rely on the work of academics to assist in this. If the claim were made that the two contexts were similar and that accordingly the Dennis approach should apply, Professor Porat’s critique would have much substance; yet exactly what role Dennis played in the Kol Ha’am opinion is not entirely clear, but appears relatively small.22 And in any event, as Professor Porat 20 For a suggestion that democracies experience ‘co-evolution’ of ideas on fundamental rights or values as a result of the effects of transnational social movements, see R Dixon, ‘A Democratic Theory of Constitutional Comparison’ (2008) 56 American Journal of Comparative Law 947. Professor Dixon develops a theory for constitutional courts’ considering the developments in other countries based on co-evolution and on the notion that there may be blockages in particular systems that prevent ordinary democratic politics from giving expression to emerging majorities’ views. But putting aside this theory for judicial decision-making, is it possible that the process of free interchange about ideas, including ideas of governance, that is characteristic of constitutional democracies, itself improves the discourse in other democracies, by opening up and expanding and subjecting to competing perspectives the debate over public issues? 21 For discussion of objections from interpretive theories grounded in originalism, contractarianism, popular sovereignty, majoritarianism or representation-reinforcement, see Jackson, Constitutional Engagement (n 19) 20–23; of objections from jurisprudential ideas about constitutional law as an expression of self-identity or of law as organic or autochthonous, ibid 18–19, 24–25, 31–32; and of objections grounded in judicial discretion, ‘cherry picking’, expertise, and the challenge of comparisons, ibid 26–27, 183–91. Other objections, based on cultural exceptionalism, political resistance, principled disagreements or fear of cultural elites are also discussed. ibid 28–30. 22 From the English version of the Israeli case that I was able to find on line, HCJ 73/53 Kol Ha’am v Minister of Interior 7 PD 871 [1953] (in Hebrew): www.cfisrael.org/a554.html?rsID=405, Dennis appeared to play a relatively small role in the opinion, with considerable discussion of the Holmes and Brandeis pro-free speech dissents in earlier cases. The conclusion of Justice Agranat that reliance on the mere possibility of a dangerous tendency was not enough to support the censorship, but rather the words ‘likely to endanger’ required that there be a probability that the publication would lead to the harm, was, in its context, plainly a speech protective conclusion. In explaining why it was not adopting the ‘clear and present danger’ test, the judgment seemed to rely more on the dictionary definition of the statutory word being interpreted, ‘likely’; and the Court suggested that where harm was not imminent, a greater degree of caution should be exercised before restricting speech. I would have said the major thrust taken from the US cases in the opinion was the importance in a democracy of adopting speech-protective rules, and that the Dennis decision, cited briefly with a ‘cf’ cite to the
212 Vicki C Jackson notes, the use made of the idea of ‘balancing’ in the Israeli context was as a speech- protecting move, rather than, as in Dennis, a speech-threatening departure from the ‘clear and present danger’ test. This reinforces the point (which I make later, in connection with proportionality) that an apparently similar doctrine can be applied in quite different ways by different courts. Professor Porat makes an important argument that, of the risks that have been identified, perhaps the greatest lies in what he calls ‘global constitutionalism’. Although his discussion about judges writing for a transnational audience emphasizes more the risks than the potential benefits (eg of considering outside perspectives in promoting impartiality in judging), I want to focus on Professor Porat’s concern, which is one I share, and that is the risk of failing to pay sufficient attention to the democratic roots of constitutionalism. In a recent review essay, I argued that some theoretical work in global con stitutionalism, designed to describe emerging relationships, or desirable directions for relationships between domestic and international law, towards concepts of ‘cosmopolitan constitutionalism’,23 paid insufficient attention to the importance of democratically legitimated and legitimating domestic sources of law.24 This point is conditioned by today’s political contexts, in which the national state or even smaller units are far more plausible locations for the exercise of democratic decision-making than are supra national or international organizations. A question here is why, as Professor Porat suggests, such concerns (of failing to attend to internal democratic sentiment through a ‘false sense of consensus’)25 would be a particular risk or a special danger in a small democracy? Why would it not be easier in a small, rather than a large, democracy for judicial and legal elites to stay in touch with local democratic opinion at the same time that they become familiar with global views? One final thought about the risks of global constitutionalism, as Professor Porat discusses it: one aspect of globalization that I noted at the outset was the increased accessibility and availability of information about legal systems all over the world. The increased information creates a situation in which comparisons – well-informed or not – are increasingly inevitable. Most jurists in the United States probably think they know something about other legal systems, and implicitly make comparisons on various issues with that assumed knowledge. Yet ‘knowledge’ acquired thus implicitly and without conscious study and reflection (to say nothing of adversarial briefing) may be in error. This possibility was vividly illustrated by the assertion, in Bowers v Hardwick,26 by Burger CJ that homosexual sodomy was condemned ‘throughout the history of Western civilization’, at a time several years after the European Court of Human Rights (ECtHR) had decided that an Irish prohibition of sodomy was invalid under the European Convention.27 This mis-impression was later corrected by the Court in Lawrence v
‘new approach of the majority opinion’ and Frankfurter’s judgment (cited for a speech-protective quotation from Sir William Haley, and for a collection of authorities on the clear and present danger test) played a smaller role. 23 See, especially, the argument and analysis by Kumm (n 2) 262. 24 See VC Jackson, ‘Paradigms of Public law; A Review of Ruling the World?’ (2010) 8 International Journal of Constitutional Law 517–62 (agreeing with other scholars that ‘distance matters’ in the degree to which mechanisms of democratic participation or accountability can function well). 25 n 18, part IVCiii. 26 Bowers v Hardwick 478 US 186, 196 (1986) (Burger, CJ, concurring). 27 Dudgeon v United Kingdom (1982) 4 EHRR 149.
Constitutional Law in an Age of Globalization 213
Texas.28 The sequence suggests the importance of encouraging candid discussion of comparative views that jurists are increasingly likely to have, in order to avoid such errors. IV. PROFESSOR MARGIT COHN AND PROPORTIONALITY ANALYSIS
In her chapter, Professor Margit Cohn identifies several important sets of analytical questions in discussions about proportionality related to the difference between pro portionality as a concept and proportionality as a doctrinal formula; the unacknow ledged and acknowledged migrations of particular formulae; the difference between proportionality as a general overarching principle and proportionality as a principle for resolving human rights cases; and the difference between proportionality (or balancing) and consequentialism. She makes a number of telling points. For example, drawing in part on her own earlier scholarship, she argues that the very complexity of the prevailing doctrinal formulation can be understood as designed to give the appearance of a formalist, scientific reasoning process that is remote from the political choices that its deployment is intended to obscure. A question this account raises is whether the particular form of complexity matters, on this account, or only that the doctrinal formula has several steps that are engaged with in their application.29 Professor Cohn also weaves an interesting narrative about the migration of the three pronged formula from Germany and Canada, suggesting that Canada borrowed in an unacknowledged way from Germany, possibly as a result of the influence of Joel Bakan when he was a law clerk in the Canadian Supreme Court after having studied in Europe.30 She points out the silence in Britain about the German, or European, origins of the proportionality analysis it now uses, noting in particular the British court’s reliance on decisions of national courts of past or current members of the Commonwealth rather than on decisions of the ECtHR. The symbolic significance of borrowing from other national courts, rather than from Strasbourg or Germany, in the complex relations the UK has had with the EU and the European Convention is worthy of the focus she brings to it. There may be multiple factors behind this phenomenon. First, one should not neglect the possibility that judges may be making these choices implicitly, without self- awareness, that is, without consciously deciding not to cite ECtHR jurisprudence; yet what ‘comes to mind’ may nonetheless be influenced by sets of strategic concerns. Second, identification with other Commonwealth countries, as well as perhaps a greater sense of familiarity with the style of opinion writing, may make those opinions feel easier to use than decisions of the German Constitutional Court or the ECtHR. Third, some degree of resistance to the idea of the Europeanization of Britain, may be in play, as in Judith Resnik’s idea of ‘law as affiliation’, which implies that what is cited may be a form of ‘disaffiliation’ as well.31 Finally, in referring to other national courts, rather Lawrence v Texas 539 US 558 (2003). A further question might be to explain what differentiates the simpler from the more complex proportionality formulae used by the European courts: are there different incentives to engage in greater or lesser degrees of complexity in formulating doctrine? 30 Ch 13, text accompanying n 43. But cf D Grimm, ‘Proportionality in Canadian and German Constitutional Jurisprudence’ (2007) 57 University of Toronto Law Journal 383, 384 (noting the possibility of influence from a recent US Supreme Court decision on Canada’s initial formulation of the proportionality test, while emphasizing the possibility of German influence). 31 J Resnik,‘Law as Affiliation: “Foreign” Law, Democratic Federalism, and the Sovereigntism of the NationState’ (2008) 6 International Journal of Constitutional Law 33. 28 29
214 Vicki C Jackson than to Strasbourg, there may be an implicit sense of greater control over one’s own domestic law; one is not risking being read as acknowledging the binding force of international or supranational norms, as the Australian debate over the citation to inter national law (which is controversial) as compared to foreign law (which is much less so) suggests might be possible.32 But in the rest of this comment, I want to focus on the difference between proportionality as an idea or concept, and proportionality as a particular doctrine, and its relationship to the distinction between systems that use proportionality as a principle for resolution of rights cases or as a more general principle, two of the very interesting distinctions drawn in the Cohn paper. Her argument made me want to understand better the significance of these distinctions for more general understanding of the role of proportionality. Is ‘proportionality’ when used as a general principle of law similar to or different from its use in resolving ‘rights’ cases?33 Very tentatively, I will suggest that proportionality – though with the same name, or even the same doctrinal formulae – may actually serve very different functions in these two areas, or in cases that align, to some degree, with the distinction Professor Cohn suggests between these two areas. My suggestion, albeit tentative, is that proportionality as a concept applied to rights – or at least as applied to some rights – has the distinctive possibility of providing a doctrinal mechanism to bring law in books closer to a felt sense of justice. Many would agree that the purpose of legal systems is to advance justice; yet many others share the cynical or ‘realist’ view of Oliver Wendell Holmes, who insisted that talking about ‘justice’ means ‘shirking thinking in legal terms’.34 The insistence on a separation between law and justice is a hallmark of most jurisprudential schools. And yet law that strays too far from a widely held sense of justice can lose its socio-legal legitimacy,35 and thereby its capacity to serve rule of law functions. Thus, for example, in evaluating the constitutionality of punishment, many legal systems have adopted some form of proportionality review. The operation of this form of review holds out the possibility of bringing the coercive force of law, as it metes out punishment in the criminal justice system, closer to broader understandings of justice and morality. True, there will often be contests about these issues, as divergent conceptions of justice exist.36 But there may be overlap between the results required by different conceptions of justice, and to this extent application of legal standards of proportionality may well move law closer to justice. In the United States, proportionality is not regarded as a general tool of constitutional analysis, at least not by a majority of the court. Yet traces of proportionality as a concept can be found, especially in the court’s capital sentencing jurisprudence. Thus, in 32 See, eg Al-Kateb v Godwin [2004] HCA 37 (compare the opinions of McHugh and Kirby, respectively). For discussion, see Jackson, Constitutional Engagement (n 19) 34, 300 fn 82, 362 fn 51. 33 While Professor Cohn’s discussion of proportionality as a ‘general principle’ focused primarily on its use in administrative law settings, where claims of fairness in the administration of programmes vis-à-vis individuals were at stake, I focus on its use as a ‘general principle’ particularly in disputes over allocations of powers. 34 See letter from Oliver Wendell Holmes to Dr Wu, in OW Holmes and HC Shriver, Oliver Wendell Holmes: His Book Notices and Uncollected Papers (New York, Da Capo Press, 1936) 201, quoted in R West, ‘Re-Imagining Justice’ (2002) 14 Yale Journal of Law and Feminism 333 (‘I hate justice’, Holmes wrote. ‘I know that if a man begins to talk about that, for one reason or another he is shirking thinking in legal terms’). 35 See RH Fallon Jr, ‘Legitimacy and the Constitution’ (2005) 118 Harvard Law Review 1787. 36 For thoughtful discussion of the competing conceptions of ‘proportionality’ that may exist even as to sentencing for crime, see PS Karlan, ‘“Pricking the Lines”: The Due Process Clause, Punitive Damages, and Capital Punishment’ (2004) 88 Minnesota Law Review 880.
Constitutional Law in an Age of Globalization 215
Coker v Georgia37 the Court held (by a seven to two vote) that the death penalty could not constitutionally be imposed for the crime of rape of an adult. For the plurality of four, the death penalty was ‘excessive’ and ‘grossly disproportionate’ as a punishment for rape of an adult person, based on both the actions of most state legislatures in rejecting the death penalty for rape and the Court’s own ‘independent’ judgment;38 for Powell J, the death penalty was ‘ordinarily’ a disproportionate and hence unconstitutional punishment, but he would have distinguished cases of ‘aggravated rape’ in which ‘victims are so grievously injured physically or psychologically that life is beyond repair’.39 Even the two dissenters ‘accept that the Eighth Amendment’s concept of disproportionality bars the death penalty for minor crimes’.40 Although there is some reason to think that on some kinds of matters there are close to universal intuitions of morality, perhaps grounded in a universal ‘moral grammar’, that are reflected in legal doctrines or provisions,41 there are many possible objections to the claim that proportionality helps bring law closer to justice. It might be objected that there is no sufficient agreement on what justice means to provide any form of constraint on judicial decisions, or that justice is not connected to socio-legal perceptions of morality, or that there is too little overlap between competing and important and different conceptions of justice.42 On any of these accounts, proportionality analysis cannot achieve the effect of bringing law closer to justice; at best its ‘bridging’ possibilities are between law and contestable views of public morality that contribute to socio-legal legitimacy. This bridging effect, even if shorn of its claims to ‘justice’, may still be a reason to think proportionality is a useful analytical approach if it provides for a structured and more transparent way for courts to reason. Moreover, even if proportionality review may provide a way of bringing law closer to important conceptions of justice, proportionality review on ‘human rights’ or individual constitutional rights may not always be the best tool. One might here want to distinguish between rights that are regarded as derogable, and rights that are regarded as nonderogable. A non-derogable right requires categorical reasoning; if ‘torture’ is, in a legal system, always prohibited,43 proportionality analysis will be an insufficient tool – notwithstanding the strong pragmatic claims made by those who would defend torture in some instances. One might also want to consider whether there are differences in situations involving a conflict between individual rights on the one side and general or collective public interests, Coker v Georgia 433 US 584 (1977). ibid 597 (plurality opinion). Justices Brennan and Marshall also concurred in the judgment. Ibid 600–01 (explaining their views that the death penalty is always prohibited by the Eighth Amendment). 39 ibid 603 (Powell J, concurring in the judgment and concurring and dissenting in part). 40 ibid 604 (Burger CJ, with Rehnquist J, dissenting). 41 See J Mikhail, ‘Is the Prohibition of Homicide Universal: Evidence from Comparative Criminal Law’ (2009) 75 Brook Law Review 497 (collecting sources in cognitive sciences, experimental ethics, and comparative criminal law). 42 See, eg Robin West’s effort to describe three very different understandings of ‘legal justice’, which she defines as meaning treating like cases alike. R West, ‘Is the Rule of Law Cosmopolitan?’ (2000) 19 Quinnipiac Law Review 259 (arguing that there are three different models of legal justice, so defined; a tradition oriented understanding of legal justice, which values stare decisis; a libertarian model of treating likes alike, which values contract and market relations; and an egalitarian or communitarian view, which values human equality not based on national state lines). 43 See, eg Convention for the Protection of Human Rights and Fundamental Freedoms (European Convention on Human Rights, as amended), Art 3 (‘No one shall be subjected to torture or to inhuman or degrading treatment or punishment’) and Art 15 (prohibiting derogations from Art 3 even in time of war or public emergency). 37 38
216 Vicki C Jackson on the other, and situations in which rights of different individuals come into conflict, as where road closures impair some citizens’ rights of mobility and freedom of movement but at the same time, at least arguably, protect the rights of others to practice their religion.44 In these latter cases, one could imagine a decisional process in which the enquiries as to ‘minimal impairment’ and ‘proportionality as such’ may need to be doubled, or viewed through the perspective of each affected party, and not merely the party who challenges government action as burdening his or her rights.45 On the other hand, courts often apply proportionality analysis only to examine infringement on the rights of those challenging the regulation, incorporating the ‘interests’ of others in its analysis of the government’s justifications for action. Whether the individual rights that are being protected or advanced by government regulation in such cases should be addressed separately as ‘rights’ or considered only within the rubric of government justification may depend in part on whether the state is conceived to have positive duties to protect such rights, or is only optionally empowered to protect them. Nonetheless, in many cases involving protected individual rights, proportionality as a tool has benefits, in requiring governmental justifications for action, in encouraging the protection of rights by limiting intrusions thereon, and in bringing the law close to broader understandings of just government or right resolutions. But if proportionality has utility with respect to a wide range of individual rights and assists in bringing the law closer to public understandings of justice, with respect to other kinds of constitutional issues – those going to structure and allocation of powers – proportionality as a concept may have quite a different function. In earlier work, I suggested that proportionality analysis has a more limited application to structural constitutional issues of federalism, and other structural issues going to the allocation of powers vertically or horizontally in a polity.46 Let me revisit this suggestion in the context of Professor Cohn’s analytical distinction between proportionality as a general principle and proportionality as it is used in human rights cases. 44 For a case involving such a conflict – at least of interests – see HCJ 5016/96 Horev v Minister of Transportation 51(4) PD 1 [1997] (in Hebrew); cf Attorney General v Irwin Toy [1989] 1 SCR 927 (Can) (suggesting that more deference is accorded, in applying proportionality review, to legislation that seeks to mediate between the rights or interests of two groups in society than to legislation in which the government is the principal opponent, as it were, of the individual challenger); ibid 993 (quoting Dickson CJ in R v Edwards Books and Art Ltd [1986] 2 SCR 713, 779 (Can): ‘the courts must be cautious to ensure that [the Charter] does not simply become an instrument of better situated individuals to roll back legislation which has as its object the improvement of the condition of less advantaged persons’). But the more problematic question is whether one can distinguish cases involving ‘rights’ on both sides from other rights cases; does the ‘public interest’ always entail protection of the rights of some members of the public, or are there cases in which the public interest advances merely ‘interests’ and not rights? 45 cf S de Vries, ‘Balancing Fundamental Rights with Economic Freedoms According to the European Court of Justice’ (2013) 9 Utrecht Law Review 169, 190–91 (discussing ‘double proportionality test’, argued for in the opinion of Trstenjak AG, delivered on 14 April 2010, in Case C-271/08 Commission v Germany [2010] ECR I-07091, paras 189–91). De Vries characterizes ‘double proportionality’ as an analysis not ‘confined to an assessment of the appropriateness and necessity of a restriction of a fundamental freedom for the benefit of fundamental rights’ protection’ but also including ‘assessment of whether the restriction on a fundamental right is appropriate and necessary in light of the fundamental freedom’ (ibid). For an arguably comparable analysis focusing on the marginal intrusions on the physical security and livelihood rights of Israelis and Palestinians, respectively, to different approaches to a national security problem, see HCJ 2056/04 Beit Sourik Village Council v Government of Israel 58(5) PD 807 [2004] para 40 (in Hebrew). 46 VC Jackson, ‘Being Proportional about Proportionality: A Review of David Beatty’s The Ultimate Rule of Law’ (2004) 21 Constitutional Comment 803. Portions of the several next paragraphs are drawn from this earlier work.
Constitutional Law in an Age of Globalization 217
Proportionality analysis is not uncommon in federal systems; it has been used in Canada, and it has recently been developed in the United States as an aspect of Fourteenth Amendment analysis of the scope of federal power. As David Beatty explained its use in Canadian federalism decisions, ‘proportionality’ review entailed examining the reasonableness of federal legislation, for the purpose of sustaining as much as possible the concurrent capacities of both levels of government, provincial and federal.47 In a recent decision, McLachlin CJ wrote that in evaluating challenges to federal statutes as ultra vires federal constitutional power, the Canadian Court has used a rational, functional test to describe the required connection, with the caveat that a test of necessity will apply where the encroachment on the jurisdiction of the other level of government is substantial . . . The idea of proportionality underlies the idea of a rational and functional standard for some cases and standard of necessity for others. The more an ancillary provision intrudes on the competency of the other level of government, the higher the threshold for upholding it on the basis of the ancillary powers doctrine.48
This doctrine, under which ‘ancillary provisions’ are subjected to a different required level of justification depending on how much they are seen to intrude on the competency of other levels, may be applied as well to the review of provincial legislation.49 In systems that do not have specific concurrent powers, proportionality may nonetheless be useful in analyzing federalism questions that turn on the plausibility of the connection between a measure and an enumerated power, as in the United States’ Fourteenth Amendment case law.50 Where the means chosen are grossly disproportionate to the asserted goal, the disproportion may justify a finding that the measure is beyond the scope of the enumerated power. But it is hard to say that proportionality in such cases corresponds to a public sense of justice, or even views of good governance. In such cases proportionality – or disproportionality – is being used in service of a certain kind of categorical rule of law goal, the categorical goal being to help sustain either exclusive or concurrent powers in multiple branches or levels of government. These decisions are not unconnected to human rights, equality and human liberty: for example, limiting jurisdiction over certain kinds of prohibitory laws to subnational levels may enhance liberty by creating the possibility of exit and options to live in states with different regimes; upholding national power to protect minorities may enhance their equality to a far greater extent than if legal protections are needed to be won state-by-state. But these allocations are not systematically connected to the advancement of human rights ibid 844 (citing Beatty, at 25–27). Reference re Assisted Human Reproduction Act [2010] 3 SCR 457, para 148 (Can). The Chief Justice and three others would have upheld all aspects of the federal statute, concluding that any intrusions on provincial powers were relatively small and justified under the rational, functional connection test. Four other justices found several provisions of the Act to exceed federal power, as did a fifth justice writing separately; the fourjustice opinion alluded to a similar idea, without using the language of proportionality. See ibid para 275 (Lebel and Deschamps JJ) (‘According to the General Motors test, the more serious the overflow [of ancillary provisions in a federal statute], the closer the relationship between the impugned provisions and the otherwise valid statute must be. In the instant case, our review of the effect of the impugned provisions has already shown that, if those provisions are viewed in isolation, the overflow is serious. Given the extent of the overflow in this case, we cannot find that an ancillary power has been validly exercised unless the impugned provisions have a relationship of necessity with the rest of the statute’). 49 For application of the ‘ancillary’ doctrine, in its ‘most rigorous form’, to uphold as ‘necessarily incidental’, the validity of a provision in provincial legislation authorizing assisting foreign governments in investigations of securities fraud, see Global Securities Corp v British Columbia (Securities Commission) [2000] 1 SCR 494. 50 See City of Boerne v Flores 521 US 507 (1997); see Jackson, Being Proportional (n 46) 844. 47 48
218 Vicki C Jackson relating either to liberty or equality.51 Perhaps, however, proportionality in these cases could be understood to provide an appeal to juridical/governmental communities, an effort to identify or draw on overlapping areas in the practical understandings of relevant juridical communities. Even if some kinds of allocation of powers decisions are well-served by ‘proportionality’ concepts of ratcheted levels of increasing justification for more severe departures or intrusions, there are other prototypical ‘federalism’ questions that are unlikely to be best served by application of a generalized proportionality enquiry. First, aspects of most federal systems may involve very specific compromises on which national commitments to union are thought to rest.52 For example, in the United States the two Senators per state rule reflects at once a wild deviation from more general constitutional requirements of ‘one person one vote’, and a clear decision to impose a specific rule to protect the powers of all of the states in the national government. Unless one is prepared to argue that the clarity of the constitutional text must yield to later constitutional commitments, it is difficult to see any role for proportionality in analyzing such a question, were it to come before a court. 53 In Canada, the Supreme Court has similarly rejected a challenge, on equality grounds, to the failure of Ontario to provide funding to Jewish religious schools on the same terms as Catholic schools are funded, based on the founding compromises of the 1867 Constitution Act.54 Founding compromises of federal systems, to the extent that they are clearly written in constitutional text and confirmed by decades of practice, simply may not yield to any interpretive moves, including arguments based on proportionality; amendment may be required. A second possible limitation on the use of proportionality analysis in structural cases has to do with the degree to which present occupants of public offices of the government entities involved will be fully ‘representative’ of the interests of the office or level of government on the issues before the court. On some accounts of proportionality, such as that of David Beatty, it is important to focus on ‘the perspectives of those who are most affected by whatever law or government action is under review’ as parties before the court.55 Concrete litigation over federalism issues may do less well in capturing the longer term structural interests that divisions of government powers are designed to advance than individual parties do in capturing unfair burdens imposed on particular individuals by government regulations claimed to intrude on individual rights. Individuals who 51 But cf PG Carozza, ‘Subsidiarity as a Structural Principle of International Human Rights Law’ (2003) 97 American Journal of International Law 38. It might be suggested that, in general, liberty claims are advanced by subsidiarity principles, keeping power at the lowest level, and equality claims conversely by keeping power at the most central level. But these possibilities are contingent on the nature of the politics at each of these levels: if the central government is dominated by a racist party, equality will be served by allowing the exercise of power at decentralized levels and precluding it at national levels, etc. It is possible that allocative decisions might have a systematic connection to values of democracy; participation is easiest in smaller units, generally; but to the extent democracy embraces a right to participate in decisions that affect one, externalities in a globally connected world will often exceed the bounds of existing national states. 52 Some of the next few paragraphs are drawn from Jackson, Being Proportional (n 46) 842–51. 53 See VC Jackson, ‘Multi-Valenced Constitutional Interpretation and Constitutional Comparisons: An Essay in Honor of Mark Tushnet’ (2008) 26 Quinnipiac Law Review 599, 658 (arguing that rule of law considerations should preclude interpretive evolution that ignores such clear and central textual commitments). 54 See Adler v Ontario [1996] 3 SCR 609 (Can); but cf Waldman v Canada (1999) UN Doc CCPR/C/67/694/1996 (rejecting Canada’s reliance on s 93 of the 1867 Constitution Act as a defence when claiming violation of the International Covenant on Civil and Political Rights (adopted 16 December 1966, entered into force 23 March 1976) 999 UNTS 171). 55 DM Beatty, The Ultimate Rule of Law (Oxford, Oxford University Press, 2004) 166.
Constitutional Law in an Age of Globalization 219
challenge laws on structural grounds, as in cases such as United States v Lopez,56 or United States v Morrison,57 do so opportunistically.58 And although governments may participate in litigation, there are difficulties in relying on litigation positions – typically under the control of a handful of executive branch lawyers – for a stable account of a state government’s ‘interest’. The constitutional interests at stake in federalism questions may be thus, at least in some sense, not fully represented in litigation before a court. Likewise, on structural issues of separation of powers, proportionality analysis based on the interests articulated by the parties before the court may also have a relatively smaller role to play. And there are important distinctions between challenges to executive and challenges to legislative action that the concept of proportionality will not adequately reflect. In some systems, at least on some issues, executive authorities or governments may be found to lack constitutional authority to take action without legislative authorization; this is not a question of proportional intrusion or burden but a question of authority. Both democratic and rule of law values are served by requirements of legislative authorization.59 Although courts may well be inclined to find executive action to be ultra vires when it appears to impose disproportionate or otherwise suspect burdens, this does not reduce the requirement for legislative authorization to simply an aspect of ‘proportionality’.60 The broader point I want to come to is this: not only does proportionality analysis not fit, as well, with the range of constitutional problems that arise in vertical and horizontal separation of powers issues as it does with individual rights issues, but the function of proportionality analysis in these two kinds of cases seems different as well. In the individual rights cases, proportionality can function to bring law closer to justice – or to popular understandings of justice. But on structural issues as they arise in judicial decisions the questions of justice are always at a certain remove from the issues before the court. Applying proportionality to determine a federalist balance of power issue may or United States v Lopez 514 US 549 (1995). United States v Morrison 529 US 598 (2000). 58 As I have noted earlier, whether one’s conduct is being regulated by a state, or by the Federal Government is not likely to make that much of a difference to the private entity or person who challenges the law; their litigation interest is in defeating the claim to power of the government under whose criminal statute they are prosecuted. Jackson, Being Proportional (n 46). 59 As Professor Lorraine Weinrib has argued her vision of post-World War II constitutionalism: ‘Postwar rights-protecting constitutions do not . . . authorize the judiciary to treat rights as absolute negations of otherwise plenary state authority. Nor do they simply transfer to the courts the political power or prerogatives of elected representatives. They presuppose limited government with separation of powers that maximizes the complementary institutional strengths of legislatures, the executive and the courts. Elected representative bodies continue to act as responsible policy-makers, both empowered and disciplined by the constitutional instrument. The executive acts in compliance with the rule of law’. L Weinrib, ‘Constitutional Conceptions and Constitutional Comparativism’ in VC Jackson and M Tushnet (eds), Defining the Field of Comparative Constitutional Law (Westport, Praeger, 2002) 16–27. Although it may be that issues of authority are dealt with within formal proportionality doctrine, as part of the initial predicate, in Canada for example, of being ‘authorized by law’ as the first step in Section 1 analysis, the concept that is doing the work here is not the concept of proportionality. 60 cf R Alexy, ‘Constitutional Rights, Balancing and Rationality’ (2003) 16 Ratio Juris 131, 131 (noting ‘two classes’ of norms in modern constitutions, the first being those that ‘constitute and organize legislative, adjudicative and administrative power, the central theme of which is ‘empowerment’, as distinct from the second class of norms that ‘constrain and direct public power’, of which ‘constitutional rights are the most prominent’). These classes are not quite so distinct, insofar as norms that empower might be understood at once also to constrain. The constraint arising out of the limits of a granted power is what I refer to above as authority or authorization. 56 57
220 Vicki C Jackson may not advance ‘justice’, which is achieved through the actions of governments empowered to act rather than in the determination of what power the government units have; or to put it differently, the determination of which level of law controls is only con tingently connected to questions of justice, and may vary from level to level with the politics of the day. Proportionality analysis may achieve ‘balances’ between con stitutional interests of executive and legislative, or between the centre and periphery, but whether these correspond to a widely held sense of justice (or even of how good government operates) is a real question. If unanchored by ties to widely held conceptions of justice, does application of proportionality in the context of allocation of powers offer too broad a lever for judicial discretion? If courts cannot be understood as bringing law closer to ‘justice’ in applying pro portionality analysis to the resolution of power conflicts among levels or branches of government, the question arises, how should we understand what this tool does? I suggested earlier the possibility that instead of providing bridges to popular socio-legal understandings of morality and fairness, proportionality in the setting of institutional or power allocation questions may bring law into greater alignment with the more governance-informed understandings of juridical communities. Some other alternatives: can we think of proportionality – in its more complex doctrinal formulations – as a kind of de-biasing tool, one designed to advance beyond the mere pragmatism of consequentialism, to discipline a judge’s first intuitions about the best way to resolve intergovernmental power conflicts, by offering a structured set of enquiries? Can we think of it as a good governance tool, designed to promote reasoned decision-making generally by all levels and branches of government? To promote a ‘pro-constitutional’ or ‘pro-federal’ attitude among the different parts of governments with such enquiries? And, to come back to Professor Cohn’s chapter, to what extent does the deployment of proportionality as a general principle of analysis in public law impact its doctrinal form, or its application, on the individual rights cases with which it may in other systems be uniquely associated? Or do the courts, while deploying a common rhetorical rubric, apply the doctrine with greater or lesser degrees of stringency based on the type of constitutional issue and, as among rights, the perceived fundamentality of the rights? As good scholarship does, these chapters provoke questions like this for further reflection and research. To circle back to my starting comments, it is important to recognize the complexity of the very categories by which we demarcate our research agendas – whether they be constitutional law, globalization, or proportionality. And in thinking about these concepts, a couple of words of caution. First, one must be careful not to assume that the adoption of the same doctrinal formulations – as shown in Professor Cohn’s chapter – necessarily means that the doctrine will operate in the same way, or that application of the doctrine will or should come to the same result in different polities. As Dieter Grimm has shown, the Canadian and German proportionality formulae are applied with emphasis on different parts of the doctrinal test, with the Canadian Court focusing more on the ‘minimal impairment’ part of the analysis and the German Court more on the third step of analyzing the proportional benefits and burdens of the challenged law.61 As to differences in application, consider whether the reasoning of the ECtHR in the Turkish veil ban case does not invite the possibility of a different result in another kind See Grimm (n 30) 387–88.
61
Constitutional Law in an Age of Globalization 221
of European state. In the course of its reasoning, the Court referred to the Turkish Government’s concern about the presence in Turkey of ‘extremist political movements . . . that seek to impose on society as a whole their religious symbols and conception’ as a reason that helped justify Turkey’s very rigid insistence, at that time, on the secular appearance of persons in university spaces.62 If other European states without such extremist movements adopt veil bans, a different judgment might, in theory, follow.63 The second caution is a response to a question that may be implicitly running through all of the chapters in this book, and that is the possibility of the comparative study of constitutions and constitutional law yielding insights into what we might call trans national constitutional principles, rules, or doctrine.64 Scholarly attention to the possibility of what David Law has called ‘generic constitutional law’65 has grown in recent years, as has the interest more generally in claiming ‘constitutional’ status for international and supranational legal regimes.66 The caution is this: constitutions draw much of their legit imacy, their gravitas and their roots, from democratic acceptance. This is not the only source of the legitimacy, or weight, of domestic constitutions; a triad of democratic consent or acceptance, of good and/or just principles (which might be thought to include equality and liberty, decision-making through fair procedures and public reason, and independent courts), and of the rule of law, are at the heart of commitments to constitutionalism. Transnational practice can be of real value in helping to understand the most important values that should inform all constitutions, as well as those values or principles 62 Sahin v Turkey (2005) 44 EHRR 5, para 115 (Grand Chamber quoting from prior Chamber’s decision). The ECtHR’s acceptance of this particular justification was subject to harsh critique, in the dissent, which suggested that secularists in Turkey, including Sahin, support religious freedom to veil (Tulkens J, dissenting). For subsequent developments in Turkey, see VK Vojdik, ‘Politics of the Headscarf in Turkey: Masculinities, Feminism and the Construction of Collective Identities’ (2010) 33 Harvard Journal of Law and Gender 661 (citing and discussing 2008 Turkish Constitutional Court decision, E 2008/16, K 2008/116 (22 October 2008, Official Gazette No 27032). The June 2008 decision of the Turkish Constitutional Court, holding unconstitutional an amendment to the Constitution designed to allow women to wear veils in universities, did so under Arts 4 and 2 of the Constitution, which together provide that the requirement that Turkey be a ‘democratic, secular and social state’ is not subject to amendment. See AU Bali, ‘The Perils of Judicial Independence: Constitutional Transition and the Turkish Example’ (2012) 52 Virginia Journal of International Law 235, 253– 54 and fns 59, 61 (also suggesting that the constitutional Court in effect reinterpreted the scope of its own powers under Art 148 of the Constitution from procedural to substantive review of constitutional amendments). Such a decision complicates the constitutional process of political response. Notwithstanding the Constitutional Court’s 2008 decision, it was reported as of late 2010 that many universities have dropped the ban; the government in power then opposed the ban, and took steps to add judges to the Constitutional Court (presumably who would be more in sympathy with its views); and the opposition continue to disagree on whether the ban should be maintained. See, eg J Head, ‘Quiet End to Turkey’s College Head Scarf Ban’ BBC News, 31 December 2010: www.bbc.co.uk/news/world-europe-11880622; J Head, ‘Referendum Result Fails to Mask Turkey’s Divisions’, BBC News, 13 September 2010: www.bbc.co.uk/news/world-europe-11288360. However, the situation remained in flux as of early 2011. See, eg JN Couvas, ‘Headscarf Returns to Trouble Turkey’, Inter Press Service, 31 January 2011: ipsnews.net/news.asp?idnews=54292 (stating that in January 2011, the Council of State ruled that women could not wear headscarves during university exams). 63 But cf Dogru v France (2009) 49 EHRR 8 (rejecting a challenge to a French school requiring a Muslim girl to remove a headscarf for physical education classes). 64 I should make clear that none of these chapters argue for the adoption of any particular transnational normative framework. Rather, it is an implicit question that might arise in response to these chapters: must a ‘transformative’ constitution include the normative elements identified in Professor Cohen-Eliya’s chapter? In Professor Porat’s chapter, there is a sceptical normative question about the risks of ‘global constitutionalism’. And readers of Professor Cohn’s chapter might well be inspired to ask whether ‘proportionality’ is a universal approach, as a concept or in its particular doctrinal formulation? 65 D Law, ‘Generic Constitutional Law’ (2005) 89 Minnesota Law Review 652. 66 See, eg JL Dunoff and JP Trachtman (eds), Ruling the World? Constitutionalism, International Law, and Global Governance (Cambridge, Cambridge University Press, 2009), reviewed in Jackson, ‘Paradigms’ (n 24).
222 Vicki C Jackson that can but need not be pursued consistent with having a constitutional democratic state. It can also offer insights on a range of doctrines or institutions, and their possible con sequences. At the same time, transnational practice might have a ‘dark side’, as when transnational consensus or rule develops that is nonetheless antithetical to other valuable normative demands of domestic constitutions.67 To think of the lessons of transnational comparison as leading to transnational legal doctrine in constitutional law – in the sense of a ‘generic’ set of analytical tools, or, more ambitiously, a unified or harmonized body of legal rules, enforceable by courts – could threaten the other parts of the constitutionalist triad – democratic consent, and the rule of law. On proportionality, I have argued that as a structured set of questions, it is a highly useful tool for constitutional analysis in some areas,68 across a wide range of constitutions, but cannot supplant other kinds of questions, in particular, of democratic legitimacy and legal authority. And, I suggest, it is premature at best, and possibly inconsistent with the democratic premises of constitutionalism, to try to develop definitive normative frameworks for constitutional law at the transnational level; rather, the focus should be on the explication of transnational constitutional values, recognizing a range of approaches towards their achievement and a range of real differences in how those concepts are applied, whether one is considering the value of independence in judging, or how to accommodate tensions between values of liberty and of equality.
67 See, eg KL Scheppele, ‘Other People’s Patriot Acts: Europe’s Response to September 11’ (2004) 50 Loyola Law Review 89 (raising concerns about UN Security Council Resolution 1373 for failing to comply with fundamental due process norms). 68 See Jackson, Being Proportional (n 46) 842–57.
15 Constitutional Proportionality: (Appropriate) Guidelines* MORDECHAI KREMNITZER
I. INTRODUCTION
P
ROPORTIONALITY IS PROBABLY the most important principle in Israeli law in general and in public law in particular. The Israeli approach to proportionality originated in administrative law. With the adoption of the Basic Laws concerning human rights – Basic Law: Human Dignity and Liberty, and Basic Law: Freedom of Occupation – the principle of proportionality, together with its subtests, was ‘imported’ into Israeli constitutional law. Proportionality is currently applied in many fields of law, and its development in the constitutional context has left its mark upon those fields as well.1 As a central component of the Limitation Clause of the Basic Laws, proportionality constitutes a normative framework for establishing the extent to which constitutional rights can be limited by Knesset (Israeli Parliament) legislation and other governmental actions.2 Constitutional proportionality is formally anchored in the Limitation Clause, which provides that an infringement of a constitutional right can be justified only when made by a law befitting the values of the State of Israel, enacted for a proper purpose, and to an extent no greater than is required, or by regulation enacted by virtue of express authorization in such law.3
The rules of proportionality define and give practical effect to the requirement that the infringement be ‘to an extent no greater than is required’. * This chapter is based on a more comprehensive article to be published in a book on proportionality by the Israel Democracy Institute. I thank Roey Sasson for his dedicated assistance. 1 For a comprehensive survey of the history of proportionality in Israel and the world, see A Barak, Proportionality – Constitutional Rights and their Limitations (Tel-Aviv, Nevo Publishing, 2010) 225–62 (in Hebrew). Proportionality in Israel developed in a manner resembling its development in German Law, where it developed out of the examination of acts pertaining to police activity. See L Hirschberg, Der Grundsatz der Verhältnismäßigkeit (Göttingen, Schwartz, 1981) 2–3; see also, K Stern, Das Staatsrecht der Bundesrepublik Deutschland Band III/2: Allgemeine Lehren der Grundrechte, 1st edn (Munich, CH Beck, 1994) 766–67. 2 See, eg HCJ 4330/93 Ganam v Israeli Bar Association 50(4) PD 221 [1996] para 10 of Court President Barak’s opinion (in Hebrew): ‘Any governmental action must be proportionate. No administrative act may infringe values worthy of protection in excess of what is necessary . . . these general principles – which apply to all administrative acts – will naturally also apply to secondary legislation’ (emphasis added). 3 Basic Law: Human Dignity and Liberty, s 8; Basic Law: Freedom of Occupation, s 4.
226 Mordechai Kremnitzer The prevailing view is that the proportionality principle is composed of three subtests, developed to give concrete meaning to the vague phrasing of the Basic Law. The first is the rational connection test (suitability test), according to which there must be a rational connection of suitability between the proper purpose (the objective) and the arrangements instituted for its realization (the means); the second is the test of the least restrictive means (necessity test), according to which the means chosen for realizing the purpose may infringe a basic right only if they impair the right as little as possible; the third is the test of proportionality stricto sensu, according to which there must be an appropriate relationship between the benefit gained from realizing the objective and the effect caused by the restriction of the constitutional right. These tests constitute a broadly accepted general framework for examining proportionality. Nevertheless, it is questionable whether this structure provides enough guidance to both decisionmakers and those who check their decisions.4 In what follows, I will consider the problems attendant to the application of proportionality in Israel, from the perspective of the rule of law as opposed to the rule of men. The central criticism directed at the rules of constitutional proportionality in Israel is the absence of sufficient guidance. This problem is rooted in the broad discretion given to the courts, especially in regard to the value-based nature of the third test.5 In fact, it would seem that the entire edifice of proportionality rests on the shoulders of the third test – proportionality stricto sensu – while diminishing the status of the two other tests, which have come to be deemed ‘pro-forma’ preconditions. But are the shoulders of this test broad enough to bear the weight of this heavy burden? Furthermore, it is not without significance that the third test is applied only after the law or governmental act have successfully passed the first two tests. It creates the impression that most of the path to legitimizing the means has been successfully traversed. In other words, the means already merit a considerable measure of legitimacy. This problem is aggravated by the fact that the legitimacy of this final stage itself is open to question. The third test requires difficult value-based decisions that are characteristically subject to the influence of the personal perception of the judges. To be more precise, we are not dealing with value judgements at a high level of abstraction where the scope of disagreement typically narrows, but rather with value judgements of a concrete character, where the potential for dispute increases. The judicial decision therefore becomes largely dependent upon the specific panel of the court. It becomes, ‘rule of men’. This stage (the third test) is especially exposed to the claim that it transfers power from an elected body that better expresses the public’s preferences, to the judicial
4 See D Dorner, ‘Proportionality’ in A Barak and C Berenson (eds), Berenson Book, vol 2, 2nd edn (Tel-Aviv, Nevo, 2000) 281, 283: ‘In my view, notwithstanding the sophistication and the complexity conferred by the three aforementioned subtests to proportionality, the subtests are insufficient and additional refinements are required’. 5 Not surprisingly, in HCJ 7052/03 Adalah v Minister of Interior 61(2) PD 202 [2006] (in Hebrew), para 107 of Justice Cheshin’s opinion, where he referred to this form of proportionality in the moral sense: ‘There are three subtests in the test of proportionality, and for reasons that I do not understand the third subtest is termed the test of proportionality “in the narrow sense”. This name is a mystery to me . . . the third subtest before us, the test in which we place on each pan of the scales the values that conflict with one another, the benefit values against the damage values, ought to be called the test of proportionality “in the moral sense”. This test is concerned with values, and it should therefore be given that name’.
Constitutional Proportionality 227
branch, which does not reflect those choices.6 Indeed, the infinite flexibility of the third test enables any legal conclusion to be reached, and then justified in terms of compliance with the tests. The greatest danger is that if a problematic measure passes the first two tests without difficulty, it will also pass the third stage. The stamp of approval granted following the constitutional examination confers ‘strong legitimacy’. When such approval is granted to an improper measure, the normative distortion is conspicuous and grave. Injustice in practice is preferable to an unjust practice that has been formally approved.7 In a post-modern era it is an illusion to assume that the institutional authority of the court by itself is sufficient to grant the decisions of the court real legitimacy and authority. Moreover, very little attention was paid to the role of proportionality from the perspective of the decisionmakers themselves. It is obvious that in order to achieve compliance with the requirements of proportionality, proportionality has to become a working tool in the hands of decisionmakers. The existing format of the tests does not provide sufficient guidance and clarity needed to make them useful devices for decision-making. This study is a preliminary examination of various possibilities for a more nuanced, application of the constitutional tests. It aims at ‘empowering’ (instilling more meaning) in the tests that are ahead of proportionality stricto sensu (the third test). It aims also at a more structured third test. The research method includes, inter alia, learning from the experience of others, specifically from the development of proportionality in Canadian and South African law, and the analysis of rulings from both ends of the spectrum – case law that failed to protect human rights as well as case law that successfully completed that task – with an eye towards drawing appropriate conclusions. The study focuses specifically on the tension between security demands (particularly in the context of confronting terrorism) and other human rights. In part II I will begin with the requirement of an appropriate purpose. Then, in part III, I will examine each of the subtests of proportionality. In part IV, I will discuss briefly the relationship between the proportionality tests themselves. In part V, I will end the chapter with concluding remarks. 6 Adalah v Minister of Interior (n 5) paras 107–15 of Court President Barak’s opinion; see also R Gavison, ‘The Constitutional Revolution: Reality of a Self-Fulfilling Prophecy?’ (1997) 28 Mishpatim 21, 64–68 (in Hebrew); A Marmore, ‘Judicial Review in Israel’ (1997) 4 Law and Government 133, 142–44; B Pieroth and B Schlink, Grundrechte Staatsrecht II, 25th edn (Heidelberg, Müller (CFJur), 2009) 71; N Siebrasse, ‘The Oakes Test: An Old Ghost Impending Bold New Initiatives’ (1991) 23 Ottawa Law Review 99, 107: ‘This kind of costbenefit balancing is what would traditionally be described as a matter of policy, not law’. 7 Jackson J expressed this point eloquently in his minority opinion in the infamous Korematsu case: Korematsu v United States 323 US 214, 245–46 (1944): ‘A military order, however unconstitutional, is not apt to last longer than the military emergency. Even during that period, a succeeding commander may revoke it all. But once a judicial opinion rationalizes such an order to show that it conforms to the Constitution, or rather rationalizes the Constitution to show that the Constitution sanctions such an order, the Court for all time has validated the principle . . . The principle then lies about like a loaded weapon, ready for the hand of any authority that can bring forward a plausible claim of an urgent need. Every repetition imbeds that principle more deeply in our law and thinking and expands it to new purposes . . . A military commander may overstep the bounds of constitutionality, and it is an incident. But if we review and approve, that passing incident becomes the doctrine of the Constitution. There it has a generative power of its own, and all that it creates will be in its own image. Nothing better illustrates this danger than does the Court’s opinion in this case’. The majority opinion in this case demonstrates Jackson’s point. Even though Black J, who wrote the majority opinion, used the most rigorous and meticulous approach in the American system, of ‘strict scrutiny’, he nonetheless concluded that the measure was constitutional.
228 Mordechai Kremnitzer
II. THE APPROPRIATE PURPOSE TEST
Any infringement of a right must serve an appropriate purpose (also referred to as ‘sufficient objective’ or ‘legitimate purpose’). It is quite clear that a constitutional right cannot be infringed in order to realise an illegitimate purpose, or even to realise a marginal or negligible objective. As a matter of law, the requirement of an appropriate purpose also includes the condition of consistency with the values of the state, as a Jewish and democratic state. Although the purpose is a threshold requirement in most legal systems,8 it does not appear to constitute an important component of the constitutional examination. The accepted view is that the appropriate purpose test requires, initially, the identification of the purpose.9 The central question arising in this context is whether the stand ard for identifying the purpose is an objective one or a subjective one, insofar as there may be a discrepancy between the two criteria in a given case. Comparative law does not offer a standard answer to this difficult question.10 Nevertheless, it would seem that the prevailing view in Israel supports the objective approach.11 Professor Barak recently suggested that a purpose should be considered appropriate only if, cumulatively, both the objective and the subjective purposes are found to be appropriate.12 This presents a reasonable and wise solution, especially – the inclusion of the subjective purpose. The gist of the argument is this: identifying the subjective purpose is preferable to identifying an imaginary purpose, that may serve the underlying goal of ‘purifying the impure’, or of concealing the real purpose.13 The subjective purpose – being the real 8 For a survey of the purpose requirement in Canada, see PW Hogg, Constitutional Law of Canada, 5th edn (Toronto, Carswell Legal Publishers, Student edn, 2009) chs 32–38; see also S Woolman and H Botha, ‘Limitations’ in S Woolman et al (eds), Constitutional Law of South Africa, 2nd ed (Juta, Cape Town, Pretoria University Law Press, 2006) 73–79. On the position in Germany, see D Grimm, ‘Proportionality in Canadian and German Constitutional Jurisprudence’ (2007) 57 University of Toronto Law Journal 383; and see Barak (n 1) 317–19, 345–52. 9 Hogg (n 7) 19–22; Woolman and Botha (n 8) 73–75; Barak (n 1) 359–67. 10 For a comparative survey of this subject, see Barak (n 1) 359–67. 11 CA 6821/93 United Mizrahi Bank Ltd v Migdal Cooperative Village 49(4) PD 221 [1995] para 81 of former Court President Shamgar’s opinion (in Hebrew). See also in the Menahem case, HCJ 4769/95 Menahem v Minister of Transport 57(1) PD 235 [2002] para 16 of Justice Beinisch’s opinion (in Hebrew): ‘As a rule, the examination of the purpose of a legislative act in accordance with the limitation clause does not focus on examination of the legislature’s motives, even though prima facie there is a certain overlap between the purpose of the law and its motive; rather, it focuses on the objective purpose of the law’ (emphasis added). 12 Barak (n 1) 368–69. 13 See, eg in Adalah v Minister of Interior (n 5) and the observations of some of the justices that the security considerations of the state may have been intended as a front for the demographic consideration, which is unacceptable. See, eg para 11 of Justice Procaccia’s opinion: ‘First, we must examine the degree of credibility of the claim concerning “security needs”. We must ascertain whether the security considerations that have been raised are not being used, in reality, as a cloak for other completely different purposes which are really the purposes that underlie the legislation containing the violation of the right’. See also para 24 of Justice Joubran’s opinion: ‘depriving the Minister of the Interior of discretion, ab initio, to examine the possibility whether citizenship should be given to any of the residents of the territories in order to realize the right of an Israeli citizen to family life, by ignoring the specific circumstances of the case, raises the concern whether the security consideration is not the only consideration underlying the enactment of the law and it raises questions with regard to the policy that this law wishes to achieve. This concern becomes even greater if we survey the legislative history that led to the enactment of the law, which, whether in a concealed or express manner, associates the law with the government’s demographic policy . . . Similarly, throughout the legislation process, it is possible to find remarks made by Knesset members and members of various Knesset committees, from various parties, who address the demographic policy that the law implements’.
Constitutional Proportionality 229
purpose – is frequently, albeit not always, more accessible.14 The proposed amendment to the Citizenship Law, 5712-1952 can serve as an example. The proposal is to add to the existing statement of allegiance to the State of Israel and its laws, allegiance to the State of Israel as a Jewish and democratic state. To disregard the patently real motivation underlying the law – mainly, hostility towards Israel’s Arab population – falsifies reality and transforms the Court into an instrument in the hands of other authorities, thus undermining its status as the central, and perhaps the only, restraining force in the democratic system for protecting minorities against the tyranny of the majority.15A similar analysis is applicable to other recent laws directed against the Arab citizens of Israel.16 Special care is required to avoid a distorted reading of the purpose, intended to legit imate improper means. For example, it is not difficult to discern that the real purpose of demolishing houses is to punish the individual offender and his family (being a means intentionally directed at causing suffering to the dwellers of the demolished house), and as a by-product, also deterrence for the many, and that it is not intended solely as a means of deterrence, as was determined by the Court, in order to legitimate the means.17 Even though the legislature has discretion in determining the degree of abstraction of the purpose, the court should encourage legislation in which the legislative purpose is expressed in as concrete terms as possible. Two reasons will be mentioned. First, open ended terms, apart from being ‘lazy’ and vague,18 may also serve as a convenient means for concealing a ‘covert’ and illegitimate purpose. Second, from the authorities’ (legislative and executive) perspective, the determination of a clear purpose is the necessary starting point for establishing an appropriate norm and the adoption of an effective measure. A more precisely defined purpose improves the decision-making process and strengthens its rational basis.
14 One cannot ignore the practical difficulties that arise in the attempt to discern the subjective purpose, particularly in a complex body such as the Knesset with its many members and opinions. Nevertheless, these difficulties are not insurmountable. The subjective purpose can be identified in a number of ways, for example the legislative history, including the explanatory notes of the law, the Knesset Proceedings and the debates in the committees, a report of a professional committee when the law is the product of its work, and others. In private draft proposals, special attention should be given to statements made by the initiators of the law. Nor can one ignore the circumstances that gave rise to the enactment of a law or the implementation of the harmful measure, and in certain cases it may also be appropriate to take affidavits from the serving holders of positions. See also Barak (n 1) 369–71. 15 See HCJ 6427/02 Movement for Quality Government v Knesset 61(1) PD 619 [2005] para 15 of Justice Grunis’s opinion (in Hebrew): ‘The principal justification for the existence of judicial review of legislation is the need to protect the minority and individuals from the tyranny of the majority. The court is the last barrier capable of preventing the majority harming individuals and minority groups’. Evidently, one of the courts into which this approach has filtered deeply is the Constitutional Court of South Africa, see n 52 below. 16 M Kremnitzer and S Krebs, ‘From Illiberal Legislation to Intolerant Democracy’ (2011) 26 Israel Studies Review 4. 17 See HCJ 9353/08 Abu Dheim v GOC Central Command, IDF (5 June 2009), Nevo Legal Database (by subscription) para 8 of Justice Naor’s opinion (in Hebrew): ‘Over many years the court has recognized that the use of this measure is intended for deterrence – to deter and not to punish or to avenge. In the past this court has avoided disputing the evaluation of the security authorities with respect to the effectiveness of the deterrence’ (emphasis added). For a critique of this view, see M Kremnitzer and T Hoernle, ‘Human Dignity and the Principle of Culpability’ (2011) 44 Israel Law Review 115, 128–31. 18 This is attested to by the divergent interpretations of the term ‘state security’ in the various cases, such as in the further hearing in the matter of the Lebanese bargaining chips, see CrimFH 7048/97 John Does v Minister of Defence 54(1) PD 721 [2000] (in Hebrew).
230 Mordechai Kremnitzer III. THE PROPORTIONALITY TEST (‘TO AN EXTENT NO GREATER THAN IS REQUIRED’) AND ITS SUBTESTS
The requirement of ‘to an extent no greater than is required’19 diverts the examination from the purpose to the means chosen by the state for realizing it.20 Three subtests are recognised – as mentioned above. A. Preliminary Remarks Before considering the proportionality tests themselves, a question arises pertaining to the factual and conjectural background that serves the court in applying the tests. If, in practice, the court trusts the platform presented to it by the state blindly21 – and this platform is quite frequently tainted by natural biases that will be discussed below – the value of the constitutional examination is significantly reduced. Furthermore, if the court is prepared to accept and rely upon unsubstantiated evaluations, and does not order the submission of data in order to examine it, then why should the authority bother to base its decision-making on data?22 The difference between the majority and minority opinion in the Korematsu23 case demonstrates the difference between a blind faith approach and a critical approach. If our concern is with incentives, then the judiciary can and should provide an incentive for the legislature to act in an appropriate manner. As such, the preparatory work that precedes legislation should be regarded as a crucial indicator of the quality of the Basic Law: Human Dignity and Liberty, s 8; Basic Law: Freedom of Occupation, s 4. Adalah v Minister of Interior (n 5) para 64 of Court President Barak’s opinion: ‘The requirement that the violation of the legislation be “to an extent that is not excessive” focuses on the means that the legislator chose. A law that infringes a constitutional human right is proportionate only if it maintains a proper relationship between the proper purpose that the law wishes to realize and the means that it adopts to realize that purpose . . . a proper means is a proportionate means. A means is proportionate if the law’s violation of the protected right is to an extent that is not excessive’. 21 See the comment of Supreme Court Justice (Ret) Itzhak Zamir in his article – I Zamir, ‘Human Rights and State Security’ (1988) 19 Mishpatim 17, 36–37 (in Hebrew): ‘the statements made in judgments, and even more so, the decisions given in judgments, lead to the conclusion that in practice the courts hesitate in intervening in the discretion of the security authorities in security matters, more so than with respect to their discretion in other matters . . . it could be said, by way of generalization, that the court tends to show a special measure of consideration and a great degree of restraint with respect to the security authorities because of the heavy responsibility, the gravity of the consequences, and the public sensitivity attaching to security matters’. This position was actually expressed in many of the court’s rulings. See, eg HCJ 1661/05 Hof Aza Regional Council v Knesset 59(2) PD 481 [2005] para 108 of the majority ruling (in Hebrew): ‘The position of this court since its inception has been that a presumption operates in favor of reliance on the security experts on behalf of the State. Weighty evidence is required in order to refute this presumption’. See also HCJ 6026/94 Nazal v IDF Commander in Judea and Samaria 48(5) PD 338 [1994] (in Hebrew); HCJ 7473/02 Bahar v IDF Commander in Judea and Samaria 56(6) PD 488 [2002] (in Hebrew). One of the more illuminating examples in this context is the Abu Dheim ruling (n 17). In that case the Court accepted the position of the authorities regarding the deterrent effect of the demolition of houses, without it having submitted any concrete evidence to that effect, and despite the existence of a comprehensive, conflicting report. For a critique of the ruling, see M Kremnitzer, ‘The Legitimacy of Demolishing Terrorists Houses – Comment on Ruling, in the Wake of the Judgment in the Matter of Hisham Abu Dheim v GOC Homefront Command’, Jerusalem, The Israel Democracy Institute, 24 February 2009: www.idi.org.il/breakingnews/pages/breaking_the_news_94.aspx. 22 It appears that the fundamentally weak claim that the High Court of Justice lacks the tools for confronting a complex factual framework, almost entirely collapses when considering the importance that the law attaches to disclosure of the truth in all matters relating to protection of human rights. 23 Korematsu (n 7). 19 20
Constitutional Proportionality 231
legislative result. This preparatory stage includes the collection of relevant data, meaningful procedures of examination and clarification, including the hearing of experts’ views and having a professional discussion, the presentation of alternatives, and a stage of consideration leading to a reasoned choice of the preferred measure.24 As a general rule, the more profound and transparent the internal proceedings of the authority, the less suspicion and the greater respect will be shown towards the means adopted during the judicial review. However, according to this same approach, decision-making that snatches an opportunity or that lacks transparency – warrants suspicion regarding their purpose and the means chosen for their realization. Another preliminary subject – before proceeding to the subtests themselves and to the concrete proposals – relates to the constitutionality of means that were found to be illegal under international law or the laws of comparable nations. There is no justification for a measure, found to be illegal in international law, to be subject to a fully-fledged constitutional examination in Israel. When a measure is unusual from a comparative perspective, and especially when it was declared unconstitutional in comparable legal systems, it should be regarded as prima facie ‘suspect’ of being unconstitutional here as well.25 In what follows, the subtests of proportionality will be briefly examined. B. The Rational Connection Test (Suitability Test) The rational connection test requires that the means chosen be capable of realizing the underlying purpose of the law.26 An additional condition posited by the rational connection test and adopted in the case law over the years is the prohibition against the use of a measure that is arbitrary, unfair or illogical.27 Accordingly, for example, in the Morar case an order to prevent access of Palestinian farmers to their agricultural plots in order to protect them and ensure their security against attacks by Jewish-Israeli lawbreakers was overturned. The
24 It seems that this approach was correctly implemented in a recent judgment handed down by the Supreme Court in which it partially cancelled a planning decision (the paving of a road) based on the failure to give sufficient consideration to environmental factors and alternatives that would be commensurate with their appropriate weight. The Court carefully examined the decision-making process and, inter alia, related to the duty to consider alternatives as part of the desideratum of receiving the fullest possible factual picture (which was not fulfilled in the particular case) as central components of its decision. Although the issue concerned an administrative decision as distinct from primary legislation, its internal logic is applicable for our purposes too. See HCJ 2920/94 Adam Teva Va Din v National Committee for Planning and Construction of National Infrastructures 50(3) PD 441[1996] (in Hebrew). 25 Eg in relation to the measure of collective punishment, expressed in the imposition of sanctions against persons who are not offenders (such as demolition of houses). This measure contravenes the fundamental conception of the rule of law, whereby the imposition of sanctions must be based on personal liability. The universal norm too, firmly rooted in international law, similarly prohibits collective punishment. See Geneva Convention Relative to the Protection of Civilian Persons in Time of War (Fourth Geneva Convention) (adopted 12 August 1949, entered into force 21 October 1950) 75 UNTS 287, Arts 27 and 33. Based on this it can and should be argued that collective punishment is a fundamentally illegitimate measure which cannot be suited to the realization of any purpose, even if appropriate. 26 See Barak (n 1) 373–74. See also references in fns 2–4 there. 27 See, eg Menahem (n 11) para 23 of Court President Beinisch’s opinion; HCJ 9593/04 Morar v IDF Commander in Judea and Samaria 61(1) PD 844 [2006] para 25 of Justice Beinisch’s opinion (in Hebrew); HCJ 2887/04 Abu-Medigam v Israel Land Administration 62(2) PD 57 [2007] (in Hebrew).
232 Mordechai Kremnitzer measure was held to be fundamentally unfair and actually rewarding lawbreakers.28 In Professor Barak’s view, the requirement concerning fairness does not belong to this category, because the unfairness of a measure does not necessarily detract from the rationality of its connection to the goal, and it is more reasonable to view it as detracting from the requirement of an appropriate purpose or the third proportionality test. Professor Barak expressed a similar position on the subject of an arbitrary measure, which in his view would only be annulled if it impaired the effectiveness of realizing the goal, and not by force of being arbitrary as such.29 It seems that this position overly limits the scope of the test. Indeed, any effective measure, even if arbitrary or unfair, has a connection to its goal, but this connection is not a rational connection, but rather a distorted and irrational one. The effectiveness of the measure should not be the sole parameter.30 Abolishing such measures, which should not have been adopted in the first place, at the earliest stage possible, is also important from an educational perspective. Furthermore, sometimes the more ‘outrageous’ the measure – the more effective it may be in comparison to other more moderate means, and as such may also pass the least restrictive means test, which requires that the alternative means realise the goal to an identical extent. As mentioned earlier, passing these two tests is no trifling matter because it confers a certain degree of legitimacy to the measure. It was for this reason that this requirement formed an integral component of the rational connection test in the Canadian Oakes case, which is the cornerstone of the development of the stages of constitutional examination.31 Even if the Israeli Court overlooked this additional condition at first (either intentionally or unintentionally),32 its adoption is both necessary and commendable. As mentioned in the introduction to this chapter, there are profound reasons for substantively strengthening the rational connection test, instead of automatically transferring all means to the next stage.33 What can be done in this respect? First, it is important to describe accurately the means under discussion in the concrete circumstances of the case. For example, the basic and logical foundation of administrative detention is the prevention of future harm, and as such it relies upon a threat assessment. In the case regarding the Lebanese detainees, the state itself conceded that they posed no danger, but rather that they were being held for purposes of negotiations.34 28 Morar (n 27) para 25 of Court President Beinisch’s opinion: ‘In these circumstances, the closure of the areas to the Palestinian farmers in order to contend with the aforesaid threat is not rational, since it is an extremely unfair act that results in serious harm to basic rights while giving in to violence and criminal acts’. 29 Barak (n 1) 378–80. 30 Court President Beinisch’s comments on this point in Morar (n 27), are particularly germane: ‘The rational connection test is not merely a technical causal connection test between means and end. Even when use of a certain measure is likely to lead to realization of the desired purpose, this does not mean that there is a rational connection between the means and the end and that the means is suited to achieving the end. The emphasis in the rational connection test is whether the connection is rational. The meaning of this is, inter alia, that an arbitrary, unfair or illogical measure should not be adopted’ (emphasis added). 31 See R v Oakes [1986] 1 SCR 103, 139 (Can): ‘The measures adopted must be carefully designed to achieve the objective in question. They must not be arbitrary, unfair or based on irrational considerations. In short, they must be rationally connected to the objective’ (emphasis added); see also Hogg (n 8) chs 32–38. 32 See, eg United Mizrahi Bank (n 11) para 95 of Court President Barak’s opinion; Adalah v Minister v Interior (n 5) paras 66–67 of Court President Barak’s opinion; Hof Aza (n 22) para 67 of the majority opinion. 33 For a different view of this text, see Barak (n 1) 387–89; Hogg (n 8) 34; Grimm (n 8) 389. 34 See John Does (n 18) para 2 of Court President Barak’s opinion, where he presented the facts with the observation that ‘There is no debate among the parties that the petitioners themselves do not pose a threat to national security’.
Constitutional Proportionality 233
Could it honestly be claimed that keeping them in administrative detention was rationally connected to the stated goal underlying this particular measure? Second, the test should be interpreted as requiring consideration also of the ‘negative effectiveness’ of the means. ‘Negative effectiveness’ refers to the undesirable results of the means apart from the infringement of the right under discussion (which is taken into account at the stage of examination of proportionality in the narrow sense).35 A meaningful examination of the rationality of adopting a particular measure cannot be based on the optimistic assumption that it has only positive effects (especially in the absence of the requirement that the measure realise its goal in full, and accepting that the probability of realization need not be high), nor can it rely on the naive disregard of the negative aspects involved in its utilization. For example, it is reasonable to presume that the demolition of houses or the use of torture as a means to reduce terrorist attacks may actually increase motivation to commit additional acts of terrorism, due to the humiliation and alienation experienced by the victims of such measures and those close to them, thus hindering the purpose in a long-term perspective. Use of torture also generates false information that is liable to mislead the system and thus prevent efficient prevention.36 A full answer to the question of the rationality of resorting to the particular means must include consideration of the negative effects as well. As opposed to the first two tests, which involve utilitarian balancing, the third test is seen as addressing values. From this perspective, too, it is essential that the negative factors be weighed as part of the first test. Ignoring the negative implications at this stage creates the misleading impression that from an efficiency perspective, as opposed to a value-based assessment, the measure passed the judicial critique. It should also be noted that this approach is consistent with the logic that should guide rational governmental decision-making within the authority itself. It is clear that a decisionmaker betrays her duty if she examines only the positive aspects of the means while ignoring their negative consequences.37 When the constitutional examination ignores the negative aspects at precisely the stage that they should be examined, it encourages the same kind of disregard by the legislature and the executive. Furthermore, such an approach produces a substantive distortion, both in the decision-making process and in the process of constitutional examination, by giving the advantages of the means a position of primacy over its deficiencies. 35 See, eg HCJ 316/03 Bakri v Israel Film Council 58(1) PD 249 [2003] para 14 of Justice Dorner’s opinion (in Hebrew). This judgment discussed the constitutionality of the Film Censorship Board’s decision to prohibit the screening of Jenin, Jenin. Regarding the suitability test, the argument was that the prohibition of the film would prevent the exposure of its contents and would reduce the danger to public peace and security. It should be noted that in this case there was a dark side of censoring the film from the effectiveness perspective. It was clear to both sides that the prohibition would trigger a media storm, turning the film into a ‘symbol’ and hence actually increase its public exposure. The negative effectiveness of the measure also includes the probable creation of a ‘black market’ and an ‘incentive’ to illegally obtain the film via the Internet. On this matter in general, see J Waldron, ‘Torture and Positive Law: Jurisprudence for the White House’ (2005) 105 Columbia Law Review 1681; see also J Waldron, ‘Security and Liberty: The Image of Balance’ (2003) 11 Journal of Political Philosophy 191. 36 See M Kremnitzer, ‘Coercive Interrogation in Israel – The Process of Shaping the Norms’ in M Mahlmann (ed), Gesellschaft und Gerechtigkeit: Festschrift für Hubert Rottleuthner (Baden-Baden, Nomos, 2011) 164, 177. Moreover, the positive effectiveness of these measures is similarly questionable and at the very least is not necessarily self-evident, but the discussion of this topic exceeds the scope of this chapter. 37 In view of this analysis it should be remembered that there is no advantage in legislation ‘at any price’. Under certain circumstances, when the negative effectiveness of the measure is extensive, the authority would be well advised to avoid legislation altogether.
234 Mordechai Kremnitzer Third, it should be examined whether the means is aimed at promoting the core of the goal or only its periphery. Finally, the level of the efficiency of the measure should be examined in light of the importance of the relevant aspects of the infringed right. One view that has been expressed in the Israeli ruling was that there is no requirement for a definite contribution to the promotion of the entire goal. Rather, a partial fulfillment of the goal is sufficient.38 According to this view, in order to pass the test, there is no need, at this stage, to take into consideration the relevant aspects of the infringed right. This view is vague and arguably overly lenient. A more dominant approach calls for a relationship between the importance of the right and level of efficiency.39 This approach is preferable, provided that what is considered is not the right in abstracto but the relevant aspect of the right that is being infringed. If the level of efficiency is low and the aspect of right being infringed is highly important – it is not obvious at all that the measure should pass the test, and especially so if the measure is directed to the periphery of the goal. C. The Least Restrictive Means Test (Necessity Test) The least restrictive means test is based on a comparison of alternative measures. This stage seeks to examine whether, from among the various measures rationally connected to the (appropriate) goal, the measure chosen causes the least possible harm to human rights. It seems that the main problem of the necessity test stems from the requirement that the purpose be realised by the less harmful measure in an identical manner. It should, however, also be noted that on a number of occasions the case law has expressed a view that differs from this approach. In the Academic Center of Law and Business case,40 addressing the constitutionality of a law authorizing the establishment of a private prison, the Court required that the alternative measure realise the legislative purpose to a ‘similar’ degree. In this spirit I suggest to substitute the requirement of an identical degree of effectiveness with a requirement of a similar degree. It is almost impossible to measure comparative effectiveness in exact terms. Requiring identity means requiring the impossible. The practical meaning of insisting on identity is emptying the second test from content. On top of this, the Court’s approach indicates that even when less harmful means are available, the legislature has broad discretion and is under no obligation to resort to them even if they are expected to realise the purpose of the law to an identical degree.41 38 See HCJ 466/07 Gal-On v Attorney General (11 June 2012), Nevo Legal Database (by subscription) para 11 of Justice Joubran’s opinion (in Hebrew). 39 Barak (n 1) 627–28; HCJ 6298/07 Ressler v Knesset (21 February 2012), Nevo Legal Database (by subscription) paras 51–53, 55 of Court President Beinisch’s opinion (in Hebrew); HCJ 1715/97 Investment Managers Bureau v Minister of Finance 51(4) PD 367 [1997] paras 4 and 8 of Justice Dorner’s opinion (in Hebrew); see also Dorner (n 4) 289: ‘it suffices that there is an appropriate degree of probability that the act that infringes a protected right or interest makes a reasonable contribution to the attaining of the purpose’ (emphasis added). 40 HCJ 2605/05 Academic Center of Law and Business v Minister of Finance (19 November 2009), Nevo Legal Database (by subscription) para 49 of Court President Beinisch’s opinion (in Hebrew): ‘requires us to examine whether that less harmful measure realizes the legislative purpose to the same degree or to a similar degree as the measure chosen by the legislature’ (emphasis added). 41 See, eg CrimA 6659/06 A v State of Israel 62(4) PD 329 [2008] para 31 of Justice Beinisch’s opinion (in Hebrew): ‘Consequently this court has accorded recognition to “constitutional room for maneuver” which is
Constitutional Proportionality 235
This approach should be rejected. Accepting the approach means that the second test is not being considered seriously. Even when a measure does not stand up to the test – it may still be constitutional. If so, what value has the test? D. The Proportionality Test (Stricto Sensu) The proportionality test in the narrow sense establishes a requirement for an appropriate relationship between the expected benefit to be derived from the means chosen (the degree of expected benefit multiplied by the degree of probability of its materialization) and the damage caused by the limitation of the right. In contrast to the other tests based on the fulfillment of a particular requirement or compliance with a specific condition, this test calls for striking a balance among conflicting values.42 The infringement of the right and the realization of the purpose are placed on opposite ends of the scale. The balancing process must include the full gamut of relevant considerations. Not surprisingly, in the Korematsu case, such a comprehensive analysis was only conducted in the dissenting opinion. Considering the use of torture as a means – the overall implications of which are not limited to the grave consequences for its victims (which apart from the physical harm also include long-term psychological consequences and a degrading component), but also include the impact of a license for extreme state arbitrariness on the entire legal system and on society as a whole.43 The examination of proportionality should entail: the aspect (and sometimes aspects) of the right that is being infringed – its importance; the nature of the infringement: its level of severity and likelihood. Within this examination – attention should be paid, inter alia – to the degree of discretion granted to the authorities in the implementation of the law44 and to the impact of the victim’s identity or status on the severity of the infringement.45 A similar analysis relates to the expected benefit – the exact nature of this benefit, its importance, and the realistic likelihood of its occurrence. also called the “zone of proportionality” . . . This court will not substitute its own discretion for the criteria chosen by the legislature and will refrain from intervention as long as the measure chosen by the legislature falls within the zone of proportionality. The court will only intervene when the chosen measure significantly departs from the bounds of the constitutional room for maneuver and is clearly disproportionate’. 42 See, eg HCJ 10203/03 Hamifkad Haleumi v Attorney General 62(4) PD 715 [2008] para 55 of Justice Naor’s opinion (in Hebrew); Movement for Quality Government v Knesset (n 15) para 60 of Court President Barak’s opinion; Hof Aza (n 21) paras 67 and 114 of the majority judgment. In comparative law as well, the accepted approach is that the test is one of balancing. See on this matter Barak (n 1) 424–25 and the citations therewith. 43 See Waldron, ‘Torture and Positive Law’ (n 35) 1713–17, 1737–39. 44 In the Constitutional Court of South Africa this criterion merits serious consideration, eg in the Dawood case, Dawood v Minister of Home Affairs 2000 (3) SA 936 (CC) paras 47–48 (S Afr), the Court annulled the powers given to the immigration clerks (to refuse to extend temporary residence permits to foreign spouses of South African residents) because they lacked sufficient adequate guidelines or the possibility of judicial review of their decisions. On the other hand, in the Metcash case, Metcash Trading Limited v Commissioner for the South African Revenue Service 2002 (4) SA 317 (CC) (S Afr), the appointee’s power to make exceptions in certain cases made the infringement less abstract, subjected it to objective oversight, and moved the court to dismiss the petition. 45 Consideration of the victim’s position in determining the severity of an infringement can lead to a real change only if there is a sincere, genuine examination of the actual situation, with keen awareness to the concrete reality. The South African Constitution related in a number of situations to the victim’s status as an auxiliary indication. One of the most persuasive examples of this was in the Manamela case, S v Manamela 2000 (3) SA 1 (CC) (S Afr), in which the majority opinion, related to the practical implications of reverse-onus
236 Mordechai Kremnitzer Only an absolutely compelling expected benefit may justify a very severe infringement. Only a clearly compelling expected benefit may justify a severe (but not very severe) infringement. For the sake of assisting the process of balancing, it may be helpful to differentiate between different levels of the infringement caused by the measure and of the expected benefit. The theory of proportionality should be enriched through guiding considerations. Three will be mentioned: 1. It would be shortsighted not to learn from past experience, both national and universal. Certain matters are accepted universally, or almost universally, as judicial travesties, and as the beginning of the slippery slope from democracy to a totalitarian regime.46 For instance, treating ‘unpatriotic’ ideological criminals differently than patriotic ideological criminals as we witnessed recently in Israel.47 These must be seen as warning signs, that proclaim: Beware! A slippery slope before you! No less important is learning from commendable judgments, in which the courts stood up to the challenge.48 2. It is particularly important to avoid bias in the choice and the implementation of the measure. Here I refer primarily to the players who implement proportionality before the court, namely the legislative and executive authorities, but also to the courts themselves. It is essential that the actions of the authorities (and the evaluation of these actions by the courts) be conducted with full awareness of natural biases that may influence their actions.49 Particularly where it concerns actions against the ‘other’, it is important that the measure be generalised and examined from the perspective of it being wielded against ourselves. Additional natural biases which provisions on the accused in cases concerning possession of stolen goods: ‘The practical implications of this cannot be ignored. Many of these people are not likely to keep records of the wide variety of informal transactions that they conclude daily. They, and not the professional receivers, are the persons least in a position to discharge the onus and hence become the class most vulnerable to erroneous conviction precisely because of their disadvantaged position in society. Furthermore, because of their inability to afford legal representation, they will not be well prepared either to present their case to best advantage or to meet the cross-examination to come. The risk of people being erroneously convicted, subjected to the social stigma of conviction and unjustly sent to jail is unacceptably high’. 46 See, eg Korematsu (n 7) and the Hirabayashi case, Hirabayashi v United States 320 US 81 (1943), both dealt with the United State’s policy regarding Japanese citizens. Over the years a consensus has emerged that these rulings are particularly wrong. On collective sanctions, see n 6 above. 47 In examining legislation that distinguishes between those who are patriots and those who are not we can learn from the experience of the Weimar Republic. See the criticism on the proposal to pardon Jewish rioters in the context of the ‘Gaza disengagement plan’, and the difference in the treatment of Palestinian rioters, Y Cohen ‘What Distinguishes between the Violence of the Left and the Violence of the Right, or How the State Undermines its Authority’, Jerusalem, The Israel Democracy Institute, 15 February 2010: www.idi.org.il/ BreakingNews/Pages/171.aspx. 48 An instructive case in this context is that of Anon, CHR 8823/07 John Doe (Anon) v State of Israel (11 February 2010), Nevo Legal Database (by subscription) (in Hebrew); the Court ruling prohibiting the use of torture in investigations is also worth mentioning – see HCJ 5100/94 Public Committee against Torture in Israel v State of Israel 53(4) PD 817 [1999] (in Hebrew). Another important case in this context is A v Secretary of State for the Home Department [2004] UKHL 56, [2005] 2 AC 68, where the House of Lords held that the indefinite detention, without trial, of foreign prisoners, who cannot be deported, under s 23 of the Antiterrorism, Crime and Security Act 2001 was incompatible with the Convention for the Protection of Human Rights and Fundamental Freedoms (European Convention on Human Rights, as amended) (ECHR). 49 For elaboration on these biases and their influence, see M Kremnitzer, ‘Terrorism and Democracy – An Israeli Perspective’ in T Wurtenberger et al (eds), Innere Sicherheit im Europaischen Vergleich (Berlin, LIT Verlag Dr W Hopf, 2012) 203.
Constitutional Proportionality 237
demand awareness and adjustment are the short-range perspective of the government, its need for ongoing public support and its need to demonstrate strength and confidence, the tendency to downplay the influence of our actions on opposing parties (eg the subjective–motivational effect on the other party, such as ignoring the danger of increasing terror as a result of implementing unjust measures such as demolishing houses, torture, arbitrary detention, etc) and the mistaken approach of aspiring to the impossible threshold of zero risk in the security realm. If the government fails of its own initiative to avoid these biases, the Court will find itself as a last-resort defender of human rights. 3. The central role – though not the sole role – of the Court in the constitutional realm – the protection of the minorities, especially permanent minorities – should deserve utmost attention. Protecting a permanent minority is tremendously important, since the self-restraint that is usually practiced by a majority that expects to eventually become a minority does not operate in the relationship between a majority and a permanent minority. In such a situation, one cannot rely on the majority rule.50 A minority can be politically extremely powerful (the religious sector in Israel) and it may be devoid of real political power. It is the second type of minority that deserves special protection. The Arab minority in Israel is both a permanent minority and a minority with no political power. Therefore it deserves enhanced protection. As a rule, a minority will suffer from feelings of insecurity, inferiority and even fear. Discriminating or hostile legislation against a minority enhance these feelings. When the state acts against a minority, anti-minority sentiments among members of the majority flourish. It is for these reasons that the protection of the minority should be a compass guiding the Court’s path.51 IV. THE RELATIONSHIP BETWEEN THE PROPORTIONALITY TESTS
The most important point in this context is the conception of proportionality as a ‘global’ balancing or assessment rather than as an independent examination of each one of the subtests in a manner that is not contingent upon the results of the other subtests. The stages are not distinct and alien to one another. There are different ways to pass a test, from nearly failing to passing with flying colours. When difficulties, doubts or uncertainties are encountered (including weaknesses in the process or in the evidence forming the basis of the conclusions), they should be carried on to the next stages, otherwise the bias (psychological and analytical) ‘in favour’ of the measure is liable to distort the final result. 50 See, eg JH Ely, Democracy and Distrust: A Theory of Judicial Review (Cambridge, Harvard University Press, 1980) chs 4 and 6. See also Movement for Quality Government v Knesset (n 15) para 15 of Justice Grunis’s opinion: ‘The principal justification for the existence of judicial review of legislation is the need to protect the minority and individuals from the tyranny of the majority. The court is the last barrier capable of preventing the majority harming individuals and minority groups’. 51 A full internalization of this concept can be found, perhaps not surprisingly, in the Constitutional Court of South Africa in the dissenting opinion in Prince v President of the Law Society of the Cape of Good Hope 2002 (2) SA 794 (CC) (S Afr), which adjudicated the constitutionality of applying the prohibition of cannabis to Rastafarians. See para 157 of the judgment: ‘The hydraulic insistence on conformity could have a particularly negative impact on the Rastafari, who are easily identifiable, subject to prejudice and politically powerless . . . because they are politically powerless and unable to secure their position by means of a legislative exemption, the Rastafari are compelled to litigate to invoke their constitutional rights’ (emphasis added).
238 Mordechai Kremnitzer V. CONCLUDING REMARKS
Four tests are addressed by proportionality in the broad sense. One is the appropriate purpose, the second is the rational connection, the third is the least restrictive means, and the fourth is proportionality in the narrow sense. The proposals advanced above seek to add substance to the rational connection test and the least restrictive means test, so as to reduce the growing burden placed on the shoulders of the test of proportionality in the narrow sense. In addition, they seek to structure the proportionality stricto sensu test by presenting normative and practical considerations to be taken into account in that framework. The proposals were written to guide and structure the application of proportionality, not only for the judicial authority, but for all ‘users’ of proportionality – the legislature, which is obligated to choose only goals that are appropriate and means that are proportionate, and the executive, which bears the duty of making proportionate use of the means that it is authorised by law to exercise, while according for human rights. My final statement: the commitment to take rights seriously cannot be accomplished without taking proportionality seriously. In this respect, we have a long way before us.
16 The Deficiencies of Balancing: Restricting Speech due to Offence to Feelings YAACOV BEN-SHEMESH
I. INTRODUCTION
O
VER THE PAST few decades the balancing approach has risen to the status of the leading, if not exclusive, method for resolving cases of conflicting interests and rights in Israeli law.1 In this chapter I will argue that in addition to the wellknown flaws of balancing, in some cases it is simply counterproductive, and the very willingness of the courts to balance leads to undesirable consequences. This is the case, I will argue, with respect to the willingness of the courts to balance freedom of speech with offence to public feelings. I will show that contrary to expectations, this willingness may lead, among other undeniable consequences, to more, rather than less, offence to feelings. I will conclude that in such cases, a categorical approach, that is, refusing to balance, and simply rejecting offence to feelings as a legitimate reason for restricting speech, is preferable. After describing the balancing approach in the remainder of the introduction, I will follow, in part II, its rise in Israeli free speech jurisprudence. In part III I will then examine its outcome in several cases, and offer an alternative approach. The rise of balancing is part of a broader phenomenon in Israeli law, noticeable particularly in recent decades, of the decline of formalism and the rise of the jurisprudence of values in Israeli law.2 According to the balancing approach, in a conflict between rights or interests, the conflicting interests need to be identified, their importance evaluated, the magnitude and probability of their violation estimated, and then a decision made based on a ‘balance’ of all the above. For example, regarding the question of whether to permit a demonstration outside the private house of an important public figure, one must consider, amongst other things, the importance of the right to freedom of speech as opposed to the right to privacy, the degree of violation of the respective rights and its likelihood should the other right be preferred, and all other relevant circumstances (such as whether the private house also serves for the public figure’s public activities; the possibility of harming other tenants, and the hours requested for the demonstration). All of these factors must then be appropriately balanced.3 1 For a general survey, see R Segev (under the guidance of M Kremnitzer), Weighing Values and Balancing Interests (Jerusalem, The Israel Democracy Institute, 2008) ch 4. 2 See M Mautner, Law and the Culture of Israel (Oxford, Oxford Univeristy Press, 2011) ch 4. 3 See HCJ 2481/93 Dayan v Wilk 48(2) PD 456 [1994] (in Hebrew).
240 Yaacov Ben-Shemesh An alternative to balancing is what I will refer to as the categorical approach. According to the latter, there are cases in which no weight at all should be ascribed to certain interests. For example, the position that in clarifying the property rights over a certain asset, no consideration should be given to the relative wealth of the parties, or their relative need for the property in question, is an example of the categorical approach: in determining the question of ownership, certain considerations must be ignored. The balancing approach on the other hand would claim that considerations of justice may be relevant, but would most likely ascribe them relatively minimal weight. The categorical approach is an application of Joseph Raz’s concept of ‘exclusionary reasons’.4 An exclusionary reason negates consideration for particular factors in certain contexts. In our previous example, the right of property is an exclusionary reason with respect to certain considerations of justice or utility. Naturally, there are other contexts in which considerations of justice will be relevant; the categorical approach is invariably a relative one: concerning a certain right, X, certain considerations of category Y must be disregarded. Clearly, choosing to use the categorical approach in certain cases is in itself a result of a value-based choice. Hence, the importance of the right of property, and the need to ensure stability and legal certainty as well as other considerations of social policy are the factors that underlie the categorical position outlined above. In other words, the categorical approach likewise reflects a value decision. But once the decision to use the categorical approach is made in a certain context, the resolution of future cases becomes more technical and much less value-laden. And this indeed is the strongest advantage of the categorical approach. The advantages of the balancing approach are well known, focusing mainly on its tremendous flexibility and its ability to factor in many relevant considerations, deriving a conclusion adjusted to the specific circumstances of a given case. Its shortcomings are also well recognised. Most important is the absence of clear criteria for the balancing ‘itself’, for deciding between the conflicting interests, notwithstanding the seemingly scientific format in which the balancing approach is worded.5 Certain critics have indeed concluded, based on the absence of criteria, that the balancing approach is actually devoid of any substantive content and hence useless.6 Even if we reject the radical claim, it seems undeniable that the balancing approach involves a definite dimension of subjectivity and judicial caprice, insofar as the ‘appropriate balance’ between the conflicting values differs from one judge to the next in accordance with the specific world views and personal values of the judges. The inexorable result is that it also detracts from legal certainty and may thus impair the protection of certain rights. For example, a categorical rule stating that freedom of speech may not be abridged based on claims relating to the truth of the expression, which is the current guiding principle in Israeli law,7 would 4 See J Raz, Practical Reason and Norms (Oxford, Oxford University Press, 1975); and see RH Pildes, ‘Avoiding Balancing: The Role of Exclusionary Reasons in Constitutional Law’ (1994) 45 Hastings Law Journal 711. 5 See M Birnhack, ‘Constitutional Geometry: The Constitutional Methodology of the Supreme Court’ (2003) 19 Bar-Ilan Legal Studies 593 (in Hebrew). 6 For an incisive critique on the absence of any rational criteria in the balancing of values, and the law’s despairing attempts to conceal that fact using a variety of rhetorical tools, see P Schlag, The Enchantment of Reason (Durham, Duke University Press, 1998). 7 ‘[R]egarding the freedom of expression, we do not concern ourselves with the truth of the expression’, HCJ 806/88 Universal City Studios Inc v Film and Play Review Board 43(2) PD 22, 33 [1989] (in Hebrew).
The Deficiences of Balancing 241
provide a better protection of freedom of speech than a rule stating that freedom of speech should be preferred unless the expression under discussion is blatantly and damagingly false. The reason is that the term ‘blatantly and damagingly’ invites a resort to balancing, which may result in the depletion of the protection conferred on freedom of speech. Despite its striking defects, the balancing approach has one decisive advantage, namely the absence of any reasonable alternative in a wide variety of cases.8 As Cheshin J noted, ‘the concept of balancing rules us and all that surrounds us’.9 Nonetheless, this does not mean that there is absolutely no alternative to the balancing approach and that it must be used in every case. In this chapter I explore certain contexts in which the harms of the balancing method outweigh its advantages, and therefore the categorical approach should be preferred. In a sense, the current chapter is part of a larger trend that can be identified in Israeli law, according to which the value-based balancing approach has gone too far, and a partial restoration of the formalistic dimensions of the law are needed.10 My test case is the conflict between the right to freedom of speech and the offence to the feelings of the public. In the following pages I will show how Israeli law initially refused to recognise the notion of a violation of public sensitivities as legitimate grounds for restricting freedom of speech. Using current terms, Israeli law used to invoke a categorical approach – considerations of public feeling are irrelevant when deciding issues of free speech. The rise of the balancing discourse changed this. Public feelings merited gradually growing importance as an interest deserving of protection, and in respect of which the balancing approach should be applied. This was the case regarding freedom of speech and demonstration, the laws of defamation, and the legitimate limits of elections broadcasts. Simultaneously, accepting public feelings as a legitimate consideration to be factored in triggered the ascent of the ‘feelings discourse’ in the judgments, concerned with the feelings that certain forms of speech evoked, according to the judges. The feelings of the judges started occupying an increasingly large place in the judgments. Nowadays it is common to see judges intentionally projecting us into the world of their inner feelings, pouring out their hearts, and applying emotional and even sentimental rhetoric to describe the offensiveness of certain expressions. My claim is that this phenomenon in both of its aspects is undesirable, portending an unwarranted risk to freedom of speech. In the context of the offence to feelings, I argue that the categorical approach – opposing any limitation on free speech based on claims of offence to public feelings – should be preferred and in fact restored. I will argue that using the balancing approach not only unjustifiably depletes free speech, given the problematic nature of the offence to public feelings claim, but also, in certain cases, actually magnifies the offence to the very feelings it purports to protect. I will show that the Israeli Supreme Court’s readiness to restrict freedom of speech on the grounds of offence to feelings, as expressed in its application of the balancing approach, will in certain cases 8 For a discussion of the various attempts to offer alternatives to the balancing approach, which have been at the most only partially successful, see Segev (n 1) 89–91 (mostly discussing the attempt to distinguish firstorder reasons from second-order reasons, and restricting balancing only for reasons of the same order). 9 CrimA 537/95 Ganimat v State of Israel 49(3) PD 365, 397 [1995] (in Hebrew). 10 This trend has been gaining increasing currency in other fields of Israeli law. See, for an example from contract law, the discourse in the wake of the Apropim case (CA 4628/93 State of Israel v Apropim Housing and Promotions 49(2) PD 265 [1995] (in Hebrew)). See, eg G Shalev, ‘Who Governs the Contract? Another Look at the Aproprim Case’ in A Barak, Y Zamir and Y Merzel (eds), Mishael Cheshin Book (Tel-Aviv, Israel Bar Publishing, 2009) 645 (in Hebrew).
242 Yaacov Ben-Shemesh render freedom of speech the prisoner not of a hostile audience but rather of a sensitive audience. Now, more than ever before, the question of whether a certain expression will be prohibited is dependent on the magnitude of the anticipated harm and the degree of sensitivity of the public that responds to the expression. If the offence to feelings is sufficiently powerful, it may justify the imposition of limitations on the freedom of speech. The problem is that this approach supplies an incentive for those interested in the stifling of the expression to attempt to ‘manufacture’ a stormy environment of inflamed passions and feelings and to intentionally aggravate the offence to feelings in the hope of persuading the court to restrict the speech. The paradoxical result is that the court’s willingness to restrict the freedom of speech in order to protect against the offence to feelings actually magnifies its scope. Based on these and other considerations I argue that the protection of public feelings should be categorically rejected as an illegitimate consideration for restricting freedom of speech. The rise of the ‘feelings rhetoric’ deserves special attention and research which I will not attempt here. I will mention, however, that there are reasons to believe this development impairs the quality of the judgments. First, it increases the subjective dimension of the ruling and detracts from legal certainty. What one judge finds offensive may not necessarily appear that way to another judge. The ascent of the feelings discourse therefore makes the legal result dependent not only upon the sensitivity of the public but also upon that of the judges. Furthermore, as we shall see, when a judge writes from the depth of her heart her vision narrows; the breadth of her perspective is impaired, and she becomes less receptive to the possibility of understanding the expression in other, less offensive, manners. These are undesirable consequence on their own, but they also add to the diminishment of the protection of free speech. In part I, I describe the rise of the balancing approach and the rhetoric of feelings in the Supreme Court case law. In part II, I demonstrate the undesirable dynamic created by the balancing approach in churning up public feelings with reference to two events: the dispute concerning the Pride parade in Jerusalem, and the dispute over the publication of caricatures of Mohammed in Europe. Finally, I will clarify why the use of the balancing approach should be limited, as should be the attendant discourse of feelings in case law concerning freedom of speech. II. THE RISE OF THE OFFENCE TO FEELINGS ARGUMENT
During the first three decades of the State’s existence, the Supreme Court did not recognise a general claim of ‘offence to public feelings’ as a reason for restricting freedom of speech.11 A restriction of freedom of speech was legitimate when the expression had the 11 In this chapter I have focused on the offence to feelings in the context of freedom of speech, although it bears mention that in other contexts too there is a growing tendency to ascribe importance to offence to feelings. Indeed, during the first few decades, in matters unrelated to freedom of speech too, the Supreme Court was generally unwilling to ascribe significant weight to the claim of offence to feelings. See, eg HCJ 357/61 Tatlees v Mayor of Herzliya 16(2) PD 902 [1961] (in Hebrew) (rejecting the argument that a licence should not be granted because it gave offence to feelings); HCJ 124/70 Cochavei Shemesh v Registrar of Companies 25(1) PD 505 [1971] (in Hebrew) (acceptance of a petition against the Companies Registrar, who refused to register a film titled Sirtei Kotel Ltd, on grounds of its giving offence to feelings; HCJ 230/73 STM Ltd v Mayor of Jerusalem 28(2) PD 113 [1974] (in Hebrew) (annulment of Mayor’s decision not to grant a business licence to a shop for sexual accessories because of its offence to feelings).
The Deficiences of Balancing 243
potential to cause real physical damage to public order, or peace, or to the life, person or property of another person.12 In addition the Court was prepared to consider an offence to feelings only in two specific contexts: an affront to religious feelings and a violation of public morality attendant to pornographic publications.13 Even this, however, was not the product of a general willingness to balance freedom of speech and offence to feelings, but rather was anchored in the provisions of the criminal law, which prohibited the publication of obscene publications14 and matter that may have outraged religious feelings.15 For example, in the Keinan16 case the Court confirmed the decision of the Films and Plays Censorship Board (hereinafter the Board) not to permit the presentation of the play Friends of Jesus written by Amos Keinan, on the grounds that it was offensive to the religious feelings of Christians. The Court explicitly related to the criminal prohibition, determining that the Board was entitled to prohibit the showing of a play that involved a violation of the criminal law. All the same, and notably, already in that judgment the Court questioned whether an offence to religious feelings in the generic sense, that is where not involving a breach of criminal prohibitions, provided sufficient grounds for the curbing of free speech. In its decision the Board also referred to additional offences to public feelings, apart from an offence to religious feelings, but in respect of the additional offences, Justice Landau wrote the following: There can be no doubt that offending the feelings of mourning of the individual as well as of the public attests to unheard of callousness and vulgarity. Even so, despite our revulsion, we would have hesitated in confirming the prohibition of the presentation were this the sole grounds, given our concern for a violation of freedom of expression. We can however leave this matter pending because for purposes of our decision the grounds of offense to religious feelings sufficed to justify the Board’s negative decision.17
In a gradual process however, the concept of an offence to feelings was severed from its criminal origins and religious, moral roots. Gradually, it became recognised as an independent claim concerning a general offence to public feelings. The first key judgment in which offence to public feelings merited independent status was the decision of Justice Barak in the Laor case.18 In that case the Board prohibited the staging of the play Ephraim Returns to the Army, arguing that comparing IDF (Israel Defence Forces) soldiers and the military administration in the Territories, to the Nazi occupation, was grossly offensive to the feelings of the entire Jewish public. The Court accepted the claim in principle that an offence to feelings may justify the restriction of freedom of speech, and made it clear that the Board’s authority to prohibit a play was not limited to plays See HCJ 73/53 Kol Ha’am v Minister of Interior 7 PD 871, 876, 879 [1953] (in Hebrew). ibid 879. 14 See Penal Law, 5737-1977, s 214(a) (original version in Criminal Code Ordinance, 5696-1936, s 179). 15 Penal Law (n 14) s 173 (which replaced s 149 of the Ordinance). The question of whether offending of religious feelings should be criminalized, and whether religious feelings merit protection as opposed to other kinds of feelings, has been extensively discussed. For a survey and argument in favour of abolishing these prohibitions, see M Kremnitzer, S Goldman and E Tamir, Religious Feelings, Freedom of Speech, and the Criminal Law: Proposal to Abolish the Criminal Prohibition of Expressions that Offend Religious Feelings (Jerusalem, The Israel Democracy Institute, 2003). On the question of whether religious feelings deserve special protection, see D Statman, ‘Hurting Religious Feelings’ in M Mautner, A Sagi and R Shamir (eds), Multiculturalism in a Democratic and Jewish State (Tel Aviv, Ramot, 1998) 133–88. 16 HCJ 351/72 Keinan v Film and Play Review Board 26(2) PD 811 [1972] (in Hebrew). 17 ibid 816. 18 HCJ 14/86 Laor v Film and Play Review Board 41(1) PD 421 [1987] (in Hebrew). 12 13
244 Yaacov Ben-Shemesh which if presented would constitute a criminal offence.19 On the other hand, along with this principled recognition, the Supreme Court expressed its clear reservations regarding the curbing of freedom of speech exclusively by reason of an offence to public feelings. The judgment ascribes minimal weight to the claim of offence to public feelings and in contrast elaborates on the claim that the proper response to a publication that offends public feelings is by articulating opposition to its message: ‘In a democratic regime, the confrontation with this kind of work should not be by way of governmental power but rather by way of education and persuasion; it must fail because of its exposure and not its suppression’.20 Specifically addressing the comparison between the IDF soldiers and the Nazi soldiers, Justice Barak ruled that the offence was not sufficiently severe to justify the restriction of expression: I myself was a child during the Holocaust, and I crossed fences and borders guarded by the German Army smuggling objects on my body. The parallel between the German soldier arresting a child and the Israeli soldier arresting an Arab youngster breaks my heart. Nonetheless, we live in a democratic state, in which this heartbreak is the very heart of democracy. Its strength lies not in its recognition of my right to hear pleasant matters that are music to my ears. Its strength lies in its recognition of a person’s right to express matters that are grating to my ears and which break my heart.21
The Laor judgment thus includes both the principled recognition of an offence to the feelings of the general public as a claim that may justify a restriction of freedom of speech, and reluctance to invoke those grounds on a practical level. A few years later, in the Universal case,22 the metamorphosis of the offence to feelings to the status of an independent claim moved an important step forward. The context was the Board’s decision to prohibit the screening of The Last Temptation of Christ on the grounds that it would gravely offend the religious feelings of Christians in Israel. The issue concerned religious feelings, thereby relieving the Court of the need to rule on the status of the more general offence to feelings claim; indeed, for the length of his judgment Justice Shamgar was careful to refer exclusively to an offence to religious feelings as grounds for placing restrictions on expression.23 Justice Barak on the other hand, viewed offence to religious feelings as a specific instance of offence to public feelings, which he regards as a central reason that may justify a limitation on freedom of expression.24 In his judgment Justice Barak related for the first time to the difficulty in principle of recognizing an affront to feelings as a ground for curbing expression, insofar as it gives rise to the concern that ‘every expression might harm some sentiment and recognition of harm to sentiments as a basis for restricting freedom of expression – without drawing any distinction between religious sentiments and other sentiments – could devoid the freedom of speech of any content’.25 Justice Barak’s solution to this problem consisted on the one hand of reaffirming the holding that in principle the consideration is a legitimate one, but at the same time he nearly empties it of practical significance by
ibid para 12. ibid para 14. 21 ibid para 24. 22 Universal (n 7). 23 See, eg ibid paras 7 and 10 of Justice Shamgar’s opinion. 24 ibid para 10. 25 ibid para 9. 19 20
The Deficiences of Balancing 245
requiring that the offence be ‘harsh, serious, and severe . . . that exceeds the “tolerance threshold” of a democratic society’.26 Justice Barak’s position was subsequently adopted as the settled case law of the Supreme Court, in a number of judgments handed down from the middle of the 1990s, and until about six years ago.27 The position can be summarised as follows: accepting offence to feelings as a legitimate, independent reason for curbing free speech, even if no physical harm or danger is expected, combined with the practical negation of the possibility of actually restricting speech for this reason by establishing a particularly demanding balancing formula requiring that the injury be harsh, serious, and severe to the extent of shaking the foundations of democracy and mutual tolerance. The Supreme Court’s willingness to grant legitimacy in principle to the claim of offence to feelings precipitated a wave of similarly motivated petitions, beginning from the early 1990s and until today. Indeed, during the first decade of this period (1993– 2004), in a relatively long series of judgments the Court consistently refused to give practical expression to its principled recognition, ruling time and again that the offence to feelings had not reached the high level of gravity required to justify a restriction of freedom of speech.28 However, the stream of petitions did not stop until the dam burst, and having burst, it remains open. The first time the Court imposed restrictions on free speech based on an offence to feelings was in the Shinui case, delivered in 2006.29 It concerned a petition of the Shinui party against a decision of the Chairman of the Central Elections Committee to disqualify a campaign clip that Shinui asked to broadcast during the elections to the Seventeenth Knesset (Israeli Parliament). The Court rejected the petition, leaving the disqualification intact and ruling that the broadcast was vilely reminiscent of anti-semitic propaganda, in which the ultra-Orthodox Jew becomes a faceless person, a ‘non-human’ who crawls on the floor and clings to the secular person as if he was a leech . . . this depiction is degrading and humiliating, and violates human dignity in general and specifically the dignity of the ultraOrthodox Jew, stripping him of his image and dignity as a human being.30
The Court’s conclusion was that the offence to feelings exceeded the level of tolerance required in a democratic society, and that the Chairman of the Central Elections Committee was therefore entitled to prohibit its broadcast.31 ibid para 11. See, eg HCJ 6126/94 Senesh v Broadcasting Authority 53(3) PD 817 [1999] (in Hebrew); HCJ 316/03 Bakri v Israel Film Council 58(1) PD 249 [2003] (in Hebrew). 28 HCJ 606/93 Advancement Promotions and Publishing (1981) Ltd v Broadcasting Authority 48(2) PD 1 [1994] (in Hebrew) (confirmation of the broadcast of a radio advertisement containing the expression ‘leh titztayen’ (literally – go and excel, but written in Hebrew the expression resembles a four letter expletive)); Senesh (n 27); (see English translation in: [1998–89] Isr LR 339) (confirming the television broadcast of a programme offending the character of Hana Szenes); HCJ 2888/97 Novick v Second Television and Radio Authority 51(5) PD 193 [1997] (in Hebrew) (see English translation in: [1998–89] Isr LR 362) (confirming broadcast of television programme concerning the Rabin assassination, linking the incitement that preceded the assassination to the assassination itself); HCJ 1514/01 Gur Aryeh v Second Television and Radio Authority 55(4) PD 267 [2001] (in Hebrew) (confirming broadcast of television programme on Shabbat, featuring the petitioner, who are religious people); Bakri (n 27) (see English translation in: [2002–03] Isr LR 487 (confirming the screening of the film Jenin Jenin); HCJ 5432/03 SIN v Council for Cable and Satellite Broadcasting 58(3) PD 65 [2004] (in Hebrew) (confirming the broadcast of Playboy and other pornographic channels in the framework of the cables and satellites broadcasts). 29 HCJ 2194/06 Shinui Party v Chairman of the Central Elections Committee (21 June 2006), Nevo Legal Database (by subscription) (in Hebrew). 30 ibid para 14 of Court President Barak’s opinion. 31 ibid. 26 27
246 Yaacov Ben-Shemesh It is doubtful whether the broadcast of the clip would really have shaken the foundations of mutual tolerance. It is also doubtful whether Justice Barak’s description of the clip is the sole way of viewing and understanding it. A similar tone is evidenced in Justice Rivlin’s opinion on the matter, who concurred, albeit not without reservation, with Justice Barak’s view. He added that the decision was ‘borderline’ and that the clip could have been left untouched, but that the decision to disqualify it could not be considered unreasonable to a degree that warranted the Court intervention.32 It is important to add however, that the restriction on freedom of speech was nonetheless subjected to an additional condition that must obtain in order to restrict speech based on the claim of offence to feelings: the requirement that the expression be one that humiliates, degrades and offends the dignity of a group of people. In the words of Court President Barak: We are persuaded that the case before us falls into the category of exceptional cases in which the damage to emotions amounts to humiliation and degradation and an infringement on the right to human dignity so severe that it should not be protected.33
Justice Rivlin too addressed this aspect: In this case there is not only an offense to the public. The advertisement conveys a message that offends the dignity of a specific group, on the basis of belonging to that group. The petitioner is attempting to pave its way to the Knesset by way of its degrading presentation of the ultraOrthodox public. The question of the manner in which one should treat an expression deemed as gravely offending a minority group, or any other group of diminished power, has and continues to occupy numerous systems of law.34
However, only a short time passed before the Court waived this requirement. This development was reflected in a series of judgments given between 2006–08 on petitions filed against the decisions of the police to permit the LGBT Pride parade in Jerusalem.35 As outlined below, over the years the intention to conduct the Pride parade has encountered fierce opposition on the part of religious groups in the city. Despite the opposition, each of the annual parades received police approval, subject to restrictions on the route and proceedings. In each of the years a petition was filed against the police, claiming that the licence should be cancelled because the parade would endanger the public peace and offend the public’s feelings. In each of the petitions the claim of endangering public peace was rejected both because the Court saw no grounds for interfering with the police decision on the matter, and because of its principled reservations regarding claims of this nature when raised by the opponents of the parade. Here, the Court trod the path of established case law, which ruled that when the danger to public safety originated not from the demonstrators, but rather from a ‘hostile audience’ the police are required to adopt all possible measures to ensure the safety of the demonstrators, and not to surrender to threats and prevent the demonstration.36 ibid para 8 of Justice Rivlin’s opinion. ibid para 14 of Court President Barak’s opinion. 34 ibid para 5 of Justice Rivlin’s opinion. 35 HCJ 8988/06 Meshi Zahav v Police Commissioner for the Jerusalem District (27 December 2006), Nevo Legal Database (by subscription) (in Hebrew); HCJ 5277/07 Marzel v Police Commissioner for the Jerusalem District (20 June 2007), Nevo Legal Database (by subscription) (in Hebrew); HCJ 5317/08 Marzel v Police Commissioner for the Jerusalem District (21 July 2008), Nevo Legal Database (by subscription) (in Hebrew). 36 HCJ 153/83 Levi v Police Commissioner for the Southern District 38(2) PD 393 [1984] para 9 of Justice Barak’s opinion (in Hebrew) (see English translation in: Isr SC 7, 38): ‘A person’s freedom is not to be infringed merely because of violent objection to its exercise . . . It is the task of the police, in this situation, to keep the 32 33
The Deficiences of Balancing 247
In contrast, the claim regarding offence to feelings merited a far more attentive ear. In all of the petitions the Court rejected the demand to cancel the parade because of the offence to public feelings, but its reasoning was that the restrictions imposed on the route and conduct of the parade created an appropriate balance between the right to freedom of speech and the need to prevent offence to public feelings. In 2006 and 2007 this position was only alluded to, but in the decision on the 2008 petition the claim of offence to feelings received significant weight as a factor that justified the imposition of significant limitations on the parade. In the words of Justice Procaccia: On the one hand, the Pride Parade should not be prevented, but on the other hand, consideration should be had for public feelings and all possible measures taken to mitigate the offense in various relevant aspects, including in terms of timing, location, the length of the route, the duration of the event, the substantive format of the event itself, and the anticipated general conduct and behavior of those participating in it.37
This completed the process whereby the Supreme Court accepted the claim that the offence to feelings should be ascribed significant weight in determining the limits of the freedom of speech, even where it concerned a patently political expression of an oppressed minority group, and even where it is clear that the expression is not targeting, humiliating or degrading any groups or individuals, but rather it aims to promote the rights of the LGBT community and its struggle for equality. In the words of Justice Procaccia: This gathering of the gay and lesbian community and their participation in an event in a central part of the city is intended to impart the community’s message to the public at large, and to bring itself closer to them. They are intended to further the general public’s familiarity with the community, and its own familiarity with the public, and to establish the community’s right to blend into the overall social fabric of the city and its populace. The gathering of all the members of the community for purposes of the event is instrumental in strengthening the interconnectedness of its members and crystallizing the social robustness of the group as a whole.38
In fact, in all of these cases it was the petitioners themselves whose speech and conduct were intended to humiliate and degrade.39 The necessary conclusion is that the law today permits the imposition of significant restrictions on the freedom of speech if it prevents a serious offence to public feelings, irrespective of the nature and motivations of the speech and the motivations of those opposing it.
crowd from disturbing demonstrators, and not to prevent the demonstrators from exercising their right . . . Showing deference to crowd hostility is like entrusting the key to exercise of the right of demonstration and procession to those who oppose it. This has to be avoided and the mob is not to be given a power of veto, nor violence a reward’. 37 HCJ 5317/08 Marzel (n 35) para 7 of Justice Procaccia’s opinion. 38 ibid para 5 of Justice Procaccia’s opinion. 39 I will not repeat the numerous homophobic and offensive expressions of the petitioners against the LGBT community, and will suffice with saying that following the dismissal of one of the petitions, some of the petitioners (Baruch Marzel and Itamar Ben-Gvir) organized a ‘beast parade’ that was conducted in parallel to the Pride parade (see Ari Galhar and Ronen Medzini, ‘Donkeys at the Parade in Jerusalem; 3 Teenage Girls With Eggs Arrested’, Ynet, 29 July 2010: www.ynet.co.il/articles/0,7340,L-3926962,00.html (in Hebrew)).
248 Yaacov Ben-Shemesh III. A CRITICAL EVALUATION: THE REASONS FOR NOT RECOGNIZING AN OFFENCE TO FEELINGS AND WHY THE ATTEMPT TO PROTECT FEELINGS ACTUALLY AGGRAVATES THE OFFENCE
My view, shared by many, is that freedom of speech should not be curbed on the grounds of its being hurtful to feelings, religious or otherwise.40 An offence to feelings and values is part of the reality of life in a pluralistic and multicultural society. In that reality, a person must anticipate and accept the possibility of his or her feelings hurt by conduct, values or beliefs of others who differ from him or her. The hurting of feelings is an inseparable part of life in a pluralistic society, and an inevitable and often desirable result of the exposure to a world of diverse values and customs. Furthermore, since offence to values and feelings is both inevitable and ubiquitous, the willingness to recognise it as legitimate grounds for restricting speech leads ineluctably to limiting the speech of those who are weak and lacking in political power. Limiting all offensive speech amounts to silencing almost anything controversial, and is therefore inconceivable. The result is that the offence argument is most likely to be used by those who have political power, and against those who lack it. It is no accident, for example, that in a city where there is no shortage of events that can be described as offensive to religious feelings, the Pride parade, of all things, was chosen as a target, and not, say, the mosques on Temple Mount, or the growing number of restaurants open on the Sabbath. Battles over those issues pose an overly formidable challenge for the ultra-Orthodox community. The LGBT community of Jerusalem, on the other hand, seemed a relatively convenient victim, against which part of the secular community may similarly be mobilised, and against which an inter-religious coalition can be formed.41 The danger of the selective use is then another reason for the categorical rejection of offence to feelings as grounds for limitation of free speech. Apart from all these general arguments, I believe that close examination of the offence to feeling claim reveals that often it is nothing more than an attempt to demonstrate political power and influence, and where this is the case, the hurting of feelings claim is just one more manipulative tool wielded in a power struggle. This becomes evident if we bear in mind that often the best way of diminishing the offence to feelings occasioned by certain expressions is to simply ignore them. The attempt to stifle them may spark off a societal struggle that leads to a far greater dissemination of the offending expression, and hence a graver, more widespread outrage of public sentiments. Despite the simple, practical logic of this claim, there are numerous cases in which an attempt is made to curb particular expressions based on their offensive nature, even though the parties 40 For an excellent review and summary, see D Statman and G Sapir, ‘Freedom of Religion, Freedom from Religion, and Protecting Religious Feelings’ (2004) 21 Bar-Ilan Legal Studies 5 (in Hebrew). 41 Eg in the context of the caricatures, the human rights activist Maryam Namazie wondered why no apology was demanded from the leaders of the Muslim religion: ‘I’d like the offended Islamists – from the Islamic Republic of Iran to Islamic Jihad to the Saudi government . . . – to apologise; not for their backward and medieval superstitions and religious mumbo jumbo but for their imposition of these beliefs in the form of states, Islamic laws and the political Islamic movement. If any of them want to apologise for the mass murder of countless human beings in Iran and the Middle East, and more recently in Europe, for veiling and sexual apartheid, for stoning, amputations, decapitations, Islamic terrorism and for the recent brutal attack on Tehran bus workers and so on and so forth, just email me direct’ (Maryam Namazie, ‘Apologise for what? On Caricatures of Mohammad’, Namazie’s blog, 1 February 2006: maryamnamazie.blogspot.com/2006/02/ apologise-for-what-on-caricatures-of.html).
The Deficiences of Balancing 249
attempting to stifle the expression are well aware of the inevitable consequences of their battle. I will demonstrate this argument having resort to two events from recent years: the dispute over the publications of caricatures of Mohammed in Europe in 2006 and the dispute over the Pride parade in Jerusalem in the years 2005–07. I will begin with a brief description of the unfolding of the events in each of the cases. A. The Caricatures Case In September 2005 the Danish newspaper Jyllands-Posten featured caricatures mocking Mohammed and linking Islam to terror and violence. For example, one of the caricatures displayed Mohammed with a bomb concealed under his turban. The publications triggered strident protests on the part of the Muslim factions in Denmark and demands for an official apology and the adoption of measures against the publishers; the claim being that the publication severely offended the religious feelings of Muslims. A number of weeks passed, and after the caricatures were disseminated by religious Muslim figures in other Muslim states, the protest began to make increasingly large waves. On the other hand, newspapers from all over Europe responded to the calls for censorship by publishing the caricatures, in support of freedom of speech and in a show of solidarity with the Danish newspaper, and as part of their coverage of an event that had cascaded into an event of international proportions. In the months of February and March 2006 mass demonstrations were held in many Muslim states, some of them violent, against European targets. The entire event received extensive coverage in all of the media for a protracted period, naturally leading to unprecedented dissemination of the offending caricatures. Caricatures initially viewed by a few hundred thousand readers of a Danish paper were ultimately viewed by hundreds of millions of people all over the world. The storm gradually abated, inter alia, because of a partial apology published by the paper. B. The Jerusalem Pride Parade Case The first LGBT Pride parade in Jerusalem was held in 2002, and subsequently in 2003 and 2004. The first parades were relatively quiet affairs, with a modest turnout at most (from a few hundred up to about three thousand), and passed without incident. This was mainly due to the decision of the Jerusalem ultra-Orthodox leadership to ignore the parade, refusing to confer it the legitimacy and recognition of existence implied by actively opposing it. It was also due to a silent agreement between certain rabbis of the ultra-Orthodox community and the heads of the Open House, the organization that initiated and organised the parades.42 During these years no legal issues arose that necessitated judicial intervention. This situation changed dramatically in 2005. Due to the confluence of a number of events, some of which I discuss below, and the decision of the Jerusalem Open House to conduct the International Pride Parade in Jerusalem (an intention which did not ultimately materialise due to the political circumstances in Israel that summer, concerning 42 During the years 2004–06 the author of this chapter was a member of the Board of the Jerusalem Open House, and many of the matters described here are based on first-hand experiences.
250 Yaacov Ben-Shemesh the ‘Gaza disengagement plan’ which was carried out at that time) the 2005 parade became the focus of a fierce, acerbic intercommunal battle. In a rare show of unity, various religioius leaders, headed by the ultra-Orthodox rabbis, initiated a broadly based campaign aimed at exerting intense pressure upon the police and the Court which would culminate in the cancellation of the parade in Jerusalem. Leaders of other religious communities in Israel, such as the Latin Patriarch, the Vatican’s Israeli ambassador, and the heads of the Sharia courts, joined in. This surprising show of interfaith solidarity precipitated extensive reporting of the event both in Israel and all over the world.43 The Mayor, Knesset Members, and other public leaders also added their voices in calling for the cancellation of the parade, and the whole affair dominated media reports for a number of weeks thereafter. However, stiff opposition notwithstanding, the police granted the permit for the parade. The parade did not end peacefully. The marchers encountered verbal and physical violence, climaxing in an assault by a young ultra-Orthodox man who burst into the ranks of the marchers, stabbed three of the participants, causing them serious injuries. He was arrested, charged, and sentenced to 10 years in prison. These grave events substantively altered the attitude of the Israeli police. In 2006 there was a repetition of the opposition to the event on the part of religious and ultraOrthodox sectors, and as a result, the Israeli police refused to issue the permit, fearing for the safety of the marchers and public order. Following a petition to the Supreme Court filed by the Open House, the permit was granted, but subject to numerous restrictions with respect to the route of the parade, which was significantly changed and abridged.44 Petitions filed against granting the permit were dismissed,45 but ultimately the parade was cancelled due to other security related reasons. Instead there was an assembly in the Givat Ram stadium. This pattern repeated itself in 2007. After much hesitation and delay, the Jerusalem police gave a permit for the parade, once again subject to significant restrictions regarding the route and its conduct (by this stage, the total length of the parade had been whittled down to just a few hundred meters). Opponents’ petitions calling for the cancellation of the parade were dismissed with the Court once again establishing severe limitations intended to strike ‘an appropriate balance’ between freedom of speech and the protection of public feelings.46 In 2008 matters returned to what they had been between the years 2002–04, and municipal calm was preserved, mostly due to the conclusion reached by the leaders of the religious communities that the damage of the opposition exceeded its benefits. Focusing now on the Supreme Court proceedings and decisions during these years, we can see the rise of the offence to feelings claim in action. Interestingly, when the worry of the personal safety of the public was raised, the Court was quick and decisive in rejecting it. The Court consistently reiterated the entrenched principles of Israeli constitutional law, according to which it is unacceptable to limit speech due to threats to safety caused by a ‘hostile crowd’. This, the Court repeatedly and rightly stated, 43 See, eg Meirav Levi, ‘Leaders of the Three Religions Protest the International Gay Pride Parade in Jerusalem’, News 1, 30 May 2009 available at: www.news1.co.il/Archive/001-D-67546-00.html (in Hebrew); see also the New York Times report of the same event: Laurie Goodstein, ‘Clerics Fighting a Gay Festival for Jerusalem’, New York Times, 31 March 2005: www.nytimes.com/2005/03/31/international/worldspecial/31gay. html. 44 HCJ 7348/06 Jerusalem Open House for Pride and Tolerance v Police Commissioner for the Jerusalem District (19 June 2006), Nevo Legal Database (by subscription) (in Hebrew). 45 Meshi Zahav (n 35). 46 HCJ 5317/08 Marzel (n 35).
The Deficiences of Balancing 251
would be tantamount to encouraging and even rewarding violence. Regarding the offence to feelings argument, however, the Court adopted a different approach. Whereas the restriction on the freedom of speech by reason of concern for a disruption of public order was regarded as ‘a prize for violence’ and an incentive to violent conduct, the restriction of freedom of speech by reason of offence to feelings was viewed as an expression of a reasonable balance between two competing social considerations. Closer examination on the other hand reveals that both cases are substantively similar. The Court’s willingness to restrict offensive speech provided an incentive for a hostile crowd to ‘drum up’ a conflict, to inflame emotions, and deliberately intensify the offence to feelings so as to subsequently demand the restriction of the speech. In other words, the willingness of the Court to consider the offence argument, and to balance it against the right to free speech, led to two undesirable outcomes: first, it contributed to the creation of more hurt feelings, as I explain below. Second, freedom of speech is now hostage not of a violent crowd, but of a ‘sensitive’ crowd, depositing the keys for the realization of freedom of speech in its hands. Should it so deign, it will keep the peace and enable the speech. Alternatively it may choose to precipitate a public storm, galvanizing opposition and stirring up emotions, and then, when passions are sufficiently inflamed, it will demand the restriction of free speech ostensibly to protect against giving offence to feelings. This is precisely the conclusion that must be drawn from the unfolding of events in the cases of caricatures, and the Pride parade. An examination of these cases forces the conclusion that those demanding to prohibit publication of the caricatures or to prevent the Pride parade were not really seeking to protect the feelings of religious believers. In both cases it was well known and anticipated that the attempt to suppress the speech would fuel the unprecedented dissemination of the offending speech, in dimensions that would have been unimaginable if not for the attempt to suppress it. For example, the immediate consequence of the campaign to censor the caricatures was the publication of the offensive caricatures all over Europe, and all over the world. The hostile response to the publication of the caricatures and the demand for their censorship prompted dozens of newspapers all over Europe to publish the offending caricatures in an expression of their solidarity with the Danish paper and as a protest against the attempts to silence it. Hundreds of other media outlets all over the world reported the ensuing riots, and brought the caricatures to the knowledge of millions of others, both Muslim and others, thereby compounding and intensifying both the offence to Islam and the offence to the feelings of Muslim believers all over the world. The same is true of the Pride parade. The ultra-Orthodox opposition lead to unprecedented enlistment both on the part of the parade’s supporters and on the part of those who would not, initially, have supported it, but who felt obliged to support it as an expression of their commitment to the freedom of speech, once the ultra-Orthodox had demanded to prevent it. Here too the burgeoning conflict received unprecedented press coverage so that at the end of the day there was not a single ultra-Orthodox in Jerusalem who had not heard about the Pride parade. Ultimately, the event was attended by 10,000 people, the highest turnout in the parade’s history. We therefore see that the calls to protect feelings are actually accompanied by the opposite result – dissemination of the offending speech in unprecedented dimensions, whipping up feelings, intensifying the conflict with a clear and inevitable result – an offence to feelings of incomparably larger dimensions than would have been if not for
252 Yaacov Ben-Shemesh the attempt to stifle the expression based on the claim of offence to feelings. This ironic and paradoxical result is neither incidental nor surprising, it is in fact inevitable if the balancing approach is adopted, for the two reasons already mentioned. First, given the balancing formula, requiring a severe, profound and broad offence in order to justify limitation of speech, the need therefore arises to create and intensify that kind of offence, even if it did not exist initially, by disseminating the offending speech in unprecedented dimensions, by directing the attention of the believers to it, by churning up public furore and by creating a conflict. Only after the atmosphere has been sufficiently inflamed and the emotions sufficiently whipped up, can one come and demand a limitation of the offending speech. Second, the attempt to silence speech, which as mentioned also includes the staging of inflamed emotions, naturally prompts a response from the supporters of freedom of speech, including those who initially felt that the offending speech was superfluous, in bad taste, or insensitive. Many of these who would not initially have seen the need to support such speech will nonetheless support it when its suppression is perceived as an offence to human rights and democracy. The offending speech attains formidable dimensions, far in excess of its original import, becoming a symbol and component in a culture war. This happened in Europe, when the issue of the caricatures was enlisted into a culture war that focused on the increased Muslim immigration to the European states; and it happened in Jerusalem too, when the issue of the Pride parade was enlisted into the broader culture war between the religious and the secular. Like any war, a culture war is an attractive media item, and the conflict merits unprecedented media coverage, with news items, photos, interviews, op-eds both for and against. Once again the result is the unparalleled amplification of the offence to feelings which, prima facie, we initially aimed to protect. We can now see that in many cases, those who claim that an offence to feeling must be prevented, are in fact motivated by totally different reasons. If protecting emotions was truly their motive, the best strategy would be to simply ignore the offensive speech, thus minimizing its harmful effects. Since they do not do that, we must conclude that they are in fact trying to promote other societal and political interests, that have very little to do with defending emotions. I will enumerate some of these interests without presuming to exhaust the many and varied possibilities. The conflict as a unifying event. Often a conflict becomes a focus around which the community can be rallied and unified to fortify its self-definition and identity, by contrasting it with the values and culture of ‘the other’. It becomes one of those unifying events that are so important for the internal cohesion of every community. Indeed, as exemplified by the cases of the caricatures and the Pride parade, events of this kind are particularly important to traditional minority groups battling to preserve their identity and uniqueness in the framework of the secular and liberal state. Both cases provide the opportunity for the minority communities to once again trumpet the shortcomings of the liberal, secular, degenerate and godless West that sanctifies human rights and individualism in contrast with their own traditional and religious values. The conflict as a show of strength. The conflict may serve to show both the internal and external strength of the group. Thus the Pride parade conflict in Jerusalem was intended, inter alia, to give expression to the power of the ultra-Orthodox community, enabling them to show ‘who is in charge’ in Jerusalem. And as I explained above, the target – the LGBT community – was carefully selected in accordance with that goal.
The Deficiences of Balancing 253
The conflict as a tool in internal struggles. There are times when the creation of a conflict regarding a certain form of speech becomes a tool wielded in internal community struggles. For example, in the framework of struggles between ultra-Orthodox groups in Jerusalem, criticism was levelled against the ex-Mayor of Jerusalem, the ultraOrthodox Uri Lupolianski who allegedly was not doing enough to prevent the abominable parade. In response the Mayor was compelled to be more abrasive and militant in his opposition to the parade. He even went so far as to compare the Pride parade to placing the head of a pig inside the Al-Aqsa Mosque.47 Other people, who had been anonymous when the dispute broke out, became particularly vocal in the inflaming of passions, giving rise to suspicion that they were motivated not so much by their desire to protect religious sentiments as by the desire to promote their own names and actions (such as Rabbi Yehuda Levin and the councilwoman Mina Fenton). In the cases of the caricatures too one discerns the influence of power struggles on the conduct of those involved. In the beginning of February 2006 two stormy demonstrations were conducted in opposition to the publication of the cartoons, one in Syria and the other in Lebanon. Both demonstrations ended with the Danish embassies being set on fire. In Lebanon a church too was set on fire. In the context of these demonstrations the jurist Zaid Al-Ali said the following: Very few people in Lebanon seem to believe that this weekend’s violence was a true and spontaneous reaction to the Danish caricatures. By and large, the Lebanese interpreted the Damascus protest as a message from the Ba’athist leadership to the outside world: if you topple our government, these are the people that will take our place, so tread carefully. For many Lebanese, the Beirut protest on the following day was another message from Damascus: without Syrian oversight, Lebanon is chaotic and simmers with sectarian hatred.48
The conflict as an economic resource. Finally, the conflict may carry economic dividends. A political storm and inflamed feelings are of tremendous value not only for closing ranks and enlisting supporters, but also, and for the same reasons, for raising financial resources. For example, in the book Perfect Enemies: The Battle Between the Religious Right and the Gay Movement,49 authors John Gallagher and Chris Bull describe how the two political movements exploited local ‘burning issues’ and political storms in order to augment their financial support.50 Here too I can personally attest that during the weeks before the Pride parade in Jerusalem in 2006, there was unprecedented willingness on the part of the donators to the Open House to provide economic support for the battle being waged at that time. These four factors can certainly be supplemented by others, such as the conflict as a tool for ‘letting off steam’, the conflict as a building block of communal identity, and others. At times, the conflict will simultaneously fill all of these roles. The entire phenomenon was articulately depicted in the razor sharp language of the Orthodox journalist B Michael, who related to the claim that the Pride parade would offend the feelings of the religious: See an interview with the Jerusalem weekly Kol Ha’Ir, 20 May 2005. Zaid Al-Ali, ‘The Lebanese Reaction’, openDemocracy, 6 February 2006: www.opendemocracy.net/conflicteurope_islam/muslim_cartoons_3244.jsp#three. 49 J Gallagher and C Bull, Perfect Enemies: The Battle Between the Religious Right and the Gay Movement (Darby PA, Diane Publishing, 1996). 50 As the saying goes: ‘the shriller you are, the easier it is to raise money’. 47 48
254 Yaacov Ben-Shemesh The whole myth of the delicate feelings of the Haredim [ultra-Orthodox] is nothing but a uniquely effective stunt to justify demonstrations of strength, for raising clerical and budgetary demands, and occasionally for providing vital release of the billows of frustrations that pile up in the belly of a pressured populace. For fifteen years I lived on the outskirts of Mea Shearim in the thick of the Haredim. I learnt to respect and admire them, but I also became familiar with ‘their tricks and their shticks’. Everybody who is worried about the feelings of the Haredim should come just once and see the joyful glee that engulfs the Haredi community in anticipation of each of these carefully orchestrated storms, to understand the degree to which this entire business is more for entertainment than for the sake of heaven. The children whoop it up. The young boys quiver in anticipation. The Yeshiva [institution for religious studies] dropouts eagerly seize the opportunity to prove their piety, and the shepherds of the community know that they have supplied their flock with healthy exercise and myths of bravery for future generations.51
IV. CONCLUSION
Summing up, a harsh and extensive offence to religious or other feelings is not, for the most part, a direct result of offensive speech in and of itself. Rather, it is intentionally ‘produced’ for societal, strategic, political, and other reasons. These are all of course legitimate forms of social mobilization and community building, but they have nothing to do with protecting feelings, and everything to do with the typical dynamics of culture wars. When the courts become involved in events of this kind they do not protect feelings; rather, they are dragged into the culture war itself. This, I believe, adds weight to the already strong case for declaring offence to feelings as a categorically illegitimate consideration when deciding issues of free speech in a pluralistic society.
M Brizon, ‘The Parade of Hysteria’, Yedi’ot Aharonot, Tel-Aviv, 3 November 2006.
51
17 Proportionality: Comparative Perspectives on Israeli Debates SUJIT CHOUDHRY*
I. INTRODUCTION
P
ROPORTIONALITY IS A pervasive feature of contemporary constitutional practice under most systems of rights-protection. Although its textual underpinnings are diverse, this diversity in constitutional forms has been overwhelmed by the emergence of a doctrine of proportionality with a common legal structure. After decades of doctrinal development marked by self-conscious and deliberate constitutional convergence, we are now witnessing the rise of a scholarly literature on proportionality that is genuinely comparative, and which is trying to catch up with constitutional practice. Its centre of gravity is proportionality in the strict sense. This may reflect the fact that it is at this stage where the bulk of legal analysis tends to take place in the jurisdictions that command central academic attention, notably Israel and Germany (and now perhaps even Canada).1 Proportionality in the strict sense has generated intense academic controversy on a range of issues. There have been questions over the exact methodology of balancing, with Robert Alexy and Aharon Barak setting out proposals that reconstruct and systematise German and Israeli constitutional doctrine.2 This, in turn, has raised the issue of whether this methodology is rational and hence qualifies as a form of legal reasoning, or whether it is a form of intuitionist ad hocery. These debates are analytically distinct from, but closely related to, a set of institutional concerns regarding the capacity of courts to apply proportionality in a manner that is consistent and predictable across judges and cases and is appropriately shielded from the corrupting influence of judicial preference, questions regarding comparative institutional competence, and the difficulties in adducing and assessing the evidentiary materials seemingly required by the demands of justification. These institutional concerns have spawned a series of doctrinal
* I thank Aharon Barak, Daphne Barak-Erez, and Gideon Sapir for their kind invitation to speak at ‘Israeli Constitutional Law in the Making – Comparative and Global Perspectives’. All remaining errors are mine. 1 A Barak, ‘Proportional Effect: The Israeli Experience’ (2007) 57 University of Toronto Law Journal 369; D Grimm, ‘Proportionality in Canadian and German Constitutional Law’ (2007) 57 University of Toronto Law Journal 383; Alberta v Hutterian Brethren of Wilson Colony [2009] 2 SCR 567 (Can). 2 R Alexy, A Theory of Constitutional Rights (Oxford, Oxford University Press); A Barak, Proportionality: Constitutional Rights and Their Limitations (Cambridge, Cambridge University Press, 2012).
256 Sujit Choudhry proposals that systematically under-enforce proportionality, in Larry Sager’s sense of that term.3 Yet another set of debates concerns whether proportionality undercuts the very idea of rights themselves. Habermas tells us that rights are ‘firewalls’, and draws a sharp distinction between ‘law defined through a system of rights’ that ‘domesticates, as it were, the policy goals and value orientations of the legislator through the strict priority of normative points of view’, and a proportionality analysis under which rights are merely ‘goods and values’ which ‘must compete with the others at the same level for priority’.4 To these critics, proportionality in the strict sense embodies everything that is wrong with the very idea of permitting proportional limits on constitutional rights. By contrast, the question of legitimate objectives has suffered from relative neglect in the literature. This too may reflect constitutional practice, because laws easily clear this step, and judicial scrutiny is relatively minimal. Indeed, courts have sometimes said that legislative objectives are off-limits, and will therefore accept any plausible objective offered by the government without gazing behind it. Why is this the case? David Beatty has offered the most well-developed answer, and argues that proportionality sets up an institutional division of labour between legislatures and courts, where legislatures can pursue whatever policy they want, and courts confine their role to policing legislative means.5 The policy of near complete judicial non-intervention on legislative purposes is accordingly a response to the counter-majoritarian dilemma. Closely related is the idea – set out by Grégoire Webber among others – that proportionality is a strategy to depoliticise judicial review by bracketing deep disagreement over fundamental questions of political principle.6 It purports to lower the stakes of these disagreements by refusing to set boundaries on the acceptable modes of argument in politics, and channelling disagreement into a highly contextualised enquiry rooted in the particularities of an individual case, often turning on questions of instrument choice and empirical evidence. On this view, it is important that there be few if any limits set on the range of permissible legislative objectives, and that statutes whose purposes appear to be unconstitutional be given a sympathetic reconstruction in order to permit the proportionality analysis to occur. But the difficulty is that these goals contradict the very terms of the proportionality test itself. There is an important distinction between the different elements of a proportionality analysis. Steps one and two accept the legitimacy of the objective, and scrutinise the instrumental relationship of means to ends. By contrast, proportionality in the strict sense re-engages the objective itself, and, inter alia, assesses its relative importance in relation to the limitation of rights. Now there is an intramural debate over the extent of the analytical overlap between the initial search for a legitimate objective and the final prong of proportionality.7 But if we put that debate to one side and concede that these two enquiries are not coextensive, it is nonetheless true that they raise similar questions, 3 LG Sager, Justice in Plainclothes: A Theory of American Constitutional Practice (New Haven, Yale University Press, 2004). 4 J Habermas, Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy (Cambridge, MIT Press, 1998) 259. 5 DM Beatty, The Ultimate Rule of Law (Oxford, Oxford University Press, 2004). 6 GCN Webber, The Negotiable Constitution: On the Limitation of Rights (Cambridge, Cambridge University Press, 2009). 7 For competing views, see Barak, ‘Proportional Effect’ (n 1); PW Hogg, Constitutional Law of Canada (Toronto, Thomson/Carswell, 2007).
Proportionality: Comparative Perspectives 257
and do similar work. So the choice is not whether to bracket the scrutiny of legislative objectives entirely, but whether to front end or back end their consideration. What unites the chapters by Ben-Shemesh8 and Kremnitzer9 is that they make the case for front-ending the examination of legitimate objectives, prior to engaging in proportionality analysis. Ben-Shemesh terms this the rejection of balancing in favour of categorical answers at the legitimate objective stage, and trains his sights on the Israeli jurisprudence on offensive expression to argue that the fact that speech causes offence to feelings should not count at all as a legitimate reason for restricting expression. Kremnitzer devotes considerable attention in his chapter to the idea of judicial scrutiny of pretextual purposes that serve as cover for illegitimate objectives. The chapters raise three questions: 1. What purposes are illegitimate, and hence terminate the limitations analysis before one proceeds to proportionality? 2. Can we reinterpret those elements of proportionality that describe themselves as not concerned with the scrutiny of ends and which take those ends as givens – suitability and necessity – as tools to uncover illegitimate purposes? 3. When the state acts in response to the demands of some private parties to restrict the rights of other private parties, how should a court characterise the purposes under lying its actions? II. ILLEGITIMATE REASONS
The chapters set out reasons for state action that are illegitimate and therefore cannot serve as a basis for justifiably limiting constitutional rights. But the reasons offered vary in where they lie on the continuum from the highly specific to the very abstract. BenShemesh focuses narrowly on why the protection of feelings should not count as a legitimate reason to limit free speech – because such feelings are unavoidable in pluralist democracies committed to freedom of expression; because regarding such feelings as legitimate reasons to limit rights creates the perverse incentive to politically mobilise to create these feelings, putting pressure on the state to limit expression and to provide a constitutional case for these limitations; and because only politically powerful interests will be able to conscript the state to limit expression. Kremnitzer is highly abstract. He does not reason up from particulars, and instead starts at the other end from the constitutional order itself, and posits that an illegitimate purpose is one that is inconsistent with the values of the state. The next step in this collective project among Israeli constitutional scholars is to fill the gaps between these positions, and provide a general account of illegitimate reasons that is both based in more abstract principles and explains, justifies and organises answers in particular cases. This is not just an Israeli problem. The question of illegitimate reasons, or unconstitutional motives, has arisen in India, South Africa, Germany, Canada and the United States (and no doubt in other jurisdictions). Israeli constitutional scholars might want to mine comparative constitutional experience for argumentative strategies on how to frame this problem and approach this issue. Ch 16 in this volume. Ch 15 in this volume.
8 9
258 Sujit Choudhry For example, one could take up Kremnitzer’s proposal and infer illegitimate purposes from the very structure of the state – what Indian constitutional jurisprudence calls its basic structure. For example, if the state’s basic structure encompasses a Bill of Rights protected through judicial review, one can infer a commitment to the project of liberal constitutionalism that underlies a Bill of Rights, and hence conditions its interpretation. One could argue that judicial review and Bills of Rights institutionalise the Dworkinian commitment to equal respect and concern, which forbids the state from acting on reasons that evince such a lack of concern or respect – for example, on the basis of sexual orientation, sex, religion or race.10 This kind of commitment underlies the jurisprudence of the Supreme Court of Canada, which per se prohibits sectarian, sexist, homophobic and racist preferences as reasons for the limitations of constitutional rights.11 Or one could reason, as has the Constitutional Court of South Africa, that the very idea of a constitution committed to the rule of law prohibits the enactment of laws for what Cass Sunstein has termed naked preferences. Although Sunstein defines naked preferences narrowly as the ‘distribution of resources or opportunities to one group over another simply because they held and exercised the requisite raw political power’,12 which would render constitutionally suspect any legislation that rewards groups engaging in rentseeking behaviour, the reach of the South African conception of naked preferences is broader, and encompasses any legislative purpose that is arbitrary, capricious and does not plausibly bear further any notion of the public good (and indeed, is a free-standing ground of judicial review that need not be triggered by the infringement of a constitutional right).13 The South African jurisprudence on unconstitutional motives illustrates another point – the importance of history. The notion that the state should be prohibited from acting on the basis of naked preferences is based on a barely concealed view that this is how the South African State conducted itself under apartheid. And so the constitution can be understood as a sort of remedial document, whose foundational premises involve a decisive rejection of the constitutional order on whose ashes it sits – what can be termed the ‘never again’ principle. The idea of the past as a negative, anti-model of constitutional experience, that defines a contemporary constitution by what it is not, is an under- recognised and under-utilised resource that bridges the gap between universalistic principles of political morality and particular national experiences. Another example of this way of distilling illegitimate purposes emerges from Indian constitutional jurisprudence. In Naz Foundation v Government of NCT of Delhi,14 the Delhi High Court has set out the notion of a ‘constitutional morality’ which it derives from a reading of the history of the framing of the Indian Constitution. One element of that history is the idea of the constitution as an instrument of social revolution that obliges the state to dissolve existing social, political and economic hierarchies based on caste, especially with respect to untouchables. Naz Foundation reasons from these premises to declare illegitimate any R Dworkin, A Matter of Principle (Cambridge, Harvard University Press, 1985). R v Big M Drug Mart Ltd [1985] 1 SCR 295 (Can). 12 CR Sunstein, ‘Naked Preferences and the Constitution’ (1984) 84 Columbia Law Review 1689. 13 Pharmaceutical Manufacturers Association of South Africa: Re ex p President of the Republic of South Africa 2000 (2) SA 674 (CC) (S Afr). 14 Naz Foundation v Government of NCT of Delhi (2009) 160 DLT 277 (India). For a detailed discussion of this point, see S Choudhry, ‘How to Do Comparative Constitutional Law in India: Naz Foundation, Same Sex Rights and Dialogical Interpretation’ in S Khilnani, V Raghavan and AK Thiruvengadam (eds), Comparative Constitutionalism in South Asia (Delhi, Oxford University Press, 2013) ch 2. 10 11
Proportionality: Comparative Perspectives 259
state action that is designed to reproduce a caste-like status for other social groups, such as homosexuals. A different approach would identify illegitimate purposes in connection with specific rights. Each right has a point or basic purpose of what sort of interests it is meant to protect. It is per se illegitimate for governments to act for the sole reason of deliberately robbing people of that interest, as opposed to pursuing another social goal that has the incidental effect of infringing the right. The comparative law of democracy furnishes a useful illustration of this idea in practice. One basic goal of the right to vote is to create a system of democratic accountability, whereby citizens choose their representatives through elections. A particular danger faced by democracies is attempts by incumbents to insulate themselves from democratic accountability through the manipulation of the rules governing elections. And so in a series of decisions, the German Constitutional Court has declared that laws enacted for this legislative purpose are unconstitutional, because their very purpose is to subvert the basic objective of the right to vote.15 I conclude this part on a broader note. One of the main objections to proportionality analysis is that it fails to acknowledge that rights-claims should have a special urgency that requires that they be treated differently from the goods and values pursued by liberal democracy. It is said that proportionality, especially proportionality in the strict sense, fails to take seriously the categorical nature of rights claims. These debates have been given renewed force by constitutional cases brought on the basis of the right to life, and the right to not be subject to torture. The decisions of the German Constitutional Court in the German Airplane case,16 and the Israeli Supreme Court in Public Committee against Torture in Israel v State of Israel,17 suggest that some rights should not be subject to proportionality because they are absolute. Arthur Ripstein, for example, has proposed a hierarchy of rights, some of which are immune from proportionality (eg life, bodily integrity), and the remainder that are not (eg property).18 But as these examples make clear, there is another way to structure constitutional doctrine that takes seriously the categorical nature of rights-claims even for those rights without a plausible claim to being absolute. Certain reasons for limiting constitutional rights may be per se inadmissible, either because they contradict the basic structure, mission or premises of the constitutional order, or because they deliberately undermine the very purpose of a particular right. This second kind of reason-based restraint emerging from rights is interesting. As Jeremy Waldron and Richard Pildes have argued, it has long been recognised that one dimension of rights is their reason-blocking function.19 An emerging literature seeks to move this discussion forward, and develop a taxonomy of those rights that connote excluded reasons, and those which do not.20 I cannot develop the point here, but my sense is that every right excludes certain reasons to limit it. Moreover, while some rights may have excluded reasons in common, in 15 S Issacharoff and RH Pildes, ‘Politics as Markets: Partisan Lockups of the Democratic Process’ (1998) 50 Stanford Law Review 643, 690–99. 16 115 BVerfGE 118 (2006) (Ger). 17 HCJ 5100/94 Public Committee against Torture in Israel v State of Israel 53(4) PD 817 [1999] (in Hebrew). 18 A Ripstein, ‘Proportionality without Balancing’ (unpublished). 19 RH Pildes, ‘Why Rights Are Not Trumps: Social Meanings, Expressive Harms, and Constitutionalism’ (1998) 27 Journal of Legal Studies 725; J Waldron, ‘Pildes on Dworkin’s Theory of Rights’ (2000) Journal of Legal Studies 301; RH Pildes, ‘Dworkin’s Two Conceptions of Rights’ (2000) 29 Journal of Legal Studies 309. 20 K Möller, ‘Two Conceptions of Positive Liberty: Towards an Autonomy-based Theory of Constitutional Rights’ (2009) 29 OJLS 757.
260 Sujit Choudhry many cases, those reasons will differ. The implications for proportionality analysis are important. It is often claimed that there is a trend towards a generic proportionality analysis in which rights have disappeared from the picture. This has been fuelled by the proliferation of rights, the expansion of the scope of rights to encompass a broad right to negative liberty, and the trend towards generic proportionality clauses. Bills of Rights, in the view of David Beatty, Moshe Cohen-Eliya, Mattias Kumm, and Iddo Porat, create a general right to justification.21 But there is another possibility, in which each right steers the court and orients its proportionality analysis in a different direction that flows from the very nature of that right itself. There may not be a generic proportionality analysis, but rather a family of ‘proportionalities’, each with a subtly different character shaped by the scope of permissible reasons permitted by the right in question. III. REINTERPRETATION OF MEANS/ENDS ANALYSIS AS THE SEARCH FOR OBJECTIVES
Steps two and three of proportionality proceed from the assumption of a legitimate objective, and scrutinise the relationships of means to ends. Step two assesses the suitability of the means to achieve the objective, whereas step three determines its necessity through a search for other equally effective alternatives that meet the same aim. Throughout this analysis, the objective is treated as unassailable. But can we reinterpret these steps in a manner that treats the objective not as a given, but as provisional, and which views the probing of the relationship between means and ends as a tool to uncover illegitimate purposes? The particular problem here is that of pretextualism, where the government publicly offers as justification a legitimate objective for state action that merely serves to mask its true purpose, which is illegitimate. This is a problem that Kremnitzer addresses in his chapter. The most direct way for a court to flush out and expose illegitimate motives is through a careful examination of the factual record, to see if the evidence adduced supports the existence of the problem the challenged measure purports to address, and the claim that this problem was in fact the mischief that prompted the measure. The absence of evidence for a legitimate objective would be sufficient grounds to hold for the rightsclaimant. But in the face of the suspicion that an illegitimate objective was really at work, the absence of evidence may also provide grounds for a court to draw the inference that what really motivated the objective were illegitimate considerations. But another technique to flush out illicit motives is to take an equally sceptical approach to scrutinizing the relationship between means and ends. The failure of a measure to meet the demands of suitability or necessity does not merely indicate that the government has chosen constitutionally prohibited means to pursue its stated goals. Rather, it is evidence that casts doubt on the credibility of the government’s assertion of its objective, and may point to a hidden agenda that is constitutionally illegitimate. The 21 Beatty (n 5); M Kumm, ‘Political Liberalism and the Structure of Rights: On the Place and Limits of the Proportionality Requirement’ in S Paulson and G Pavlakos (eds), Law, Rights and Discourse: The Legal Philosophy of Robert Alexy (Oxford, Hart Publishing, 2007) 131; M Kumm, ‘The Idea of Socratic Contestation and the Right to Justification: The Point of Rights-Based Proportionality Review’ (2010) 4 Law and Ethics of Human Rightss 142; M Cohen-Eliya and I Porat, ‘Proportionality and the Culture of Justification’ (2011) 59 American Journal of Comparative Law 463.
Proportionality: Comparative Perspectives 261
idea that the lack of suitability between means and stated ends is a sign of constitutional trouble for those ends is an old one. The complete ineffectiveness of means to fulfill stated ends is one scenario in which this arises. Another is where those means produce perverse effects – that is, effects precisely opposite to those intended. The inference is that a government would have known in advance that measures were ineffectual or counter-productive, and that if it nonetheless persisted in adopting them, its true motivations must lie elsewhere. Much less attention has been devoted to the power of the enquiry into necessity to do the same work. This flows from the way in which courts approach this question – through a static comparison of the comparative effectiveness of alternative means to achieve the legitimate objective. This may be a highly artificial exercise, engaged in strictly for the benefit of the court. But there is another way to conceptualise the judicial enquiry into necessity. As Kremnitzer suggests, the prospect of judicial review should create the incentive to address alternative means and their comparative effectiveness during the policy process, including in legislative debates. If so, the judicial assessment of necessity can be reframed as a probing enquiry into the executive and/or legislative process, and to examine whether and how alternatives were considered. Following David Dyzenhaus, we can think of this as a way of proceduralizing proportionality analysis.22 Dyzenhaus’ principal example is the administrative decision at issue in the House of Lords decision in Begum,23 and his target is not legitimate objectives, but the question of necessity itself. But this method can be adapted to the legislative process and reframed around whether the stated objective is pretextual. To be sure, the lack of a serious consideration of alternatives can mean different things. It may simply reveal one or more of Rosalind Dixon’s legislative blind spots (of application, perspective, and/or accommodation), and not impugn the credibility of the stated objective.24 But the failure to seriously consider equally effective alternatives to achieving the stated objective may raise the suspicion that this is not the true objective at all. Rather than a lack of foresight, a limited world-view, or indifference to rights, the true problem may indeed be bad faith. IV. THE STATE AND THIRD PARTIES
Proportionality proceeds on the basis of an account of the state’s reasons for action. If there are purposes that are illegitimate, their illegitimacy in many cases flows from the fact that the state cannot act upon these considerations. However, the question of whether the constitutional order permits private parties to act on the same sort of reasons is much more complex. Where Bills of Rights are principally vertical in their application, parties may often act on the basis of reasons off-limits to the state. Where Bills of Rights have horizontal application, the scope for private actors to act on these grounds is narrower, but still exists, especially if those actions fall within the ambit of a constitutional right. Religion is a useful illustration. The state cannot act for sectarian reasons. However, private parties can, and indeed, their right to do so is protected by freedom of religion. D Dyzenhaus, ‘Proportionality and Deference in a Culture of Justification’ (unpublished). R (Begum) v Headteacher and Governors of Denbigh High School [2006] UKHL 15, [2007] 1 AC 100. 24 R Dixon, ‘Creating Dialogue about Socioeconomic Rights: Strong-Form versus Weak-Form Judicial Review Revisited’ (2007) 5 International Journal of Constitutional Law 391. 22 23
262 Sujit Choudhry This raises interesting questions when the state acts on behalf of third parties, as it did in the dispute underlying the Gay Pride Parade litigation (discussed by Ben-Shemesh). When the state acts in response to the demands of one group of private individuals to restrict the rights of another group of private individuals, because the first group objects on religious grounds to the activities of the second group, how should a court characterise the purposes underlying the state’s actions? The answer to this question depends on how the state describes its relationship to the private parties. There are two options. On one account, the state stands above social divisions and adjudicates impartially among them. When the state restricts one party’s rights at the other party’s behest, the state acts for its own reasons and not those of any private party. But on a second account, the state sides with one private party against another – it represents it, is a front for it, is captured by it, and/or is beholden to it. By siding with one party, it adopts and acts for its reasons. Which of these two accounts are correct matters when the state acts in response to religious objections raised by private parties, because religious reasons are illegitimate objectives. From an outsider’s perspective, this is the issue at the heart of the Gay Pride Parade litigation, not the somewhat different question of whether it is legitimate for the state to regulate offensive speech. So the Court had to choose between two accounts of the state’s objectives. The state’s version would be that it was not acting for religious reasons, but in response to the consequences of other parties holding religious reasons that the state was itself prohibited from holding. The counter-narrative would be that this is precisely what the state was doing. This is the problem of pretextual purposes, with a twist – the state’s constitutional case is dependent on the existence of private reasons, yet would fail if those reasons were attributed to it. This is a delicate line to walk. Ben-Shemesh’s arguments for why the state should be constitutionally prohibited from regulating offensive speech provide the beginning of an answer as to how a court should tackle the question of the state’s true motives. He writes that the regulation of offensive speech ‘is most likely to be used by those who have political power, and against those who lack it’. The question is how one determines who wields political influence. In the liberal constitutional tradition, John Hart Ely provided a highly influential answer to this question.25 For Ely, the central determinant of the intensity of judicial review was the notion of a ‘discrete and insular minority’, which is permanently excluded from the exercise of political power. In justifying restrictions on speech, the state may assert that it is acting as a third party to protect a minority that lacks political power and cannot commandeer the state to serve its ends. The state is acting as a mediator among the interests of competing groups. But a court should not accept the state’s bare assertion that a discrete minority is necessarily a political minority. To do so would be to conflate a minority’s discreteness and insularity – along the dimensions of race, ethnicity, religion or degree of religious observance – with a lack of political power. As Kremnitzer acutely observes, minorities that are discrete and insular may nonetheless be ‘politically extremely powerful’. Although he does not develop the point and explain why, the answer is clear to even a casual, outside observer of Israeli politics. In a highly fractured polity such as Israel, discrete and insular minorities may be indispensable coalition partners, which grants them real political 25 JH Ely, Democracy and Distrust: A Theory of Judicial Review (Cambridge, Harvard University Press, 1980).
Proportionality: Comparative Perspectives 263
power. But the precise configuration of political power and influence is a complex empirical question, the answer to which varies across countries and within countries over time. And so to determine the real purposes underlying the state’s actions, one would have to engage in a detailed analysis of the political dynamics at play in each case. Israeli constitutional scholars need to integrate this kind of fine-grained political analysis into the application of the doctrine of proportionality. Indeed, the raw materials of Israeli constitutional politics may provide them with a better opportunity to do so than in perhaps any other rights-protecting liberal democracy. But Israel is not alone in grappling with the issue of how to interpret a Bill of Rights in a divided society, and more precisely, in calibrating the doctrine of proportionality to this kind of political context. As this volume attests to, Israel is a constitutional laboratory in many areas of constitutional law. If Israeli scholars and courts wrestle openly and courageously with the problem of unconstitutional purposes in a fraught and fragmented political context, they will have much to teach the world.
18 Human Dignity as a Central Pillar in Constitutional Rights Jurisprudence in Israel: Definitions and Parameters TAMAR HOSTOVSKY BRANDES
I. INTRODUCTION
I
N 1992, THE Knesset (Israeli Parliament) enacted two Basic Laws, Basic Law: Freedom of Occupation and Basic Law: Human Dignity and Liberty. The enactment of the two laws was hailed as the first step in establishing an Israeli Bill of Rights and was referred to by jurists as marking the beginning of a ‘Constitutional Revolution’ in Israeli law.1 The two laws were enacted as part of an ongoing process rooted in a 1950 Knesset decision known as the ‘Harari Decision’,2 according to which the Knesset would enact a series of Basic Laws which, together, were intended to eventually form the Israeli Constitution. The Harari Decision was a result of a political compromise which aimed to resolve political disagreements regarding the desired content of the future Constitution. The gradual process was thought to enable the Knesset to first address less controversial rights while postponing discussion of rights perceived as more problematic to a later stage. Among the issues considered problematic were the right to equality, which some members of the Knesset viewed as possibly conflicting with religious laws governing matrimonial matters and with certain aspects of the character of Israel as a Jewish state, and freedom of religion, which raised similar issues. However, more than 40 years after the Harari Decision, no consensus was reached with regard to these rights, and they, together with other rights, were thus intentionally left out of the two 1992 Basic Laws. The 1992 Basic Laws were therefore not only born out of a political compromise, but also shaped by this compromise. Two main issues needed to be resolved by the courts after the enactment of the two Basic Laws. The first was the scope of the Basic Laws; the second was their status within Israeli law and in particular their implications for judicial review of primary Knesset 1 The term ‘Constitutional Revolution’ referred to the enactment of the laws and in particular to the Supreme Court’s ruling in the case of United Mizrahi Bank which determined that the Court could declare as void laws that were contrary to the Basic Laws. See: CA 6821/93 United Mizrahi Bank Ltd v Migdal Cooperative Village 49(4) PD 221 [1995] (in Hebrew). 2 DK 5 (1950) 1743 .
268 Tamar Hostovsky Brandes legislation. The second question was resolved relatively shortly after the Laws’ enactment. In the 1995 decision of United Mizrahi Bank,3 the Israeli Supreme Court determined that the Basic Laws had supra-legislative, constitutional status and, accordingly, that the Court had the power to declare void primary legislation that contradicted the Basic Laws. While the decision was initially criticised by many, it has since been reaffirmed on numerous occasions.4 The dispute with regard to the scope of the Basic Laws, however, and in particular the scope of Basic Law: Human Dignity and Liberty (hereinafter: the Basic Law), is ongoing. The Basic Law explicitly protects the right to life, body and dignity of the person (sections 2 and 4), the right to property (section 3), the right to personal liberty (section 5), the right of all Israeli nationals to enter the country and the right of all people to leave it (section 6), and the right to privacy (section 7). All of these rights had already been recognised by the Court prior to the enactment of the Basic Laws. With the Basic Law’s enactment, however, and following the United Mizrahi Bank decision, these rights were elevated to the status of constitutional rights.5 The enactment of the 1992 Basic Laws raised questions regarding the status of other rights previously recognised in case law but not explicitly named in either of the two Basic Laws. These included rights as important and central as the right to equality, freedom of religion and freedom of expression. Such rights were referred to as the ‘unnamed’ or ‘unenumerated’ rights.6 The dispute centred on the question whether the Court could interpret the 1992 Basic Laws as awarding constitutional protection to such unenumerated rights. The present chapter examines the manner by which the Court determines which unnamed rights fall within the scope of the Basic Law. It examines the models of interpretation applied by the Court when recognizing certain unnamed rights while refusing to recognise others. The chapter argues that different approaches with regard to recognition of particular rights are rooted in different value-based world-views regarding the essence of human dignity, and argues that the lack of a clear test for the recognition of unnamed rights results from the absence of a robust, well-developed notion of human dignity in Israeli constitutional law. The chapter purports to demonstrate the import ance of developing a concept of human dignity that encompasses the different values and world-views that exist in Israeli society. The chapter proceeds as follows: part II presents the controversy regarding deriving unenumerated rights from the right to human dignity; part III examines various models of interpretation of the right to human dignity; part IV proposes an understanding of the concept of human dignity that is based on the protection of different world-views regarding the unique worth of human life; and part V summarises the proposed concept of human dignity and indicates the challenges the Court will face when interpreting this concept. United Mizrahi Bank (n 1) 352. For general criticism, see M Landau, ‘Giving Israel a Constitution Through Supreme Court Rulings’ (1996) 3 Law and Government 697 (in Hebrew); R Gavison, ‘The Constitutional Revolution – a Description of Reality or a Self-fulfilling Prophecy?’ (1997) 28 Mishpatim 21 (in Hebrew). A controversial Bill currently before the Knesset proposes to confer the right to declare a law void upon a nine-justice panel of the Supreme Court. According to the Bill, a majority of 65 Knesset Members would be able to reconstitute a Bill declared void by the Supreme Court. See Bill Memorandum of Basic Law: The Legislation, 2012, published by the Ministry of Justice: www.tazkirim.gov.il/Tazkirim_Attachments/41283_x_AttachFile.doc (in Hebrew). 5 United Mizrahi Bank (n 1). 6 The use of the term ‘unenumerated rights’ has been criticized. See, eg A Barak, Proportionality in Law: The Constitutional Right and its Limitation (Tel-Aviv, Nevo, 2010) 78–80 (in Hebrew). 3 4
Human Dignity: Definitions and Parameters 269
II. ANCHORING UNNAMED RIGHTS IN THE RIGHT TO HUMAN DIGNITY
In an article written shortly after the enactment of the Basic Laws, Hillel Sommer outlined the different aspects of the debate spurred by the enactment of the Basic Laws.7 One aspect of the debate, explained Sommer, regarded the question whether unenumerated rights could be at all derived from the Basic Law. Among those who believed that unenumerated rights could, in general, be derived from the Basic Law, there were disputes regarding the methods by which such rights could be recognised, as well as the particular rights that should be recognised.8 Although a number of prominent jurists argued, at the time, that the Court should exercise restraint in interpreting the scope of the Basic Law,9 the position that the Basic Law could indeed be interpreted to include additional rights eventually prevailed. In the two decades that have passed since the establishment of the Basic Law, the main basis for recognizing unenumerated rights under the Law has been the interpretation of the right to human dignity, protected by section 2, which determines that ‘there shall be no violation of the life, body or dignity of any person as such’ and section 4, which provides that ‘All persons are entitled to protection of their life, body and dignity’. Considering that the language of the law provides for at least four other explicit rights, it is surprising how little judicial deliberation there has been regarding the possibility of deriving unenumerated rights from sources other than the right to human dignity.10 The right to liberty, for example, has been interpreted as referring mainly to the loss of physical freedom. The right to privacy, explicitly protected by section 7 of the Basic Law,11 has also been interpreted rather conservatively, certainly in comparison to some interpretations of privacy suggested in US constitutional law. With few exceptions, then, the right to human dignity has served as the main source of unenumerated rights.12 The right to human dignity has been argued to encompass, among other rights, freedom of religion,13 freedom of expression,14 the right to equality15 and the right to a minimal standard of living.16 Some have argued that it includes the 7 H Sommer, ‘The Non-Enumerated Rights: On the Scope of the Constitutional Revolution’ (1997) 28 Mishpatim 257 (in Hebrew). 8 ibid 264–67. 9 Landau (n 4) and Gavison (n 4). 10 Sommer indicates that two other possible sources of unenumerated rights were noted by jurists, but quickly rejected. The first was s 1A of the Law (the ‘purpose clause’), which states that ‘the purpose of this Basic Law is to protect human dignity and liberty, in order to establish in a Basic Law the values of the State of Israel as a Jewish and democratic state’. The second was the right to liberty. See Sommer, ‘The Non-Enumerated Rights’ (n 7) 281–87. 11 Art 7 of the Basic Law determines that: ‘(a) All persons have the right to privacy and to intimacy. (b) There shall be no entry into the private premises of a person who has not consented thereto. (c) No search shall be conducted on the private premises of a person, nor in the body or personal effects. (d) There shall be no violation of the confidentiality of conversation, or of the writings or records of a person’. 12 See HCJ 6427/02 Movement for Quality Government v Knesset 61(1) PD 619 [2005] paras 31 and 33–35 of Court President Barak’s opinion (in Hebrew). 13 See A Barak, Interpretation in Law: Constitutional Interpretation, vol 3 (Jerusalem, Nevo, 1994) 225 (in Hebrew); HCJ 10907/04 Solodoch v Municipality of Rehovot (1 August 2010), Nevo Legal Database (by subscription) (in Hebrew). 14 See, eg HCJ 4804/94 Station Film v Council of Film Supervision 50(5) PD 661, 674–75 [1997] (in Hebrew); HCJ 10203/03 Hamifkad Haleumi v Attorney General 62(4) PD 715 [2008] paras 21–27 (in Hebrew). 15 Movement for Quality Government v Knesset (n 12) paras 36–41. 16 HCJ 366/03 Commitment to Peace and Social Justice Association v Minister of Finance 60(3) PD 464, 481 [2005] (in Hebrew).
270 Tamar Hostovsky Brandes right to education,17 the right to an adequate standard of living,18 and the right to a clean environment.19 The debates on the constitutional status of these rights produced an extensive body of case law referring to the right to human dignity. The concept of human dignity has thus become the central pillar of constitutional rights jurisprudence in Israeli law. Little research has been conducted on why the right to human dignity has obtained such a prominent role in comparison to other rights explicitly protected by the Basic Law. One possible explanation is that, in contrast to other sections of the Basic Law, the sections pertaining to human dignity provide little guidance on the content of the right to human dignity or the direction in which it should be understood, thus leaving it open for interpretation. For example, section 7(a), establishing the right to privacy, states that ‘all persons have the right to privacy and to intimacy’, sections 7(b), 7(c), and 7(d) specify protections awarded against ‘entry and searches of private premises’ and the body, as well as against violations of the privacy of conversations and records. Section 5, establishing the right to liberty, determines that ‘there shall be no deprivation or restriction of the liberty of a person by imprisonment, arrest, extradition or otherwise’ (emphasis added). While it is possible to interpret section 7(a) or the term ‘otherwise’ in section 5 elaborately, these sections do provide some indication of the types of protections they guarantee. No similar specifications are offered for the right to human dignity, thus leaving it completely open to judicial interpretation. This explanation, however, is only partial, as there are other rights, most notably the right to property, which are also phrased in general terms. The explanation for the frequent turn to human dignity lies not in the generalness of the term human dignity but in its vagueness, and in particular in the lack of agreement on what the ‘natural’ meaning of the term human dignity is. While there appears to be general agreement on the core of rights such as liberty and privacy, even if there are disputes regarding the inclusion of certain peripheral issues, the cases discussed below demonstrate that there is little agreement on the core of the right to human dignity. This entails, on the one hand, that judicial interpretation of the right to human dignity will be more controversial, but, on the other hand, that it is also inescapable. The main weakness of the term human dignity is thus also its main strength. III. NARROW VERSUS BROAD MODELS OF INTERPRETATION AND DELIBERATE VAGUENESS IN CONSTITUTIONAL INTERPRETATION
The controversies regarding the recognition of both unenumerated rights in general and specific rights thus revolved around the interpretation of the term human dignity in the operative articles of the Basic Law. Early attempts of framing the debate portrayed it as a dispute between narrow or broad methods of interpretation. In his 1994 book on inter See discussion in Y Rabin, The Right to Education (Tel-Aviv, Nevo, 2002) 381–87 (in Hebrew). See discussion in A Gross and D Barak-Erez, ‘Social Citizenship: The Neglected Aspect of Israeli Constitutional Law’ in D Barak-Erez and A Gross (eds), Exploring Social Rights: Theory and Practice (Oxford, Hart Publishing, 2007) 243–61. 19 The claim that a right to a clean environment can be derived, in whole, from the right to human dignity was rejected in HCJ 4128/02 Adam Teva Va Din v Prime Minister of Israel 58(3) PD 503, 518 [2002] (in Hebrew). 17 18
Human Dignity: Definitions and Parameters 271
pretation, Barak, for example, distinguished between a narrow model of interpretation, under which the right to human dignity encompasses protection against violations he viewed as ‘classic’ violations of human dignity such as humiliation, and a broad model, according to which the term human dignity encompasses all human rights. Between these two approaches, argued Barak, there was a third, intermediate approach. Under this intermediate approach the notions of dignity and liberty in the Basic Law include more than just protection against humiliation but do not include every right that can philosophically be derived from them. According to the middle ground model, which Barak viewed as the correct model, the right to human dignity includes those rights that the enlightened public in Israel views as closely related to human dignity and liberty. These included, for example, the right to equality, freedom of expression and freedom of religion, but not the right to education.20 Yehudit Karp offered a similar model, based on three ‘circles of rights’.21 The first circle, according to Karp, includes the immediate, ‘natural’ meaning of the term dignity, which she viewed as the opposite of humiliation, embarrassment and disgrace. The second circle included additional specific rights that derive from the principle of human dignity, such as the right to personal freedom, the right to life, the right to property, the right to privacy, and the right to enter and exit the country. The third circle included freedom and autonomy in the more general, broad sense.22 Under both Karp’s model and Barak’s model the question becomes how far judicial interpretation should venture beyond the ‘core’ right protected by the right to dignity in protecting rights under the term human dignity. It would appear that under both Karp’s and Barak’s models this ‘core’ right would encompass protection against humiliation, which they perceived as the ‘plain’ meaning of the words of the Basic Law. The interpretation models offered by Barak and Karp do not provide a satisfying explanation for why protection against humiliation constitutes the core of the right to human dignity and the nature of the conceptual linkage between the ‘core’ of the right to human dignity and the rights protected under the broader models of interpretation. There are two ways of understanding the proposition that protection against humiliation is the central protection granted by the right to human dignity. The first is that protection against humiliation is the core of the right to human dignity. Under this interpretation, the broader interpretations should somehow be linked to this core, meaning, they should, in some sense, be an expansion of the right not to be subject to humiliation. Neither Barak nor Karp seem to adopt this view, and as will be demonstrated below, with the exception of a few justices, it has not been adopted by the Court. The second way of understanding the centrality of the protection against humiliation is by defining some other right or value as the core of the right to human dignity and then arguing that protection against humiliation is closely linked to this right or value, that the rights included in the second ‘circle’ under Karp’s model or the ‘intermediate’ approach under Barak’s model are more remotely linked to it, and the rights included in the third ‘circle’ or the ‘broad’ approach are even further remote. Neither Barak nor Karp, however, provide an indication of what such right or value may be. The justification provided by Barak, Interpretation in Law (n 13) 413–18. Y Karp, ‘A Few Questions Regarding Human Dignity According to Basic Law: Human Dignity and Liberty’ (1995) 25 Mishpatim 129 (in Hebrew). Karp’s analysis is also discussed by Barak, Interpretation in Law (n 13) 418–19. 22 ibid 136–42. 20 21
272 Tamar Hostovsky Brandes Barak, in his early writings and ruling, for identifying the rights included within the right to human dignity is that the ‘enlightened public’ viewed certain rights as ‘closely related’ to human dignity. A more accurate definition, perhaps, would be that the ‘enlightened public’ views such rights as components of the right to human dignity. The term ‘enlightened public’ has been criticised widely for implying that some sectors of Israeli society were more ‘enlightened’ than others.23 It has also been argued that the term disguises the fact that the Court is actually imposing its own values. In addition to undermining the legitimacy of the Court’s interpretation of the concept of human dignity, the ‘enlightened public’ test added little to the understanding of the nature of the notion of human dignity or to explaining its prominent role in Israeli rights jurisprudence. The lack of a substantive notion of human dignity is evident not only in scholarship but also in case law. The earliest cases recognizing ‘unnamed’ rights were based, for the most part, on the perception that the rights recognised were identifiable as falling within the ‘natural’ meaning of the term human dignity, which encompassed the first and perhaps second circles under Karp’s approach and fell within the ‘intermediate’ model under Barak’s approach. In many of these cases, little or no explanation was offered for why particular rights were perceived as ‘natural’ parts of the right to human dignity. Differences between judges were described as differences in their willingness to interpret ‘broadly’. In the 1994 judgment of Hupert v Yad Vashem, for example, Justice Or expressed the view that ‘today the principle of equality can be anchored in Basic Law: Human Dignity and Liberty’.24 He provides no explanation for this statement other than a reference to Barak and Karp’s articles discussed above. In the judgment in El-Al v Daniloviz, also from 1994, which concerned the right of a flight attendant’s same-sex partner to receive spousal benefits, Justice Barak adopted Justice Or’s position in Hupert and expressed the view that the ‘enactment of Basic Law Human Dignity and Liberty anchored equality as a constitutional right within the framework of human dignity’.25 Similarly, in the 1994 case of Dayan v Wilk, which revolved around the right to demonstrate outside the private residence of a public figure, Justice Barak simply stated, with regard to the right to freedom of expression that it appears that now this right can be derived from the Basic Law: Human Dignity and Liberty, which provides a statutory constitutional basis for the human right to dignity and liberty. The freedom to express ones’ self – in words alone or by expressive actions – is a major expression of human dignity and liberty.26
This lack of explanation for and analysis of why a particular right is ‘anchored’ in the right to human dignity or is a ‘major expression’ of it is surprising considering the implications such recognition had on its status. It is especially surprising in the light of the fact that the Court’s rulings were closely scrutinised by both the public and the Knesset that was waiting to see how the enactment of the two Basic Laws played out, and the fact that the question whether unnamed rights could even be derived from the Basic Law 23 See, eg R Shamir, ‘The Politics of Reasonableness: Reasonableness and Judicial Power in Israel’s Supreme Court’ (1994) 5 Theory And Criticism 7, 13–14, 20–21 (in Hebrew). Barak himself expressed restrained regret for using the term, claiming that it was misunderstood. 24 HCJ 5394/92 Hupert v Yad Vashem 48(3) PD 353, 362 [1994] (in Hebrew). 25 HCJ 721/94 El-Al v Daniloviz 48(5) PD 749, 760 [1994] (in Hebrew). 26 HCJ 2481/93 Dayan v Wilk 48(2) PD 456, 468 [1994] (in Hebrew).
Human Dignity: Definitions and Parameters 273
was still disputed.27 In the sections below, I will try to examine the underlying con siderations and justifications for which the Court recognised some rights but refused to recognise others and the changes in the Court’s reasoning throughout the years. A. The Relevance of the Legislator’s Intention In evaluating the different models of interpretation, specific attention was given by jurists to the question of the weight that should be accorded to legislative intent in interpreting the Basic Law. The debate was particularly heated with regard to recognition of rights that were intentionally omitted from the text of the 1992 Basic Laws, the right to equality being the paradigmatic example. The legislative history of the two Basic Laws clearly indicates that the right to equality was left out of the text of the Basic Law as part of the political compromise that enabled its enactment. Amnon Rubinstein, who was actively involved, as a Knesset Member, in the passing of the Law, explained that the Law included no general equality clause because such a clause would have prevented its acceptance.28 The same applies to the exclusion of the freedom of expression – in fact, separate Bills proposing to accord freedom of expression constitutional status were rejected by the Knesset – and freedom of religion.29 Yet each of these rights has been determined by the Court to be protected by the Basic Law. The legal and political debate echoed, to some extent, the debate in the United States between originalists and ‘living constitution’ advocates.30 Indeed, Barak referred to the American example in his book mentioned above. Two things, however, distinguish the Israeli discussion from the American debate. The first is the short time that has passed since the passing of the Basic Law in comparison with the time that has passed since the enactment of the US Constitution. One of the central claims made by proponents of the ‘living constitution’ approach was that, in order for the constitution to remain relevant over time, constitutional interpretation must take into consideration the conditions, norms and values of the time of interpretation. The relevance over time argument, however, loses much of its force in the Israeli case in light of the short time that has passed since the enactment of the two Basic Laws. This is particularly true in light of the fact that there is little indication that either the social understanding or the disagreements that existed at the time of enactment have since changed. If anything, the contrary could be argued. The relevance over time argument thus could not, on its own, justify interpreting the right to human dignity, for example, contrary to legislative intent. Another argument often made by non-originalists regards the difficulty of identifying the framers’ intention, 27 Justice Barak did indicate, in his scholarly writing, his position with regard to the centrality of the autonomy of free will. See, eg A Barak, ‘Human Dignity as a Constitutional Right’ in H Cohen and I Zamir (eds), Selected Writings (Tel-Aviv, Nevo, 2000) 417, 425–26 (in Hebrew). This was noted by Knesset Member Gafni, who was quoted by Sommer saying that ‘the Supreme Court announces its opinions in lectures’, H Sommer, ‘From Childhood to Maturity: Open Questions in the Execution of the Constitutional Revolution’ (2004) 1 Law and Business 59 (in Hebrew). 28 A Rubinstein and B Medina, The Constitutional Law of the State of Israel, 5th ed (Tel-Aviv, Schocken, 1996) 956. 29 ibid 961. 30 See G Liu, PS Karlan and CH Schroeder, Keeping Faith with the Constitution (Oxford, Oxford University Press, 2009) 25–26.
274 Tamar Hostovsky Brandes especially with the passing of time. Here, too, the Israeli situation differs, as the deliberations of the Knesset are clearly documented and the Knesset has restated its position by refusing, for example, to recognise freedom of expression as a constitutional right in a separate law designated for that purpose. The second important difference between the debate in Israel and in the United States on this issue is the fact that the enactment of the 1992 Basic Laws was not intended to be the last step of the Israeli constitutional enterprise. The plan and hope of the proposers and drafters of the Basic Laws was that they will be followed by additional Laws addressing rights that were not included in the two Basic Laws. Sommer expressed the concern, which some may argue has materialised, that judicial interpretation of the Basic Law in a manner that is clearly contrary to the legislator’s intention would cause the Knesset to refrain from enacting additional Basic Laws and hinder the creation of an Israeli Constitution.31 While there is clear indication that some rights were intentionally left out of the Basic Law, there is little indication of what protections the legislator had intended to include in the term human dignity. This is probably due to the fact that no single understanding of the term human dignity would be acceptable to all Knesset Members that voted in favour of the Basic Law. The concept of human dignity is a vague, value-based term that was used, to a large extent, precisely because no agreement could be reached on more concrete principles. Legislative intention thus provides little guidance for positive interpretation of the right to human dignity. The position reflected in current case law with regard to the importance of the Knesset’s intention has been somewhat inconsistent. The approach that attributed significant weight to the intention of the Basic Laws’ drafters has, in general, been rejected by the Court,32 and while there appears to be agreement that the legislator’s intent cannot be ruled to be of no value at all as a source of interpretation, it is accorded, in most cases, minimal value. With regards to the principle of equality, for example, while the Court has acknowledged the fact that the principle of equality has intentionally been omitted from the Basic Law, the dominant view within the Court is that it is nonetheless protected by the Basic Law, at least to a certain extent. However, the legislator’s intention, and in particular the fact that there was no polit ical agreement which enabled the Knesset to explicitly recognise certain rights, was perceived by the Court as relevant when examining whether, for example, social rights could be derived from the Basic Law. In the Commitment to Peace and Social Association judgment, for example, which will be discussed in length below, the petitioners argued that reductions in income supplement benefit provided by the state violated the right to human dignity, and were thus unconstitutional.33 Justice Barak indicated that ‘social rights are not mentioned expressly in the Basic Laws’ and that ‘various legislative proposals exist in this regard, but these have not yet matured’. In a striking distinction from his position with regard to the protection of the principle of equality by the Basic Law, this leads him to the conclusion that ‘In such a situation it cannot be said that the existing Basic Laws give full and complete protection to social rights’.34 31 Sommer, ‘The Non-Enumerated Rights’ (n 7) 333–34. It can be debated, of course, whether courts should weigh such considerations at all. 32 This position was expressed in earlier decisions by, eg Justice Dorner in the Miller decision discussed below, see HCJ 4541/94 Miller v Minister of Defence 49(4) PD 94 [1995] (in Hebrew). 33 See n 16 above, 475. 34 ibid.
Human Dignity: Definitions and Parameters 275
In practice, thus, the Knesset’s intention when enacting the Law, while cited by the Court when rejecting the recognition of certain rights, cannot explain the recognition of unenumerated rights or the Court’s interpretation of the concept of human dignity. B. Justifying the Recognition of Unnamed Rights and the ‘Partially Encompassed’ Doctrine I argued above that early case law provides little insight on the basis for recognizing some rights and refusing to recognise others. Later cases elaborate in somewhat greater detail about the underlying justifications for recognizing certain rights. As will be demonstrated below, the elaborative modes of judicial reasoning came alongside a narrowing of the scope of the rights recognised. Justice Dalia Dorner’s opinion in the case of Miller v Minister of Defence35 is the first to utilise this approach. Though Justice Dorner’s opinion in this case has been widely referenced in both judicial decision and in scholarship, it has influenced and shaped the judicial discussion on unnamed rights in ways that are yet to be addressed. Alice Miller, the petitioner, was an Israeli woman who challenged the Israeli Army’s refusal to allow women to become combat pilots. The question that arose was whether the Army’s refusal constituted prohibited discrimination.36 The discussion revolved, in part, around the constitutional status of the principle of equality following the enactment of the Basic Law. Justice Dorner argued that certain violations of equality, those that were based on discrimination on the basis of group membership, were prohibited under the Basic Law as they amounted to violation of human dignity. Such discrimination, argued Dorner, was based on perceiving members of certain groups, in this case women, as inferior to other members of society, in this case men.37 Treating someone as inferior amounted to humiliating treatment, which was prohibited by the protection against violations of human dignity provided by the Basic Law. Quoting from the US Supreme Court decision in Brown v Board of Education,38 Dorner explained that treating people differently based on group membership sends a message of inferiority to the members of the group and therefore constitutes humiliation forbidden under the Basic Law.39 It is interesting to note the different roles played by the principles of equality and dignity in the two decisions, a difference that Dorner failed to note. In the United States, the right to equality is constitutionally protected, and the notions of dignity and protection against humiliation were employed to interpret it. In Israel, it is the right to dignity that enjoys constitutional status, and the question discussed was whether, and to what extent, the right to equality was included in it. Both judgments, however, relied on the crossroad between equality and dignity, determining that classifications which are humiliating were unconstitutional.40 Miller (n 32) 131–35. ibid. Justice Mazza thought the differences the military cited were not proven to be relevant, pp 109–16. Justice Strasberg-Cohen thought they were relevant, but could be rectified, p 119. Justices Kedmi and Tal thought the differences were relevant in the current situation, 117–18 and 126–27. 37 ibid 129–33. 38 Brown v Board of Education 347 US 483 (1954). 39 Miller (n 32) 132–33. 40 A similar approach was taken in the Canadian case of Law v Canada (Minister of Employment and Immigration) [1999] 1 SCR 497 (Can). 35 36
276 Tamar Hostovsky Brandes Dorner’s analysis of the constitutional right to equality in Miller rests primarily on two pillars. The first is her perception of the right to human dignity, which she understands as protection against humiliation. Dorner’s position on this issue can thus be classified as adopting the ‘narrow’ mode of constitutional interpretation. The second is the position that in order for an unnamed right, in this case the right to equality, to be protected under the Basic Law, it must be established that violations of the unnamed right amount to violation of the protection against humiliation. Since only some instances of violations of equality may be humiliating, the Basic Law protects only against such instances.41 Dorner’s opinion thus offers a refined version of the question of recognition of unnamed rights, according to which the relevant question is not only which previously recognised rights came within the scope of the Basic Law, but also which aspects of such rights enjoyed constitutional status. In many cases, only certain aspects of unnamed rights will fall within the scope of the law. Dorner’s approach will thus be referred to as the ‘partial protection’ approach, although entire unnamed rights may also be recognised under this approach if each and every violation of such rights would amount to humiliation. The main innovation in Dorner’s approach was the argument that unnamed rights may be only partially protected by the Basic Law. Dorner’s analysis was also different from previous case law, however, in the fact that she clearly indicated her own perception of human dignity and emphasised the relationship between the core of the right to human dignity according to this perception and the unnamed rights it gave rise to. The main difference from previous case law, other than the possibility of partial protection, was thus the extent and specificity of the legal reasoning offered for the recognition of unnamed rights. While not explicitly adopted, Dorner’s opinion in Miller has actually greatly influenced the judicial rhetoric used with regard to recognition of unnamed rights. As indicated above, in early decisions recognizing the right to equality such as the Hupert and Daniloviz decisions the possibility that only some aspects of equality came within the scope of the Basic Law was not discussed and no explanation was offered for why the right to equality came within the scope of the right to human dignity.42 In later cases, a more refined approach was offered. Thus, for example, in the Movement for Quality Government v Knesset decision, discussed below, Justice Barak recognised that the right to human dignity includes only some aspects of the right to equality.43 In Adalah v Minister of Interior, Barak emphasised that not all aspects of equality that would have been included, had it been recognised as an independent right that stands on its own, are included within the framework of human dignity. Only those aspects of equality that are closely and objectively connected to human dignity are included within the framework of the right to human dignity.44
The need to examine which aspects were ‘objectively connected’ to the right to human dignity led judges to deviate from previous practice and state explicitly their view with regard to the content of the right to human dignity. Review of the case law reveals two main approaches with regard to the nature of the right to human dignity. Under the first ibid. Hupert (n 25); Daniloviz (n 25). 43 Movement for Quality Government v Knesset (n 12) para 40. 44 HCJ 7052/03 Adalah v Minister of Interior 61(2) PD 202 [2006] (in Hebrew). 41 42
Human Dignity: Definitions and Parameters 277
approach, the right to human dignity is perceived primarily as protection against humiliation.45 Under the second approach, the right to human dignity is perceived primarily as protecting the autonomy of free will.46 The relationship between the two interpretations is not always clear. In the case of the Academic Center of Law and Business, which concerned the privatization of prisons in Israel, Court President Beinisch determined that a violation of human dignity may also exist as an ‘independent’ violation, when a certain act or the establishment of a certain institution do not violate other human rights, but their existence reflects, from a social perspective, disrespect to the individual and his value as a person.47
Beinisch’s opinion recites the Kantian notion of dignity by determining that transforming the prisoners to a means of financial gain is incompatible with human dignity.48 While the Kantian notion of dignity is often argued to reflect the centrality of individual autonomy, Beinisch’s opinion actually echoes the humiliation approach. The humiliation approach is also reflected in Justice Levi’s minority opinion, which recognises the humiliation embedded in the control of one person over another. Justice Naor, on the other hand, determines that the primary violation in turning the prisoner to ‘a means of gaining profit’ is due to the injury caused to ‘the autonomy of the individual’, and explicitly indicates that the ‘humiliation model’ is not necessary.49 The differences between different judges’ recognition of unnamed rights are thus based on the underlying values they place at the core of the right to human dignity. Such difference is not a difference between narrow and broad models of interpretation, but between value-based world-views. The adoption of the ‘partial recognition’ rhetoric had the potential of developing clearer legal guidelines for the recognition of unnamed rights. In reality, this did not happen, for two reasons. The first is that judges provided no explanation as to why they chose a particular perception of the right to human dignity over another. But more importantly, in many cases, they failed to apply the ‘closely linked’ test and made no serious attempts to establish the necessary linkage to the core values. As a result, in some cases the scope of the rights protected seems overly broad in relation to the core values they purport to protect. In others, they seem too narrow. The sections below will demonstrate these claims. C. Protecting Societal Values as Part of Human Dignity One case in which the recognition of an unnamed right appeared to go overbroad was the case of the Movement for Quality Government v Knesset referred to above.50 The case examined the constitutionality of exemptions from military service granted to 45 See Justice Dorner’s opinion in Miller (n 32) and in CA 4463/94 Golan v Prisons Service 50(4) PD 136, 191 [1996] (in Hebrew). 46 See, eg Justice Barak’s approach in Movement for Quality Government v Knesset (n 12), Hamifkad Haleumi (n 14) para 26 of Justice Naor’s opinion. 47 HCJ 2605/05 Academic Center of Law and Business v Minister of Finance (19 November 2009), Nevo Legal Database (by subscription) para 38 (in Hebrew). 48 ibid para 39. 49 ibid 18. 50 Movement for Quality Government v Knesset (n 12).
278 Tamar Hostovsky Brandes ultra-Orthodox men in Israel. Both Justice Barak, writing the majority opinion, and Justice Cheshin, writing the minority opinion, agreed that the current legislation under which the exemption was granted violated equality. They differed, however, on whether such violation of equality amounts to violation of human dignity. Justice Barak restated his position that at the basis of the right to human dignity lay the autonomy of free will. He then argued that requiring some members of society to risk their lives for the sake of society while exempting others from the same duty violates the serving members’ ‘personal identity’ and amounts to violation of the right to human dignity as expressing the autonomy of free will, freedom of choice and freedom of action. Justice Cheshin argued in response that while the law exempting ultra-Orthodox men indeed violated equality, he failed to see how either the autonomy of free will of those drafted or their freedom of choice was violated simply by the fact that other candidates for military service were granted unjustified exemptions. He also examined the linkage between the violation of equality at stake and the value of protection against humiliation, and determined that in contrast to the Miller case, in which the Army’s refusal to consider the petitioner’s candidacy on the grounds that she was a woman was indeed humiliating, granting exemptions from military service to certain sectors of the population did not humiliate those who did serve. While they may perceive the granting of the exemption as unjust, and, according to Cheshin, rightly so, neither of the values that were viewed as protected by the right to human dignity were violated. Instead of basing his opinion on the violation of human dignity, Cheshin thus based it on the principle of equality as a foundational principle of the State of Israel.51 The Movement for Quality Government v Knesset decision is an example of a case in which, despite using the ‘partial recognition’ model, the Court fails to convincingly establish the ‘close link’ between the notion of human dignity, which is understood as protection of personal autonomy, and the right recognised, which is equality in the duty of military service. D. Recognizing Social and Economic Rights While the protection granted in the Movement for Quality Government v Knesset decision to the right of equality under human dignity seems to be too elaborate, the protection granted to other rights, and in particular social and economic rights, appears to be too narrow. In the Commitment to Peace decision, for example, the petitioners argued that reductions in income supplement benefits provided by the state violated the right to human dignity.52 The decision thus revolved around the meaning of the right to human dignity with regard to the standard of living. Justice Barak, in the majority opinion, determined that, in this regard, ‘the right to live with dignity is the right that a person should be guaranteed a minimum of material means, which will allow him to subsist in the society where he lives’.53 Others have noted that a perception of human dignity that rests on the autonomy of free will is more likely to lead to recognition of civil rights and freedoms than to social 51 The majority opinion actually rejected the petition on the grounds that the Government should be granted additional time for the implementation of the law. 52 Commitment to Peace (n 16). 53 ibid 482.
Human Dignity: Definitions and Parameters 279
rights.54 Yet even under such a perception, no explanation was provided for how and whether the minimal subsistence standard was enough to fulfill the right to autonomy of free will and to guarantee conditions in which it can be realised. The minimal subsistence standard is even more problematic from the view point of the ‘protection against humiliation’ approach. The majority decision contains no discussion and provides no guidelines for what amounts to humiliation in this regard. All parties agree that the right to human dignity protects against degrading living conditions. Yet despite recognition of the fact that ‘a state with the economic strength of a developed nation cannot be compared to a state with a weak economy’,55 the decision contains no guidelines for how the distinction is to be made in practice and no satisfying explanation for why ‘degrading living conditions’ in Israel, taking into consideration the average standard of living in it, should be interpreted only as conditions worse than those needed for minimal subsistence.56 It should be noted in this regard that Justice Dorner, who chaired the original panel, ordered the state to determine a standard required for living in dignity. The determination of such a standard would have likely led to it being challenged by the petitioners and its being examined by the Court. In the process, the core questions raised above would have, at the very least, been discussed. Dorner, however, retired during the hearings and the order was not imposed on the state. The Court instead simply concluded that ‘the human right to dignity is also the right to conduct one’s ordinary life as a human being, without being overcome by economic distress and being reduced to an intolerable poverty’.57 Justice Levy, writing the minority opinion, questioned whether it can indeed be determined that ‘living conditions, which only permit a purposeless subsistence that does not contain any potential for human achievement, do not violate the constitutional human right to dignity’58 and doubted the determination that living conditions, which do not allow even a minimal degree of correlation with the accepted standard of living in society, or which prevent a person having an opportunity, no matter how small, of developing himself, of defining his goals and ambitions and of acting in order to achieve them, do not violate the constitutional right to dignity.59
According to Levy’s first argument, the standard set by the Court is inadequate under the ‘personal autonomy’ approach. According to the second argument, it fails to guarantee both the personal development and protection against humiliation. IV. RETHINKING THE CONCEPT OF DIGNITY
I examined above the different approaches applied by Israeli Supreme Court justices in their interpretation of the right to human dignity and the disputes regarding the meaning Gross and Barak-Erez (n 18). Commitment to Peace (n 16) 480. 56 The notion of humiliation is, at least under some theories, relational, that is, it only has meaning in the context of human interaction (even if such interaction is imagined or theoretical) because it depends on the relations between individuals. Under this theory, the notion of humiliation, as opposed to the notion of auto nomy, has no meaning outside of a society (the feeling of humiliation assumes the existence of a viewer, if only in thought). It could thus be argued that the humiliation-based perception of human dignity is more ‘com munitarian’ than the autonomy-based perception of human dignity. 57 Commitment to Peace (n 16) 481. 58 ibid 496. 59 ibid. 54 55
280 Tamar Hostovsky Brandes of the notion of human dignity. Comparative research indicates that the concept of human dignity is interpreted differently both across jurisdictions and by different judges within the same jurisdiction, thus rendering it difficult to talk about a ‘natural’ meaning of the term human dignity. In examining the use of the concept of human dignity in the adjudication of human rights, Christopher McCrudden identifies the minimum content of the term ‘human dignity’ that all who use the term seem to agree about as follows: ‘This basic minimum core’, explains McCrudden, seems to have at least three elements. The first is that every human being possesses an intrinsic worth, merely by being human. The second is that this intrinsic worth should be recognised and respected by others, and that some forms of treatment by others are inconsistent with, or required by, respect for this intrinsic worth.60
The third element, argues McCrudden, regards the relationship between the individual and the state. The human rights texts, he explains, reflect the principle that ‘recognizing the intrinsic worth of the individual requires that the state should be seen to exist for the sake of the individual human being, and not vice-versa’. While there is general agreement with regard to this minimum core, explains McCrudden, there is little agreement with regard to the nature of the intrinsic worth of the individual, the forms of treatments that are inconsistent with it and the implication of these for the role of the state.61 McCrudden argues that in international human rights documents such as the Universal Declaration of Human Rights, the concept of human dignity is used as a placeholder where a theory of human rights was needed but a consensus with regard to such theory could not be reached.62 In adjudication of human rights issues in different countries, the term human dignity serves as an open channel through which domestic values could be applied. As a result, applying the concept of human dignity often leads to opposite results in similar cases arising in different countries. The main value of the concept of human dignity in the context of international human rights, concludes McCrudden, is in providing a framework for talking about rights. He is pessimistic, however, about the ability of the concept of human dignity to contribute to a common understanding of, or agreement on, the substance of rights. McCrudden’s observations are relevant to our analysis because the concept of human dignity served a similar function in the Basic Law. As no political agreement could be reached with regard to more specific rights, the general term human dignity was used as a way to overcome, or, some may say, mask, disagreements regarding the underlying values of the Basic Law. However, once disputes regarding the rights embedded in the concept of human dignity started reaching the courts, questions regarding these under lying values could no longer be avoided. Comparative review reveals the difficulty in talking about a ‘natural’ meaning of the concept of human dignity, the difficulty of reaching agreement regarding the nature of the ‘intrinsic worth’ of the individual that stands at the centre of the concept of human dignity and the extent to which the concept of human dignity is based on social, reli-
60 C McCrudden, ‘Human Dignity and the Judicial Interpretation of Human Rights’ (2008) 19 European Journal of International Law 655, 679. 61 ibid 679–80. 62 ibid 722–23.
Human Dignity: Definitions and Parameters 281
gious, and personal values.63 In a multicultural society such as Israel, such disagreements can easily become a source of strife. A legal concept of human dignity, however, does not necessarily require adopting one world-view of the nature of the ‘intrinsic worth’ of individuals. An alternative notion of the right to human dignity is as a right that guarantees protection of the various world-views that exist in society regarding the ‘intrinsic worth’ of the individual and the nature of humanity. The right to human dignity, under this approach, is comprised from the universal protections that are needed in order to ensure respect of different social perceptions of the nature of human worth.64 At the core of the proposed model of the right to human dignity is the intrinsic worth of every individual, the essence of his humanity, together with the recognition that there is, and needs to be, no agreement on the nature of such essence. The premise this model is based on is that agreement on such essence is not a necessary condition for identifying rights and measures that are needed in order to protect different beliefs regarding its nature. The model reflects a commitment to mutual respect between various social groups in a multicultural society, and does not require the Court to adopt a single view regarding the essence of human worth, which may undermine its legitimacy. The main role of the Court in interpreting the right to human dignity is identifying those universal protections that ensure respect of various notions of human worth and define the scope and nature of such protections. A. Universal Protections and Particular World-Views I suggested above that both recognition of the importance of the ‘intrinsic worth’, or ‘essence of humanity’ and the recognition that there is more than one view or belief with regard to such worth or essence lie at the core of the right to human dignity. I then suggested that the test for determining whether a particular right falls within the scope of the right to human dignity is whether such right is needed in order to ensure protection of various notions of human worth. This claim is susceptible to at least three challenges. The first challenge regards the protection of world-views according to which some people are worth more than others. The second challenge is that the definition suggested above adds little to the notion of autonomy and therefore is identical to the position that views human dignity primarily as a protection of personal autonomy. The third challenge regards the over-expansion of the concept of human dignity. The answer to the first challenge is relatively straightforward. While the model recognised that there are different world-views regarding the essence of humanity, a key component of it is that whatever such essence is, it is held universally. A belief that some 63 One of the interesting observations that can be drawn from the research is that the concept of human dignity is understood differently even among countries that otherwise share many values in the sense that they are perceived as liberal, Western countries. 64 This concept of human dignity draws on the ‘politics of dignity’ offered by philosopher Charles Taylor. Taylor distinguishes between what he refers to as ‘the politics of universal dignity’ and the ‘politics of difference’. Traditional politics of universal dignity are based on the Kantian notion of universal human potential, an abstract capacity that all humans share. Taylor advocates instead for a ‘politics of difference’, which is based on what humans have made’ of their potential: ‘actually evolved cultures’. Taylor believes that cultural belonging is an important component of personal identity and that since human beings’ self-perception is, at least in part, formed by recognition, we have a duty to respect the cultural aspects of each other’s identity. See C Taylor, Multiculturalism: Examining the Politics of Recognition (Princeton, Princeton University Press, 1994) 41–42.
282 Tamar Hostovsky Brandes human beings are unworthy will thus not be accommodated under this model, although other aspects of cultures from which such beliefs stem may be protected under it, to the extent that they can be applied universally and that the cultures do not revolve exclusively around denying the worth of some people. The answer to the second challenge, which claims that the protection of various world-views is already protected under the notion of human dignity which revolves around autonomy, is somewhat more complex. Indeed, certain notions of autonomy, in particular those encompassing a robust right to privacy, protect the right to hold different world-views on fundamental issues and act upon these views. The legal concept of autonomy, however, in Israel as well as in other countries, emphasises the notion of free choice. As a result, it is limited in its ability to recognise, in a meaningful manner, beliefs and world-views which, according to their holders’ perception, are not based on individual choice. It is important to distinguish, in this regard, between a view according to which the ability to exercise autonomy is the essence of humanity or the nature of the worth of the human being, and a view according to which autonomy is one of the means of protecting different perceptions regarding such essence. Under the first view, the value of autonomy is intrinsic. Under the second, it is instrumental. The belief that autonomy lies at the core of human dignity is one of the various beliefs that are entitled protection, but the primary justification for deriving the right to autonomy from the right to human dignity is the instrumental value of autonomy. The third challenge that may arise with regard to the proposed model is that it expands the notion of human dignity to an extent that, effectively, includes every act and world-view. It is important, in this regard, to stress that the extent of harm a particular act causes is crucial to it constituting a violation of human dignity. Thus, not every act of humiliation or every violation of autonomy will amount to a violation of human dignity, but only those violations that are severe enough to constitute an attack on the ‘essence of humanity’ under different world-views. B. The Protection against Humiliation and Protection of Autonomy under the Proposed Model The two protections that were intuitively identified by judges as deriving from the right to human dignity, namely, the protection against humiliation and the protection of personal autonomy, are arguably indeed universal protections necessary to ensure respect of various world-views regarding the essence of humanity. The experience of humiliation is both a universal experience and a unique human experience. While different acts constitute humiliation in different societies, some characteristics of the concept of humiliation, for example, the notion that it involves the power relations between several parties is common to many societies. The role of the Court in construing the constitutional protection against humiliation involves identifying those acts that are socially perceived as humiliation that is incompatible with respect to human worth, and extending these protections to all members of society. The Court has identified such acts as including discrimination on the basis of race,65 discrimination on the basis of sex,66 violations 65 HCJ 1067/08 Noar Kahalacha v Minister of Education (14 September 2010), Nevo Legal Database (by subscription) (in Hebrew). 66 Miller (n 32).
Human Dignity: Definitions and Parameters 283
of the right of an accused to due process,67 violation of the right to adequate incarnation conditions,68 and abuse of authority in the workplace. According to certain views, paying less than the minimum wage, under certain conditions, may also amount to humiliation, as may broadcasting of pornography.69 Protection of the autonomy of free will may also be necessary in order to protect different perceptions of the human worth. As indicated above, it is important, in this regard, to distinguish between a justification under which autonomy is the essence of humanity or human worth, and a justification for protection of personal autonomy according to which autonomy should be protected because it ensures respect of personal choices made under different world-views regarding human worth. Under the former justification, autonomy has an intrinsic value and is the core of the right to human dignity. This justification cannot accommodate those who believe the value of human life is based, for example, on collective identity, social ties or the sense of one’s belonging to the society within which she lives, all of which are not chosen, at least not entirely. Under the second justification, the value of autonomy is instrumental, and its role is to ensure respect of acts and practices of individuals holding different beliefs, where such acts are perceived by the individuals to be required by their beliefs. In either case, the constitutional protection only extends against violations that touch upon the very core of the human experience. The Court has yet to clearly define the scope of the constitutional protection of autonomy and, in particular, clarify the distinction between constitutionally protected autonomy and non-constitutionally protected autonomy. Another important issue, that has yet to be adequately addressed, is the conditions under which autonomy can be exercised. The Court has identified specific instances in which certain conditions must be met for a person to be able to exercise autonomy. It determined, for example, that when a patient was not provided with all relevant information, consent to medical treatment will not be valid,70 and that certain types of treatment may negate the demand that a suspect’s confession is freely made. A more comprehensive analysis of the conditions for exercising autonomy is, however, still warranted. Other protections may be included in the right to human dignity, in addition to the protection against humiliation and the protection of personal autonomy already recognised by the Court. The process of identifying the protections needed in order to safeguard respect for different world-views and value systems is an ongoing and evolving process. V. CONCLUSION
Human dignity is, in many respects, the central pillar of constitutional rights jurisprudence in Israel. Like many constitutional concepts, the concept of human dignity is an open-ended concept, which may lend itself to various interpretations. The central controversy around the interpretation of the right to human dignity in Israel has revolved around the ability of deriving from the right to human dignity additional, unenumerated CA 5121/98 Yissacharov v Chief Military Prosecutor 61(1) PD 461 [2006] (in Hebrew). HCJ 1163/98 Sadot v Prisons Service 55(4) PD 817 [2001] (in Hebrew). 69 HCJ 5432/03 SIN v Council for Cable and Satellite Broadcasting 58(3) PD 65 [2004] (in Hebrew), Justice Dorner’s opinion. 70 CA 2781/93 Daaka v Carmel Hospital 53(4) PD 526 [1999] (in Hebrew). 67 68
284 Tamar Hostovsky Brandes rights. This chapter argues that closer attention should be given to understanding the notion of human dignity itself. The chapter argues that the notion of human dignity is based on the unique, intrinsic value embedded in human life. The chapter further argues, however, that there is more than one world-view with regard to the nature of such value, and that the right to human dignity should be understood as ensuring respect for such various world-views. The role of the Court in interpreting the right to human dignity includes identifying the types of treatments that are incompatible with respect to different world-views regarding the value of human life. The process of interpretation may include deriving rights from the general right to human dignity, where such rights are needed as safeguards against such treatments. The Court is also entrusted with the task of specifying the acts that amount to prohibited treatment. Thus, for example, if the right to human dignity includes protection against humiliation, clearer guidelines should be offered for what constitutes humiliation and whether the notion of humiliation is objective or subjective, universal or culture dependant.71 If the right to human dignity protects the autonomy of free will, indication must be given for the conditions that are required for a person to be able to exercise such autonomy.72 Similar questions should be examined with regard to other types of treatment determined to be incompatible with a commitment to human dignity. Constitutional interpretation is an ongoing process, and our understanding of con stitutional rights and principles changes over time, reflecting changes in social values. At the same time, in order for constitutional rights to serve as effective safeguards against abuse, their meaning at a given time must be made clear and known. Despite the signific ance of the right to human dignity in Israeli constitutional law, judicial interpretation of the right to human dignity has failed to offer a clear indication of the protections it accords. This chapter argues in favour of a notion of human dignity that accommodates the different value systems and world-views prevalent in Israeli society, and claims that the Court should focus less on the question of unenumerated rights and more on developing such a concept.
71 For scholarly work that addresses these issues see D Statman, ‘Humiliation, Dignity and Self Respect’ in D Kretzmer and E Klein (eds), The Concept of Human Dignity in Human Rights Discourse (Hague, Kluwer Law International, 2002) 209. 72 Jeremy Waldron, eg suggests a notion of dignity that is based on ‘ranking up’ and according all people dignities once granted only to noblemen. See J Waldron, ‘Dignity, Rank and Rights: The 2009 Tanner Lectures at UC Berkeley’ (2009) New York University Public Law and Legal Theory Working Papers, Paper No 151.
19 The Inherent Authority of Judges in a Three-Track Democracy to Recognise Unenumerated Constitutional Rights: The Israeli Story of a Judicial Mission with No Ammunition SHARON WEINTAL
T
HIS CHAPTER IS the second part of my ongoing project,1 namely the quest for an holistic constitutional system and theory – three-track democracy – suited to Israeli society (as a divided one) and instrumental towards universally reconciling two seemingly rival aspirations of free women and men everywhere; the aspiration for evolution and revolution in their collective project. Here, my focus is unenumerated constitutional rights (UCRs)2 as essential judicial ammunition for taking both aspirations seriously. Considering the importance I attribute to holistic theories in constitutional law, it is hardly surprising that I am proposing one for the Israeli Supreme Court in part II. After establishing three-track democracy as an holistic constitutional system and theory in part II, it is only natural that part III focuses in on the function of the UCRs doctrine as a multifunctional bridge to the past, to the present and to the future in a nation’s life cycle. Finally, part IV reviews the Israeli Movement for Quality Government v Knesset case3 using an alternative constitutional source for deriving UCRs – a nation’s founding formula for sustaining political unity.
1 For the first part, see S Weintal, ‘The Challenge of Reconciling Constitutional Eternity Clauses with Popular Sovereignty: Toward Three-Track Democracy in Israel as a Universal Holistic Constitutional System and Theory’ (2011) 44 Israel Law Review 449. 2 The term ‘unenumerated constitutional rights’ (UCRs) is somehow misleading since some of them can be derived directly from a written constitutional concept; as such, they can actually be perceived as enumerated – an implied constitutional provision presumably written between the lines with ‘invisible ink’ (A Barak, ‘Unconstitutional Constitutional Amendment’ (2011) 44 Israel Law Review 321, 337). Therefore, whenever I use the term UCR I am referring to all constitutional rights, which are not explicitly mentioned in the text and cannot be naturally derived from a written ‘mother right’. 3 HCJ 6427/02 Movement for Quality Government v Knesset 61(1) PD 619 [2005] (in Hebrew).
286 Sharon Weintal
I. INTRODUCTION
From an Israeli perspective, the endless intellectual battle between textualists and nontextualists could have been perceived for decades as a privilege for the rich. Starting from scratch in a total textual void, from day one, Israeli judges had to sharpen their skills at spinning gold out of straw.4 This early romantic era ended in 1992 with the enactment of two (then) new Basic Laws: Basic Law: Human Dignity and Liberty and Basic Law: Freedom of Occupation and the so-called ‘Constitutional Revolution’ identified with the United Mizrahi Bank case.5 Nevertheless, the tradition of non-textualism has managed to keep its momentum as the Supreme Court was willing to expand the scope of the new written constitutional concepts in order to host a large variety of seemingly UCRs.6 In order to properly address a claim for an alleged UCR, judges need to consolidate their constitutional theory and be open about it, without which it would be impossible to apply any coherent body of constitutional law. In other words, I believe that the only way constitutional law can evolve to the benefit of society is, in Posner J’s words, by ‘top-down reasoning’ and ‘top-down theories’7 – looking vertically from an holistic constitutional theory8 defining the constitutional courts’ role in a free democratic society down to the specific case or clause at stake – transparent for public scrutiny. It is precisely the fiction of ‘bottom-up reasoning’ (the imagined option of looking horizontally, and eventually vertically from the bottom up, by addressing the ‘plain meaning’ of the text or using ‘reasoning by analogy’ from case to case) that jeopardises the role of constitutional adjudication – be it under a monist constitutional theory, a dualist one, or under three-track democracy. 4 The first era in Israel’s constitutional evolution is typified by a one-level system of written norms (setting aside regulations by the government) where Basic Laws are perceived as regular laws, alongside a judicial willingness to recognize various unwritten rights and liberties. For famous examples: HCJ 1/49 Bejerano v Minister of Police 2 PD 80 [1949] (in Hebrew) (freedom of occupation); HCJ 73/53 Kol Ha’am v Minister of Interior 7 PD 871 [1953] (in Hebrew) (freedom of speech); FH 10/69 Boronovski v Chief Rabbi of Israel 25(1) PD 7 [1971] (in Hebrew) (right to equality); HCJ 355/79 Katalan v Prisons Service 34(3) PD 294 [1980] (in Hebrew) (right to human dignity). 5 CA 6821/93 United Mizrahi Bank v Migdal Cooperative Village 49(4) PD 221 [1995] (in Hebrew). The case is identified with the Israeli historic switch from a one-level system of written norms to a constitutional demo cracy where Basic Laws are perceived as normatively superior. 6 One can argue that these rights are not actually UCRs since they can be derived from ‘human dignity’ and thus should be perceived as enumerated (see n 2 above), however at least with regard to the first case cited here (as discussed at section IVB) the argument seems unconvincing. For famous examples: Movement for Quality Government v Knesset (n 3) paras 30–43 of Court President Barak’s opinion; HCJ 7052/03 Adalah v Minister of Interior 61(2) PD 202 [2006], Court President Barak’s opinion (in Hebrew); CHR 8823/07 John Doe (Anon) v State of Israel (11 February 2010), Nevo Legal Database (by subscription) paras 15–16 of Court Vice President Rivlin’s opinion (in Hebrew). 7 RA Posner, ‘Legal Reasoning From the Top Down and From the Bottom Up: The Question of Unenumerated Constitutional Rights’ in GR Stone, RA Epstein and CR Sunstein (eds), The Bill of Rights in the Modern State (Chicago, University of Chicago Press, 1992) 433, 433–36. Posner seems to agree that bottom-up reasoning merely purports to be reasoning (ibid 449–50). 8 Posner seems to favour clause-by-clause (‘modest’) theories – tracing the purpose of one clause – over holistic (‘immodest’ or ‘ambitious’) top-down theories grappling with the purpose of the constitution as a whole (ibid 439–50). It seems that any ‘modest’ theory (eg a theory of speech or equality) will have to eventually interconnect with a holistic ‘immodest’ one since only the latter – placing speech or equality in the context of the role of constitutional courts – can face constitutionalism’s legitimacy challenges (discussed at section IIA below).
Unenumerated Constitutional Rights in Israel 287
Monist theories perceive normal politics (legislation by the representative in Parliament) as the ultimate expression of popular sovereignty; as such, any intervention in its output suffers from the counter-majoritarian difficulty;9 nonetheless, some still endorse constitutionalism for the sake of reinforcing the system. Process-based monism perceives democracy as a procedural system for making political decisions as an aggregation of private preferences, which unfortunately involves some inherent failures that need to be addressed by constitutional judges;10 conversely, rights-based monism perceives democracy as a substantial system committed to a moral imperative – to treat people with ‘equal concern and respect’ – in which constitutional judges serve as a morality oversight mechanism over the system’s output.11 Either way, a ‘bottom-up reasoning’ – avoiding an holistic top-down theory and obscuring the judgment – fails to provide a normative basis for the counter-majoritarian act of overruling a political decision by the people’s representatives. Where dualism is concerned, normal politics is only a limited expression of popular sovereignty, while the higher lawmaking process – which involves the people directly in unique constitutional moments – is a more authentic and superior one. Accordingly, constitutional courts play the pivotal role of signalling to the people and other branches of government that normal politics seem to trespass into the constitutional sphere and that they better raise their level of political awareness accordingly.12 Obviously, an obscure message would fail to facilitate the badly needed constitutional dialogue. Finally, where three-track democracy is concerned,13 constitutional courts are equally important since they signal to members of a nation that their trustee body appears to be breaching its trusteeship duties.14 Such a breach of trust occurs whenever a nation’s trustee seems to radically divert from its dynamic founding narrative as the ultimate expression of popular sovereignty.15 The court’s judgment carries another important message whereby if members of a nation actually find the course of their trustee body acceptable, perhaps the time is ripe for a third track in order to collectively break out of the founding narrative. Here again, a vague judicial message misses the whole point of having a constitutional court. The appeal for an holistic top-down theory in constitutional law goes far beyond the subject of this chapter, namely as a legal basis for deriving UCRs; it actually covers the full spectrum of constitutionalism. It seems that even where an enumerated right or an 9 AM Bickel, The Least Dangerous Branch: The Supreme Court and the Bar of Politics (Indianapolis, BobbsMerrill, 1962) 16–23; J Waldron, ‘The Core of the Case Against Judicial Review’ (2006) 115 Yale Law Journal 1346, 1348–53. 10 JH Ely, Democracy and Distrust: A Theory of Judicial Review (Cambridge, Harvard University Press, 1980) 87–88, 102–03. Those failures encompass the tyranny of the majority syndrome (the tendency of the ins to favour themselves at the expense of the outs) and the discrete and insular minority syndrome (the vulnerability of certain minorities). For a critical approach of Ely’s theory, see L Tribe, ‘The Puzzling Persistence of Process-Based Constitutional Theories’ (1980) 89 Yale Law Journal 1063, 1067–79. 11 R Dworkin, Freedom’s Law: The Moral Reading of the American Constitution (Cambridge, Harvard University Press, 1996) 7–19. 12 B Ackerman, We the People: Foundations (Cambridge, Harvard University Press, 1991) 6–16, 266–94. 13 For further discussion of three-track democracy, see Weintal (n 1) 470–89; below, at section IIA. 14 For a similar insight on MPs’ status as public trustees, see S Navot, ‘Members of the Knesset as “Public Trustees”’ (2000) 31 Mishpatim 433, 465–86, 518–25 (in Hebrew). 15 A nation’s dynamic founding narrative tells the unfolded story of free women and men joining forces to organize society according to several founding formulas in order to secure their well-being and basic unity. The dynamic nature of the story lies in the fact that those formulas are constantly being evaluated and reshaped in the wake of every founding event – every new chapter written in the nation’s biography – whether the constitutional text changes accordingly or not.
288 Sharon Weintal implied right (derived from an enumerated ‘mother right’) is concerned, the broad language of the Bill of Rights provides virtually no guidance for constitutional courts; therefore, failing to connect with an holistic top-down theory regarding its role in a free democratic society is liable to leave the audience with a bitter taste of legislation from the bench that undermines the legitimacy of constitutional adjudication. II. THREE-TRACK DEMOCRACY AS THE ULTIMATE TOP-DOWN THEORY FOR THE ISRAELI SUPREME COURT
A. Three-Track Democracy in a Nutshell Three-track democracy can be associated with an holistic constitutional system and theory: as a constitutional system, it discriminates between evolutionary and revolutionary decision-making and identifies the basic norm of the legal system with a nation’s collective will (or dynamic founding narrative). As a constitutional theory, it justifies constitutionalism and judicial review mainly for the purpose of regulating a nation’s evolution – protecting its evolved founding formulas for organizing society against harmful normative mutations – until a nation directly embraces a revolutionary leap under a special third track. Constitutional eternity clauses (non-amendable constitutional provisions)16 as well as the unconstitutional constitutional amendment doctrine (the judicial power to strike down amendments to the constitution on substantive grounds)17 seem to play a pivotal role in three-track democracies by blocking the first two tracks in order to force revolutionary movements to engage in the consensual third track and protect society from a unilaterally-imposed revolution – an act liable to be detrimental to the delicate evolutionary process. This can be achieved when a legal system distinguishes between three decision-making tracks: two tracks for evolutionary decision-making (normal legislation and normal constitution-making) both in compliance with a nation’s dynamic founding narrative, and a special third track for revolutionary decision-making in order to break out of the current narrative.18 The idea is to regulate revolutionary decisions, design a special process to ensure that a nation’s timeless achievements are embedded in the new emerging narrative and that the basic interests of individuals invested in the former narrative are taken into consideration. A successful third track – under the supervision of the constitutional court – is based on an historic compromise within an extraordinary ad hoc body of representatives (equal to the original constituent assembly), between the guardians of the current founding narrative and their rival revolutionary political movements, later to be submitted for ratification by the people directly. Thus, a new sustainable political unity will be likely to emerge from the ruins of the dying one.
16 On constitutional eternity clauses and the challenge of integrating them in a democratic system adhering to popular sovereignty, see Weintal (n 1) 456–63, 489–96. 17 Barak, ‘Unconstitutional Constitutional Amendment’ (n 2) 322–38. 18 Compare with Schmitt’s insight of the normally fading out sovereign (or nation) which makes a comeback in rare historical times to form a cycle of full incarnations: C Schmitt, Constitutional Theory (Durham, Duke University Press, 2008) 109–11, 126–28.
Unenumerated Constitutional Rights in Israel 289
Such a scheme seems to be relatively successful when it comes to addressing constitutionalism’s legitimacy problems: first, by placing narrative above text, three-track democracy encourages evolution under the first two tracks, as long as the continuity and inner logic of the narrative remain intact – thereby easing the tension with regard to the precommitment19 and intertemporal precommitment problems20 and giving more leeway to the current generation. Second, free members of a nation need not wait forever for the slow evolutionary process and benefit from the option of breaking out of their narrative in the here and now through the third track. Three-track democracy faces another legitimacy challenge as to what makes constitutional judges any better for the job, especially due to their counter-majoritarian nature.21 One response – the formal argument – would be that a nation gave them the job described in the provisions of the constitution; however, this argument is more convincing in a system that has formal eternity clauses where the three-track scheme can be figured out intuitively from the basic structure of the document. Another response would be that under the basic assumptions of three-track democracy (described below) other branches wield limited governmental powers as a nation’s trustees; it follows that only an unbiased arbitrator can effectively impose these boundaries according to the ultra vires doctrine.22 Finally, the institutional edge argument is that the judiciary seems to be a more trustworthy guardian of a nation’s will; it is precisely these independent, counter-majoritarian and unaccountable aspects that render the court competent to bravely stand in the way of a majority illegitimately promoting a unilaterally-imposed revolution.23 Moreover, it seems that other branches – as serial elected bodies – are much more likely to distort the narrative (as the collective DNA) at the first revolutionary challenge in order to please their revolutionary voters. In contrast, a slight revolutionary wind blowing from a temporary majority is probably not enough to divert the court from its familiar trail; only strong indications coming from a third track can persuade judges that the time has come to format their judicial hard drive, since a new constitutional order has been legitimately applied in society.
19 S Holmes, ‘Precommitment and the Paradox of Democracy’ in J Elster and R Slagstad (eds), Constitutionalism and Democracy (Cambridge, Cambridge University Press, 1988) 195, 199–205. The precommitment issue deals with the tendency of constitutionalism to frustrate majority will. 20 E Parnes, ‘They the People: A Third-Party Beneficiary Approach to Constitutional Interpretation’ (2003) 43 Santa Clara Law Review 495, 500–02. The ‘intertemporal precommitment problem’ deals with the tendency of constitutionalism to frustrate majority will not just for the sake of society’s current fundamental values, but in order to express the will of past generations. 21 For the ‘counter-majoritarian difficulty’, see Bickel (n 9). For the argument whereby the legislature is up for the job of protecting individual rights, see Waldron, ‘The Core’ (n 9) 1371–95. It seems as though at least one of Waldron’s basic assumptions (ibid 1359–69) – a commitment to rights – is under serious doubt regarding Israel. As a divided society – obtaining a large ultra-Orthodox community (placing God above rights) and grappling with a long national conflict (placing minority rights at risk) – giving up on judicial review is not something Israelis should seriously consider. 22 Compare with the classic ultra vires doctrine (W Wade and C Forsyth, Administrative Law, 9th edn (Oxford, Oxford University Press, 2004) 33–37; D Barak-Erez, Administrative Law (Tel-Aviv, Israeli Bar Publishing House, 2010) 97–98, 116–19 (in Hebrew)), which stresses that any governmental act falling outside the boundaries of statutory governmental powers is subjected to the judiciary’s inherent authority of judicial review. 23 For a different approach calling for further accountability of constitutional courts in direct proportion to their judicial review powers, see Y Dotan, ‘Judicial Review and Accountability – A Comparative Analysis’ (2007) 10 Law and Government 489 (in Hebrew).
290 Sharon Weintal B. Basic Assumptions of Three-Track Democracy i. Nation’s Collective Will Serves as the ‘Basic Norm’ of a Legal System I have previously defined and analyzed the relationships between the three legal entities dominating the public sphere of a free society – individual, nation and state – placing them in a pyramidal structure in the same order.24 In such a pyramidal structure, on the one hand, a nation’s collective will serves as the ‘basic norm’ of the legal system, since any other norm-producing entities in society (namely state organs) are lower in the structure and thus subordinate to the upper entity’s will.25 On the other hand, a nation’s free will is somehow limited: a nation cannot legitimately abolish the modern state or its democratic regime, since they both serve as universal preconditions for securing the well-being and the essence of freedom of the individual, which stands above the nation in the structure. It is the individual’s free will that is associated with the origin of free nations; it is its natural rights and well-being that serve as the ultimate test for the morality and legitimacy of any collective project in society. Another implication of the pyramidal structure is that all branches of government which function as institutions of the state – due to its lower position – are actually hybrid entities serving a dual role in society of both an organ of the state and a trustee of the nation. For instance, Parliament serves as an organ of the state – the legislative branch – by introducing sovereign-made rules for the sake of peaceful coexistence between members of society; it further serves as the nation’s trustee26 responsible for guarding its collective will (or dynamic founding narrative) by capturing its essence in written legal norms. Similarly, the judiciary serves as an organ of the state, delivering sovereign-made justice and resolving conflicts which would otherwise be resolved vigilante style; and here again, it serves as the nation’s trustee guarding its constitution as a manifestation of its narrative against an unauthorised attempt to speak on its behalf. ii. Nation’s Collective Will is Not Frozen in Time – the Emergence of a Dynamic Founding Narrative A nation is an evolving entity whose collective will and identity are constantly changing.27 Therefore, rigid written constitutions – regardless of the constitution-making process – are liable to become a pale, sometimes unreliable, still image of a dynamic evolving identity. As the gap between identity and text widens, the basic legitimacy problems of constitutionalism – that is the precommitment problem, the intertemporal precommit Weintal (n 1) 464–66. For the description of the ‘basic norm’, see H Kelsen, General Theory of Law and State (Cambridge, Harvard University Press, 1945) 110–11; H Kelsen, Pure Theory of Law (Berkeley, University of California Press, 1967) 193–95. 26 For the view whereby any authority holding ‘constituted power’ (embedded in the text) acts as trustee of the ‘constituent power’ (embedded in the people’s hands), see UK Preuss, ‘Constitutional Powermaking for the New Polity: Some Deliberations on the Relations Between Constituent Power and the Constitution’ in M Rosenfeld (ed), Constitutionalism, Identity, Difference and Legitimacy: Theoretical Perspective (Durham, Duke University Press, 1994) 143, 155–58. 27 For the dynamic nature of a nation’s identity, perceptions and fundamental values, see HCJ 265/87 Beresford v Minister of Interior 43(4) PD 793 [1987] para 10 of Justice Barak’s opinion (in Hebrew); HCJ 2605/05 Academic Center of Law and Business v Minister of Finance (19 November 2009), Nevo Legal Database (by subscription) paras 18–20 of Justice Levy’s opinion (in Hebrew). 24 25
Unenumerated Constitutional Rights in Israel 291
ment problem and the counter-majoritarian difficulty – intensify. The amending process is a limited tool for bridging the gap since its procedural requirements are sometimes too hard to follow28 and it tends to suffer from the same failures facing normal politics.29 In order to effectively deal with the inherent gap of constitutionalism, one has to rule out the option of counting oneself among the ‘originalist’ camp. Addressing the original will of the founding fathers or their understanding of the nation’s founding formulas for organizing society, frozen in time, means letting the ‘dead hand’ of the past govern the living30 and ignoring the hierarchical structure described above by placing the pale remnants of a dying nation above the free will of living individuals. On the other hand, joining the ‘presentist’ camp is not an attractive option in its own right; adopting an extreme version of the ‘living constitution’ approach, is actually giving up on the whole idea of constitutionalism as we know it,31 letting a current accidental majority ignore tremendous collective achievements of past generations and carelessly cutting off centuries of delicate evolutionary process. In contrast, three-track democracy urges the constitutional court to count itself among the ‘traditionalist’ camp. Since it serves as a nation’s trustee entrusted with the role of keeping its integrity – as a multi-generational political unity – intact, it should address the constitution differently. Judges ought to accept that a written constitution is merely an imperfect symbol of a supreme founding narrative that holds the cumulative wisdom of all generations regarding a nation’s ideal formulas for organizing society. The founding narrative forms a chain of continuity which binds the founding fathers and the current generation into a single living, evolving nation.32 iii. Dichotomy between Universal and Particular Founding Formulas and Values Three-track democracy assumes that there is a clear dichotomy between universal founding formulas (and values)33 and particular founding formulas. The former represent timeless formulas for organizing society – identified with the last phase in any nation’s evolution – thus, must be kept off the table for all decision-making tracks. Such formulas encompass the modern state and democracy which act to ensure individuals’ well-being and basic freedom. A fine example is the Via Dolorosa of the amending process in the US, see US Constitution, Art V. The common super-majority requirement as part of an amending process suffers from the same failures as normal politics (see n 10 above) although it tends to reduce some of the risks due to the extra power of minorities. 30 For the ‘dead hand’ terminology, see MW McConnell, ‘Textualism and Democratic Legitimacy: Textualism and the Dead Hand of the Past’ (1998) 66 George Washington Law Review 1127, 1127. For the natural right of living generations to govern themselves, see T Jefferson, ‘Letter to James Madison’ (6 September 1789) in JP Boyd (ed), The Papers of Thomas Jefferson, vol 15 (Princeton, Princeton University Press, 1958) 392, 392–97. 31 For the ‘living constitution’ approach as deception since a non-binding constitution implies giving up on constitutionalism altogether, see McConnell (n 30) 1128–29. 32 For the adoption of the founding narrative metaphor in constitutional interpretation, see HCJ 466/07 Gal-On v Attorney General (11 January 2012), Nevo Legal Database (by subscription) paras 9–18 of Justice Levy’s opinion (in Hebrew). For a similar view addressing ‘a nation’s aspirations throughout the course of history’, see Movement for Quality Government v Knesset (n 3) para 24 of Court President Barak’s opinion. 33 A nation’s founding values are derived directly from the dynamic founding narrative as their protection is essential to preserving the narrative’s continuity and inner logic. For the adoption of the ‘founding values’ terminology, see Academic Center of Law and Business (n 27) para 20 of Justice Levy’s opinion. For the ‘fundamental values’ terminology, see United Mizrahi Bank (n 5) para 47 of Court President Barak’s opinion. For the ‘depth values’ terminology, see Movement for Quality Government v Knesset (n 3) para 10 of Court Vice President Cheshin’s opinion. 28 29
292 Sharon Weintal Conversely, the latter represent imperfect formulas identified with a premature, or at least questionable, phase in a nation’s evolution adopted in order to sustain political unity. For instance, the Jewish state formula alienates Palestinian citizens of Israel and is clearly far from perfect; the French civil republican model, denies the collective rights of minorities and thus generates all sorts of identity problems; and the Turkish devotion to Atatürk’s legacy appears to be the act of a zealous minority imposing modernism on a traditional Muslim majority. Thus, founding formulas in this category must remain open for discussion, at least for the political players who engage in the third track.34 iv. Parliament is Incompetent at Running the Third Track I have addressed this assumption at length elsewhere – namely the incompetence of Parliament at speaking on behalf of a nation where revolutionary decisions are concerned (cutting off the continuity of a nation’s dynamic founding narrative or inflicting harmful normative mutations onto its delicate founding formulas), as if it were an original constituent assembly. Moreover, I have ruled out the option of super-majority in Parliament to serve as a nation’s organ, and even raised some serious doubts about a national referendum as the ultimate expression of a third track, especially in divided societies.35 My conclusion was that Parliament can only serve as a nation’s trustee bound to its dynamic founding narrative – a derivative constituent assembly – whereas a legitimate revolutionary leap could only take place beyond its boundaries;36 therefore, if it purports to wield the powers of an original constituent assembly – speaking directly on behalf of a nation in revolutionary decisions – then its action exceeds its limited powers and courts hold the inherent authority to declare it null and void.37 C. Indications for Counting Israel among Three-Track Democracies i. The Inclusion of an Implied Eternity Clause – Section 7A of Basic Law: The Knesset Constitutional eternity clauses and the unconstitutional constitutional amendment doctrine – due to their important function in the system mentioned above – serve as the ultimate indication, the earmark, of any three-track democracy;38 however, only those 34 For extra elaboration on the dichotomy and its pivotal place in a three-track democracy, see Weintal (n 1) 473, 492–95. Compare with Grey’s insight on values that are perceived as ‘permanent and universal features of human social arrangements’ and conversely ever-changing values ‘relative to our particular civilization’ (TC Grey, ‘Do We Have an Unwritten Constitution’ (1975) 27 Stanford Law Review 703, 709). 35 Weintal (n 1) 486–89. 36 Compare with Schmitt’s assertion that both a super-majority in the Reichstag and a majority in the British Parliament are similarly incompetent at eliminating ‘the constitution’ (as the fundamental decision of a political unity). Only a separate constitution-making assembly can claim such a right (Schmitt (n 18) 79–80). 37 For the terminology of ‘original constituent power’ (or assembly) and ‘derivative constituent power’, see C Klein, ‘After Bank Hamizrahi Case – The Constituent Power as Seen by the Supreme Court’ (1997) 28 Mishpatim 341, 355–58 (in Hebrew); AL Bendor, ‘The Legal Status of Basic Laws’ in A Barak and C Berenson (eds), Berenson Book, vol 2 (Tel-Aviv, Nevo, 2000) 119, 132–36, 142–45 (in Hebrew). 38 Compare with Germany’s three-track democracy scheme. On the one hand, the eternity clause (Grundgesetz für die Bundesrepublik Deutschland (Grundgesetz) (GG) (Basic Law), 23 May 1949, BGBl I, Art 79(3) (Ger)) blocks the first two tracks and protects the integrity of the Basic Law against a revolutionary amendment. On the other hand, the Basic Law recognizes a third track for breaking out of its current basic structure by freely expressing the German people’s new collective will (ibid 146).
Unenumerated Constitutional Rights in Israel 293
protecting particular founding formulas (or values) can help us figure out the three-track scheme. Where universal founding formulas (or values) are concerned (representing the last phase in any nation’s evolution), eternity clauses can be interpreted as an absolute collective commitment not to drift away from the path towards progress and thus the recognition of a third track to accommodate a revolutionary aspiration seems unnecessary and even troubling. Conversely, where particular founding formulas (or values) are concerned (representing a premature, or at least questionable, phase in a nation’s evolution), recognizing a third decision-making track – in order to facilitate the option to rectify the imperfections embedded in the collective project through a revolutionary leap – seems to be the only way to reconcile eternity clauses and popular sovereignty.39 Although Israel’s Basic Laws do not include formal (explicit) eternity clauses, its legal system tends to limit the scope of amending power where the Zionist project is concerned – thereby freezing the imperfect particular founding formula of the Jewish state and seemingly desperate for the holistic three-track scheme. Israel’s Basic Laws actually embody an implied eternity clause – section 7A of Basic Law: The Knesset – which authorises the Elections Committee to ban a list of candidates (or only one) on the grounds of having political goals or acting in a manner that negates the existence of Israel as a Jewish and democratic state, inciting to racism, or supporting an armed struggle against the state by an enemy state or terrorist organization. Such a clause sets the barrier against a revolutionary amendment, not at the end of the road in order to strike it down, but rather at the gates of the constituent assembly, thereby neutralizing its potential initiators. Section 7A sends a clear message throughout the Israeli legal system whereby any revolutionary aspiration (regarding the particular founding formula of the Jewish state) should find refuge outside the gates of Parliament since the Knesset (Israeli Parliament) perceives itself – in correlation with the premises of three-track democracy – as the nation’s trustee (rather than organ) loyal to its founding narrative. As a derivative constituent power, it is unauthorised to speak for the nation where revolutionary decisions are concerned; thus, it holds inherently limited amending power.40 ii. The Emergence of the Doctrine of Unwritten Fundamental Principles Considering the various judicial eternity clauses41 stemming from the bench, especially the doctrine of unwritten fundamental principles, the Israeli Supreme Court appears to be very much on board with the Knesset’s self-perception of its inherently limited amending powers.42 An initial indication was Court President Agranat’s opinion in the famous Yardor decision.43 Although the ruling cannot be considered a typical judicial eternity clause For a full discussion, see Weintal (n 1) 470–75, 489–96. For a similar view: Barak, ‘Unconstitutional Constitutional Amendment’ (n 2) 338–41; Bendor (n 37) 135–36; Navot (n 14) 518–25. For another approach whereby the Knesset wields the powers of an original constituent assembly: Klein (n 37) 355–57. 41 A judicial eternity clause can be defined as any judicial doctrine that tends to limit the scope of amending power on substantive grounds as if the constitution embodied an explicit eternity clause. Weintal (n 1) 460–63. 42 See Barak’s clear statement whereby the Knesset was never authorized by the people to eliminate Israel’s fundamental principles as a Jewish and democratic state (Movement for Quality Government v Knesset (n 3) paras 73–74). 43 EA 1/65 Yardor v Chairman of the Central Elections Committee to the Sixth Knesset 19(3) PD 365, 384– 89 [1965] (in Hebrew). 39 40
294 Sharon Weintal because it did not address a revolutionary amendment directly, it still has an indirect effect on the amending power. Since the Court reaffirmed the Elections Committee’s44 unwritten power (20 years before the enactment of section 7A) to ban a list of candidates from national elections, and since the Knesset is also the constituent assembly,45 closing its gates to revolutionary political movements generates an indirect limitation to the amending power. Agranat based his judgment on the interpretation of the Basic Law: The Knesset whereby since the Knesset is ‘the representative assembly of the state’,46 potential MPs cannot turn against the deepest aspirations of the nation, which include its Jewish identity – an observation which is perfectly in tune with the basic assumptions of three-track democracy. However, the most influential judicial eternity clause is the doctrine of unwritten fundamental principles, which perceives such principles as an integral part of a legal system above any other written norms, and thus resembles the Indian Supreme Court’s basic structure doctrine.47 Although it has yet to be applied,48 since it was first introduced by Court President Barak in the Laor Movement case49 it has gained more and more momentum as some of the Supreme Court judges seem to be convinced.50 Court President Beinisch recently reaffirmed its pivotal role in the Israeli legal system and was prepared to test an amendment to the Basic Laws postulating a unique two-year budget scheme in light of Israel’s higher unwritten principles as part of an Israeli unconstitutional constitutional amendment doctrine.51 Recognizing the fact that certain unwritten fundamental principles are an integral part of the legal system, above and beyond the Basic Laws, is like protecting them in a formal eternity clause, similarly to Norway’s general eternity clause which addresses the spirit of the Constitution.52 iii. The Origin and Nature of the Basic Laws Its seems as though the Israeli constitutional project of the Basic Laws fits a three-track democracy like a glove. The Basic Laws enacted by the Knesset, which are entirely declarative in nature, introduce a reliable copy of the already well-established dynamic founding narrative; as such, they could be perceived as a nation’s Trustee Act applying The Elections Committee in Israel serves as a regulative authority at times of elections. United Mizrahi Bank (n 5) para 32 of former Court President Shamgar’s opinion, paras 10–17, 38–39 of Court President Barak’s opinion. For a dissenting opinion limiting the Knesset’s powers to regular legislative ones, see ibid paras 9–29, 33–34, 63–64 of Justice Cheshin’s opinion. 46 Basic Law: The Knesset, s 1. 47 For the basic structure doctrine, see M Abrahams, ‘Judicial Role in Constitutional Amendment in India: The Basic Structure Doctrine’ in M Andenas (ed), The Creation and Amendment of Constitution Norm (London, British Institute of International and Comparative Law, 2000) 195, 201–04. The doctrine follows Schmitt’s footsteps (Schmitt (n 18) 74) by stressing that the concept of ‘amendment’ has inherently limited scope and cannot be used to undermine the basic structure of the constitution. 48 Only one dissenting judgment suggested that the time to apply the doctrine had come, see n 88 below and the accompanying text. 49 HCJ 142/89 Laor Movement v Speaker of the Knesset 44(3) PD 529 [1990] paras 25–30 of Court President Barak’s opinion (in Hebrew). 50 For a few examples among others: CA 733/95 Arpal Aluminum Ltd v Klil Industries Ltd 51(3) PD 577 [1997] paras 31–32 of Justice Cheshin’s opinion (in Hebrew); Academic Center of Law and Business (n 27) para 3 of Justice Hayut’s opinion. For a more sceptical approach: ibid paras 18–20 of Justice Levy’s opinion. 51 HCJ 4908/10 Bar-on v Knesset (7 April 2010), Nevo Legal Database (by subscription) paras 31–35 of Court President Beinisch’s opinion (in Hebrew). For the option of applying an Israeli unconstitutional constitutional amendment doctrine, see Barak, ‘Unconstitutional Constitutional Amendment’ (n 2) 338–41. 52 Kongeriget Norges Grundlov (Constitution) 17 May 1814, Art 112. 44 45
Unenumerated Constitutional Rights in Israel 295
its inherent authority to capture the essence of a nation’s collective project in written constitutional norms as a safeguard against unauthorised acts by actual members of the trustee legislative body. Considering the fact that Israel’s Basic Laws, at least until the so-called ‘Constitutional Revolution’ identified with the 1995 United Mizrahi Bank case,53 were never perceived as pivotal to the nation’s building process (a red alert of sorts that calls for higher levels of political awareness and involvement in constitution-making)54 – be it by Knesset Members,55 the courts56 or the public57 – they come across as a decisive indication of the limited nature of the Israeli constitutional project and of the Knesset’s limited authority as a constituent assembly. Since the enactment of the Basic Laws appears to be an act of a devoted trustee, the Israeli public could afford the luxury to fade out from the constitution-making process just like any other nation under a three-track democracy naturally does when nothing seems to interfere with its normal evolution or with the continuity of its dynamic founding narrative.58 III. THE UNENUMERATED CONSTITUTIONAL RIGHTS DOCTRINE AS AN ESSENTIAL ELEMENT OF THREE-TRACK DEMOCRACY
A. Bridge to the Past The UCRs doctrine serves as an essential multifunctional bridge in three-track demo cracy. It is first a bridge to the past (addressing lacuna in a thin constitutional text) since the written constitution, even if it is a masterpiece, is ultimately an imperfect social product – that is, a desperate effort to capture the elusive essence of a nation’s identity with limited linguistic tools designed for ordinary human communication. This gap is extremely wide in Israel since – from day one – the whole Basic Laws project was meant to be a partial, step-by-step process. Instead of exercising its constituent power, the First Knesset – as an elected constituent assembly – reached the so-called ‘Harari Decision’59 in order to enact one Basic Law at a time to eventually be codified into a constitution. This gradual process is ongoing and no one can claim that the project is over and done with or that the time for final codification has arrived. Thus, the current list of Basic Laws fails to capture the full capacity of the Jewish and democratic state. For instance, the Law of Return60 – which fulfills the deepest aspiration of Zionism in its revival of the original political unity of the Jewish people – was shamefully enacted as See n 5 above and accompanying text. Compare to Ackerman’s levels of collective awareness (Ackerman (n 12)). 55 Note that some of the Basic Laws were enacted with very poor MP attendance – in certain cases, with barely over a quarter of the Knesset Members voting for the Bill. 56 For years under a one-level system of written norms, the Israeli Supreme Court suggested that the Basic Laws were no different than regular laws (HCJ 148/73 Kaniel v Minister of Justice 27(1) PD 794, 796 [1973] (in Hebrew); HCJ 107/73 Negev Automobile Service Ltd v State of Israel 28(1) PD 640, 642 [1974] (in Hebrew); HCJ 60/77 Ressler v Chairman of the Elections Commission 31(2) PD 556, 560 [1977] (in Hebrew)). 57 The enactment of Basic Law: Human Dignity and Liberty and Basic Law: Freedom of Occupation – which in hindsight triggered the ‘Constitutional Revolution’ (see n 6 above) – went completely under the Israeli public radar. 58 Compare with Schmitt’s insight on the normally fading out sovereign (Schmitt (n 18)). 59 DK 5 (1950) 1743. 60 Law of Return, 5710-1950 (hereinafter: Law of Return). 53 54
296 Sharon Weintal a regular, rather than a Basic Law. This is obviously a problematic gap between text and narrative to be addressed by constitutional courts. There are several judicial techniques for bridging such a gap; however the UCRs doctrine is probably the more elegant among them. One technique consists of upgrading the Law of Return, which appears to be a regular law, to the status of a Basic Law by applying a court-made ‘rule of recognition’ as implied by Court President Shamgar.61 However, this step inevitably entails high levels of uncertainty regarding the content of the Israeli constitutional text. Another technique consists of expanding the scope of section 7A of Basic Law: The Knesset, Israel’s aforementioned implied eternity clause; accordingly, if striving to annul the Law of Return is cause for banning a list of candidates (or one) from participating in national elections,62 then handing the same proposal to the Knesset table after the elections are over, seems impossible.63 Using the UCRs doctrine can serve as an elegant bridge to the past by recognizing the unwritten constitutional right of every Jew to join the safe Jewish haven of Israel. After grappling with the possibility of applying the doctrine of constitutional lacuna in Israel, Aharon Barak concluded that since the Israeli constitutional project is a gradual one, recognizing a constitutional lacuna in Israel is impossible since any textual gap reflects a specific stage in the project and not a lacuna.64 Nonetheless, in a three-track democracy such as Israel, courts serve as a nation’s trustee with the specific social mission of guarding its dynamic founding narrative from an unauthorised act by a state organ. Therefore, even if the constituent authority (a state organ) decides to take things step-by-step, courts – as the nation’s trustee – cannot wait forever and need to step in and recognise a constitutional lacuna whenever the written ammunition falls short of fulfilling its role in society. B. Bridge to the Present The second type of bridge is a bridge to the present. As explained above, a nation is an evolving entity with a constantly changing identity; as such, even a text that was relatively successful at capturing its original elusive essence eventually becomes outdated and irrelevant – desperately chasing a dynamic founding narrative. The written constitution either overprotects rights and values that can no longer be derived from the founding narrative – as in the classic case of the right to bear arms in the Second Amendment to the US Constitution65 – or underprotects rights or values that have become essential to keeping the inner logic and continuity of the narrative intact. A fine example is gay rights. The rights of gay people (or the GLBT community) are often lacking in the constitutional text since the whole human phenomenon was ‘in the closet’ at the time of its foundation. As society evolves and a nation’s identity evolves, the UCRs doctrine becomes an essential bridge to the present, thereby preventing intense United Mizrahi Bank (n 5) paras 37–38 of former Court President Shamgar’s opinion. EA 11280/02 Central Elections Committee to the Sixteenth Knesset v Tibi 57(4) PD 1 [2003] paras 11–13 of Court President Barak’s opinion (in Hebrew). 63 Compare to the Knesset Regulations, s 75(e). This section authorizes the Knesset’s presidency to bury a private legislative proposal if it happens to negate the existence of a Jewish state or incites to racism. 64 A Barak, Proportionality: Constitutional Rights and Their Limitation (Cambridge, Cambridge University Press, 2012) 45–69. 65 US Constitution, amend II. 61 62
Unenumerated Constitutional Rights in Israel 297
collective dissonance between the free democratic element in the founding narrative and the status of gay people in society.66 Collective dissonance can be defined as a situation in which a nation radically deviates from its founding narrative, without any clear intention of breaking out of it. Thus, resolving the constitutional crisis through a corrective normative act – resuming the good old narrative – should not be considered as constitutionally problematic. The UCRs doctrine can provide the legal basis for striking down the deviating norm or practice as a bridge to the present; thereby, helping to avoid a collective dissonance in the first place. C. Bridge to the Future The last badly needed type of bridge is a bridge to the future. Three-track democracy is a system in which a nation speaks directly during rare historical times through the operation of the third track. In such a scheme, one of the court’s pivotal roles is to clear the channels for a third track since if it is stuck, there is a serious democratic deficit in the system as free women and men find themselves unable to freely express their will to reorganise as a polity according to a completely new particular formula. It appears as though a large variety of UCRs are closely linked to this judicial mission – for example freedom of speech, freedom of association, right to human dignity, right to equality, etc. Protecting these basic rights is vital to facilitating an open and fair intergroup dialogue – which is desperately needed to reach a broad historic compromise identified with a successful third track – between the founding narrative zealots and their revolutionary opposition. IV. THE ART OF DERIVING RIGHTS FROM A NATION’S FOUNDING NARRATIVE – A NATION’S FOUNDING FORMULA FOR SUSTAINING POLITICAL UNITY AS THE HOLY GRAIL OF CONSTITUTIONAL LAW
A. Israeli Dilemma: Expanding the Scope of ‘Human Dignity’ or Expanding the Search beyond the Text The Israeli constitutional project is clearly an unfinished business as the Basic Laws left huge black holes – constitutional lacunas – in the written Bill of Rights. Basic Law: Human Dignity and Liberty and Basic Law: Freedom of Occupation omit some of the most basic rights and liberties in any free democracy, for example the right to equality, freedom of speech, freedom of association, freedom of religion, freedom of movement, right to due process, etc. 66 See Justice Dorner’s opinion in HCJ 721/94 EL-AL Israel Airlines v Danilovich 48(5) 749 [1994] paras 1–9 (in Hebrew). Recently, Justice Amit, followed Dorner’s footsteps by basing his judgment on social and legal changes in order to recognize sexual orientation as a ‘suspect classification’ subject to strict scrutiny (APA 343/09 The Jerusalem Open House for Pride and Tolerance v Municipality of Jerusalem (14 September 2010), Nevo Legal Database (by subscription) paras 53–57 (in Hebrew)). Obviously, in constitutions including ‘equality’ or ‘human dignity’ explicitly, the UCRs doctrine could be reserved for other challenges as gay rights can be naturally derived from the ‘mother right’.
298 Sharon Weintal While the Israeli ‘Constitutional Revolution’67 successfully established itself by virtually turning doubt into consensus, the Supreme Court faced another painful dilemma: either stop for a while and let the dust settle or proceed to the next stage immediately by expanding the scope of ‘human dignity’, thereby risking the same consensus it worked so hard to achieve. Three models of interpretation were put forward over the years: the broad model – advocated by Justice Mazza – whose notion of ‘human dignity’ encompasses almost any aspect of equality and individual autonomy;68 the narrow model – advocated by Justice Dorner – proposed that the original clear intentions of the Knesset (to exclude several rights and liberties) be taken into account,69 thereby only including aspects of unenumerated rights which – should they be violated – lead to humiliation as a clear antithesis to human dignity;70 and Court President Barak’s intermediate model, which seems to have won the hearts of his fellow judges by suggesting a middle path. Barak’s intermediate model urges judges to expand the scope of ‘human dignity’ beyond protection from humiliation, but rejects the idea of a ‘single-provision Bill of Rights’ accommodating virtually each and every constitutional right. According to Barak, Israeli ‘human dignity’ can only accommodate rights when there is a ‘close connection’ between the alleged right and the hosting concept;71 in other words, it only covers violations of central aspects of equality and individual autonomy, which are tantamount to a breach of the state’s duty to treat the individual as an autonomous complete entity (physically and spiritually), or as an end in itself with an intrinsic value.72 Note that the ‘Basic Principles Clause’ in the Basic Laws seems to corroborate Barak’s winning model by explicitly addressing its inner elements of autonomy and intrinsic value.73 It is somewhat ironic that the prevailing model of interpretation – which has become so pivotal in Israel’s constitutional law – was shaped in the Movement for Quality Government v Knesset case,74 the worst possible case for establishing the model as its application there (as discussed below) seems unconvincing.75 The appeal of the intermediate model is obvious: its flexibility, as well as its connection to the concept of individual autonomy, provide the court with powerful constitutional ammunition, potentially hosting various UCRs. Judges have two ways of addressing a case when applying the model: either by using autonomy as their focal point – as they did in the Adalah v Minister of Interior case76 – or by using the Kantian element as they did in the landmark cases of the Academic Center of Law and Business,77 See n 5 above and accompanying text. Justice Mazza’s opinion in HCJ 4541/94 Miller v Minister of Defence 49(4) PD 94 [1995] para 14 (in Hebrew); PPA 4463/94 Golan v Prisons Service 50(4) PD 136 [1996] para 14 (in Hebrew); HCJ 205/94 Nof v Minister of Defence 50(5) PD 449 [1997] paras 8–10 (in Hebrew). For another view supporting the broad model where equality is the UCR, see HCJ 6784/06 Shlitner v Director of Pension Payments (12 January 2011), Nevo Legal Database (by subscription) paras 2–3 of Court Vice President Rivlin’s opinion (in Hebrew). 69 For a similar famous criticism, see H Sommer, ‘The Non-Enumerated Rights: On the Scope of the Constitutional Revolution’ (1997) 28 Mishpatim 257, 303–14 (in Hebrew). 70 Miller (n 68) para 4 of Justice Dorner’s opinion; Golan (n 68) paras 7–9 Justice Dorner’s opinion. 71 Movement for Quality Government v Knesset (n 3) paras 33 and 38 of Court President Barak’s opinion. 72 ibid para 35. 73 Basic Law: Human Dignity and Liberty, s 1; Basic Law: Freedom of Occupation, s 1. 74 n 3. 75 See below, section IVB. 76 Adalah v Minister of Interior (n 6) paras 30–46 of Court President Barak’s opinion, paras 7–15 of Justice Joubran’s opinion, paras 7–10 of Justice Rivlin’s opinion. 77 Academic Center of Law and Business (n 27) paras 36–39 of Court President Beinisch’s opinion, paras 3–5 of Justice Arbel’s opinion, para 1 of Justice Hayut’s opinion, Justice Joubran’s opinion, para 18 of Justice Naor’s opinion. 67 68
Unenumerated Constitutional Rights in Israel 299
Yissacharov78 and John Doe (Anon).79 Effective as it may be, there are still hard cases which cannot be addressed by using the model; hence, recognizing the alleged right as a constitutional right would make it necessary to search beyond the text and anchor the UCR in a nation’s unwritten founding narrative. Extremely hard cases are those that not only require exploring unwritten territory, they actually require a controversial particular founding formula because exploring the more consensual universal founding formula (the modern state and democracy) seems to get the judge nowhere. Here, a nation’s founding formula for sustaining political unity – which serves as a precondition for the stability and viability of the whole constitutional order80 – could actually prove to be the holy grail of the UCRs doctrine supporters or the non-textualist camp. Since any nation is built on some kind of political unity – whether inside a national group such as Israel (the cultural formula), or between all citizens of the state (or confederation) such as France and the United States (the civic formula) – the specific formula sustaining this precious resource serves as a particular founding formula at the basic structure of the constitutional order.81 B. Reviewing the Movement for Quality Government v Knesset Case Using the Nation’s Founding Formula for Sustaining Political Unity as an Alternative Source The Movement for Quality Government v Knesset case is one of those extremely hard cases when upholding a law seems to undermine a nation’s founding narrative, but it is quite difficult to pinpoint the constitutional provision or principle for resolving the case. Although the Supreme Court gave the law some extra time to achieve its goals, its problematic constitutional base went on to guide the Court in the subsequent Ressler v Knesset case where under Court President Beinisch the law was finally declared null and void.82 The law, known as the ‘Tal Law’,83 reaffirmed the exemption of ultra-Orthodox Yeshiva (institution for religious studies) students from mandatory military service, as long as they dedicate their life to religious studies. The Knesset had to resolve the issue following the Court’s ruling in the Rubinstein v Minister of Defence case84 whereby since the number of exemptions had increased dramatically from a couple of hundred to (then) nearly thirty thousand, it became a substantial issue that exceeded the limited powers of the Minister of Defence.85 To give the Knesset due credit, the law did attempt to address the issue of equality by encouraging Yeshiva students to opt for a short milit ary service that would facilitate their integration into the labour market. As noted above, this was the case for which Court President Barak designed the winning intermediate model (defining the scope of ‘human dignity’), which ironically seems 78 CA 5121/98 Yissacharov v Chief Military Prosecutor 61(1) PD 461 [2006] paras 65–68 of Court President Beinisch’s opinion (in Hebrew). 79 John Doe (Anon) (n 6) para 16 of Court Vice President Rivlin’s opinion. 80 Compare with Schmitt’s observation (Schmitt (n 18) 126–27) of ‘political unity’ as a precondition for constitution-making and for the legitimacy of ‘the constitution’. 81 For the civic perception, EJ Sieyès, Political Writing (Indianapolis, Hackett Publishing, 2003) 93, 97. For the cultural perception, Preuss (n 26) 150. 82 HCJ 6298/07 Ressler v Knesset (21 February 2012), Nevo Legal Database (by subscription) (in Hebrew). 83 Deferment of Military Draft for Yeshiva Students Whose Occupation is the Study of Torah Law, 5762-2002. 84 HCJ 3267/97 Rubinstein v Minister of Defence 52(2) PD 481 (1998) (in Hebrew). 85 ibid paras 17–41 of Court President Barak’s opinion.
300 Sharon Weintal to fall short of resolving the case. It appears as though both elements of the intermediate model cannot apply in this case: first, as Court Vice President Cheshin insisted,86 the exemption has nothing to do with individual autonomy since what seems to be closely tied with service women’s and men’s autonomy is the military service itself and not the exemption of others; second, from a Kantian perspective, their draft serves the purpose of sustaining a modern state – a universal formula for securing their natural rights as an end in itself. Moreover, addressing ‘human dignity’ in its plain meaning adds to the uneasiness since Israeli society glorifies those who serve in the IDF (Israel Defence Forces) and its elites welcomes them with open arms. Giving up on the constitutional text does not mean giving up on the case altogether since in three-track democracy courts have the option to search beyond the text. Court Vice President Cheshin tried to do precisely that by using the doctrine of unwritten fundamental principles at his disposal and exposing two founding values which justify striking the law down: the particular founding value of the IDF derived from the Jewish state formula and the universal founding value of equality derived from democracy;87 however, both should have led him to a dead end. Where the former is concerned, the detection could not have been more precise but something is profoundly wrong with the final conclusion; one can hardly say that as the number of those benefitting from the exemption has grown dramatically, the IDF – by far the strongest army in the Middle East – is under any threat. Where the latter is concerned, Cheshin’s conclusion is still standing on shaky ground. Democracy has many faces, however none of them can support the conclusion that service women and men have the constitutional right to force the state to draft members of a minority group who practice a unique way of life that would be threatened if the alleged right were protected. Facilitating democracy Ely style (resolving process-based failures that the majority could ride on)88 should have led judges to set aside decisions which are sensitive to the preference of a minority at the expense of the majority, since these are actually a sign of a healthy, functioning democracy.89 Switching to a Dworkin style democracy (as a moral commitment to treat people with ‘equal concern and respect’ that could be deserted by a careless majority)90 is equally unhelpful since assessing the majority conduct in this case could lead to the conclusion that nothing here is profoundly wrong; the decision respects the ultra-Orthodox way of life and perception of the ‘good’ rooted in their religious belief, thereby treating members of the minority ‘as equals’ (rather than ‘equally’) as it should according to Dworkin. Whenever a nation’s universal founding formulas for organizing society lead to a dead end, judges in a three-track democracy may turn to its particular formula for sustaining political unity as a last resort. Expanding our search, brings us to the Zionist project which aspires to integrate the universal achievements of enlightenment and secularism (by introducing the modern state and democracy) into a particular founding formula 86 Movement for Quality Government v Knesset (n 3) paras 68–75 of Court Vice President Cheshin’s opinion. 87 ibid paras 6, 9–11, 16–18, 27–30, 47–52. 88 See n 10 above and the accompanying text. 89 For this argument and the assimilation of Ely’s ideas in Israel, see Movement for Quality Government v Knesset (n 3) paras 7 and 9 of Justice Grunis’s opinion. For the rejection of Ely’s ideas as an excessively narrow top-down theory, see ibid paras 75–81 of Court President Barak’s opinion. 90 See n 11 above and the accompanying text.
Unenumerated Constitutional Rights in Israel 301
of a Jewish state – a safe Jewish haven in the historic homeland, a public sphere which supports Jewish culture and connects to its historical roots, all for the sake of reviving and sustaining the historical political unity of the Jewish people. It seems as though a political decision which treats the ultra-Orthodox community as an indigenous national minority – in its full capacity – undermines the basic assumptions embedded in the founding narrative and could lead to a disintegration of the political unity at the core of the constitutional order; therefore, it could be perceived as exceeding the Knesset’s limited powers as the nation’s trustee. When applying the UCRs doctrine, judges should ask themselves two distinct questions: first, could this unenumerated right soon prove vital to fulfilling my mission as a constitutional judge according to the theory I adopted? I believe that under three-track democracy, the answer should be affirmative since a right to equality in barring the burden of sustaining a Jewish state is already essential to fulfilling the Supreme Court’s (as the nation’s trustee) mission to keep the founding narrative intact. Second, is this an appropriate case for establishing such a right, even if I am not going to apply it this time? This is where judges should look at the bigger picture and since the number of ultraOrthodox escaping the draft has grown dramatically, and the issue has become a major obstacle undermining political unity, the time was definitely ripe to recognise such a right, even if the Court ultimately decides not to strike down the law. Obviously, recognizing such a right is only the first stage of judicial review, to be followed by the application of a Limitation Clause as part of the balancing stage. One can argue that since we deal with an ambiguous constitutional sphere beyond the constitutional text, the use of the Limitation Clause applied should be narrower than the textual one,91 thereby leaving Parliament more room for discretion. In any case, it seems as though the Supreme Court ruled correctly that since the law is a minor step on the long road to equality and since it is a provisional arrangement soon to be reviewed by the Knesset, the time is not ripe for striking down the law. Within a few years, everything had changed; the Supreme Court realised that the law had failed miserably at achieving its goals and that the number of exemptions hit a new record – hence the time had come to declare the ‘Tal Law’ null and void.92 Two final remarks with regard to the alternative way of addressing the case discussed here: first, anchoring the right to equality in the nation’s particular founding formula to sustain political unity inside the Jewish people – rather than in its universal founding formula of democracy – limits the UCR scope to the Jewish community and cannot serve as a constitutional basis for striking down a similar exemption benefitting the Palestinian citizens of Israel. Second, the Jewish State’s formula for sustaining political unity inside the Jewish people can serve as a legal basis for deriving all sorts of UCRs essential to keeping the founding narrative intact, inter alia: the right of Jews worldwide to become part of the Jewish State, the right of all streams of Judaism (Orthodox Judaism, Conservative Judaism, Reform Judaism, Progressive Judaism or secular Jews) to be treated with ‘equal concern and respect’, the right of secular Jews to freedom from religious coercion and the right of ultra-Orthodox Jews to communal autonomy to the extent that it can coexist with the basic political unity.
Basic Law: Human Dignity and Liberty, s 8; Basic Law: Freedom of Occupation, s 4. Ressler v Knesset (n 82) paras 49–62 of Court President Beinisch’s opinion.
91 92
302 Sharon Weintal V. CONCLUSION
Judicial restraint, which includes restraint in applying the UCRs doctrine, could be seen as a virtue by many constitutional theories (including three-track democracy) – a reminder of the limits of judicial power according to a specific theory. The problem with some textualists is their tendency to make it an intrinsic virtue, a main pillar of their textualist faith; however, it is an empty virtue if it fails to interconnect with constitutional judges’ roles in a free democratic society. In three-track democracy, constitutional judges play a pivotal role in guarding a nation’s deepest collective will (dynamic founding narrative) against subversive normative acts by unauthorised lower entities – a nation’s trustee such as the Knesset (Israeli Parliament) – through the wrong decisionmaking track. Judicial restraint whenever a nation’s founding formulas (or values) seem to be under attack – letting the Knesset go wild – could mark the beginning of a rapid disengagement process from the core of the narrative, deepening the collective dis sonance in a nation’s daily practice and ending up with a total collapse of the whole constitutional order when no one could argue that the Israeli people has collectively spoken for the second time. The UCRs doctrine is an essential element in every three-track democracy for bridging the gap between the role of constitutional judges in society and the textual ammunition at their disposal. Removing this arsenal from the constitutional courts through a textualist leap back could have severe repercussions for the legal system: first, by undermining the written constitution and turning it into an uninteresting object for legal archeologists, frozen in time and hardly relevant; second, by undermining the Court’s integrity as the nation’s trustee, which systematically fails to guard the nation’s founding values.93 Recognizing a UCR inherently invites criticism; judges could be blamed for rewriting laws that Parliament was elected to write – legislating from the bench – when all we ever wanted from them was to uphold the written constitution as is. Rather than hide what they really do or obscure their judgments, the courts should be educating the public regarding their social mission. Therefore, applying the UCRs doctrine should naturally begin from stage one of the reasoning – introducing an holistic top-down theory for the role of constitutional judges in a free democratic society; only then can they progress to stage two – deriving UCRs as essential ammunition to allow them to succeed at their mission. I believe that three-track democracy, as presented here, is a powerful holistic topdown theory for stage one which perfectly fits the Israeli constitutional evolution. Moreover, it could serve as a compelling direct message from the Supreme Court to members of the Israeli nation that goes over the heads of some politicians in the Knesset who unfortunately try to drive a wedge between them.
93 Compare with Grey’s assessment (Grey (n 34) 710–14) whereby applying a strict textualist approach (‘the pure interpretive model’) would mean giving up almost every important constitutional doctrine ever introduced by the US Supreme Court.
20 Perpetual Constitutional Moments: A Reply to Hostovsky Brandes and Weintal DAVID FONTANA
I
WAS HONOURED to be a part of the fascinating conference and I will keep my reply brief and informal, as was suggested by the conference format.1 I also do not pretend to be an expert on Israeli constitutional law, so I offer my perspective as an outsider and someone whose work focuses substantially on comparative constitutional politics more generally.2 I see in the three wonderful chapters (originally papers) I am commenting on some themes which relate to elements of comparative constitutional development more generally, and so that will be the focus of my brief reply. I will sketch out some general concepts about constitutional development and tie them to these chapters. One of the key questions that has to be asked in understanding each constitutional system is ‘how many constitutional moments have there been in that system’? I use ‘constitutional moments’ in the way that Bruce Ackerman does, as describing a state of heightened interest and participation in debating ‘the permanent interests of the community’.3 There might be three different categories of answer to this question: none, somewhere between one and constant moments, and then constant moments. Let us call these three ideal types the non-revolutionary, fixed revolutionary, and perpetual revolutionary cases. What strikes me in reading these papers and learning more about the Israeli constitutional system is that it seems to be one of the few (if there are any other) countries which fall into this third ideal type, and that has significant implications for understanding the Israeli constitutional system and these chapters. I. THE PATH DEPENDENT NATURE OF CONSTITUTIONAL MOMENTS
Not all moments are created equal when it comes to interest and participation in discussing and formulating the fundamental rules of a constitutional democracy. Bruce 1 If citations or more information about any of my arguments is desired, please do not hesitate to contact me. 2 Many of my arguments here are related to a larger paper I am writing, entitled D Fontana, ‘Revolutionary and Reorganizational Constitutionalism: How it Matters for Constitutional Development’ (forthcoming). For a preview of some of these arguments, see D Fontana, ‘Comparative Originalism’ (2010) 88 Texas Law Review 189. 3 B Ackerman, We the People: Foundations (Cambridge, Harvard University Press, 1991) 240, 272–74.
304 David Fontana Ackerman captured this in distinguishing between ‘normal politics’ and ‘constitutional politics’. During the former period, one significant feature Ackerman highlights is disinterest in and apathy towards fundamental understandings of the rules of the constitutional system. During the latter period, one significant feature Ackerman highlights is interest and participation in defining fundamental understandings of the rules of the constitutional system. Constitutional moments can be critical junctures during the process of constitutional development, creating a path dependence logic summarised by Margaret Levi: ‘once a country or region has started down a track, the costs of reversal are very high. There will be other choice points, but the entrenchments of certain institutional arrangements obstruct an easy reversal of the initial choice’.4 Constitutional moments can create this path-dependent dynamic precisely because they are constitutional moments. Constitutional moments are sticky, and from that come much of their significance. There are substantial learning effects, meaning that ‘Knowledge gained . . . leads to higher returns from continuing use’.5 This happens on the part of the ruled and the rulers of a system where there are constitutional moments. For the ruled, during the constitutional moment citizens are paying attention to and engaging in discussions about constitutional issues. They are spending time learning and investing in particular understandings of the Constitution, and learning and investing in particular individuals as being central to those understandings of the Constitution. In essence, they are making ‘asset-specific investments’6 which have their greatest value so long as those fundamental features and individuals of the constitutional system remain the same. For the rulers, as well, there is a path-dependent nature to constitutional moments. They also learn a series of norms and follow a series of leaders to create their specific understanding of the constitutional system. For rulers and ruled, constitutional moments have large set-up costs, making it hard to change constitutional understandings through normal politics because understandings are so heavily created during constitutional politics. Normal politics are different and do not create the same sorts of path-dependent dynamics. During normal politics, because interest in and participation in debating fundamental constitutional issues is less, more practical settlements can be reached. Ideologically pure groups can become less ideologically pure, and so can modify understandings to accommodate new situations. This transpires because the pure groups might become less pure because they are less invested in all issues, and because however ideologically modified they become they can do so in part because their constituencies are not paying attention to their activities. The result is less of a series of asset-specific investments, and a series of more malleable and adaptable series of constitutional understandings accommodated to the situation. 4 M Levi, ‘A Model, a Method, and a Map: Rational Choice in Comparative and Historical Analysis’ in MI Lichbach and AS Zuckerman (eds), Comparative Politics: Rationality, Culture and Structure (Cambridge, Cambridge University Press, 1997) 28. 5 P Pierson, Politics in Time: History, Institutions and Social Analysis (Princeton, Princeton University Press, 2004) 24. 6 See PA Gourevitch, ‘The Governance Problem in International Relations’ in DA Lake and R Powell (eds), Strategic Choice and International Relations (Princeton, Princeton University Press, 1999) 144–45 (‘Political actors develop investments, “specific assets”, in a particular arrangement – relationships, expectations, privileges, knowledge of procedures, all tied to the institution at work . . . [This] helps to explain institutional persistence. actors . . . have incentives to protect their investment by opposing change’).
A Reply to Hostovsky Brandes and Weintal 305
II. HOW THE N MATTERS FOR CONSTITUTIONAL MOMENTS
How many constitutional moments does a country have, and how does it matter? We can create three different ideal types along this continuum – systems where there have been no constitutional moments, systems where there have been some but no perpetual moments, and systems where there have been perpetual moments. For the first ideal type (‘non-revolutionary’), the constitutional system is a combination of malleable and adaptable constitutional understandings, with no enormous returns to investment and so lower costs to change. In the second ideal type (‘fixed revolutionary’), there are a series of fixed understandings which guide the operation of practical, normal constitutional politics during other times. In the third ideal type (‘perpetual revolutionary’), the fixed understandings are constantly being revisited because of the persistence of constitutional moments. Without normal politics to settle and make practical these constitutional moments, the system always features foundational debates without much opportunity to make these debates more operationalised. In the non-revolutionary category, we might think of the modern experience in the United Kingdom. There have been major constitutional changes in the United Kingdom in the past 40 years. At first there was the European Communities Act of 1972, and then the Human Rights Act of 1998. Both laws created major changes to the British system, creating forms of constitutional review (in form and in practice because of the political popularity of the courts). In neither situation, though, were these legal changes key reasons for voting for the leaders supporting these changes. In neither situation were these changes the result of major public discussions of and participation in making these reforms. A non-revolutionary system might present its own, unique series of complications. I have mentioned before how this might push a system towards non-originalism. It might make the constitutional law of that system more practical and adaptable since it was created during moments when compromises were possible and perhaps necessary and desirable. This might also make these principles adapted during these zero-moment creation times more fragile, because there is not the halo effect of a creation moment surrounding them. Without higher asset-specific investments, the system can change quickly and dramatically. In the second category of fixed revolutionary moments, we might place perhaps most, if not almost all, constitutional systems. In the United States, originalists might argue that we are a one (1789) or a two (Civil War) or a three (New Deal) moment system, but the important point is that these systems have had greater than zero moments. These systems have heroic figures and principles, supporting originalism (of less relevance here), but also the heroic first principles created during these moments are codified and implemented during normal politics. The ideological purity of the creation moment is tempered and made more practical and durable by the disinterest and practicality of the normal political moment. In the third category of constant moments might lie Israel (and perhaps India and a few other places), a system with something approaching perpetual constitutional moments. Most of the available research by social scientists seems to suggest a constitutional politics different from other constitutional democracies. Discussions of politics and elections feature debates about and potential changes to core issues of the polity in
306 David Fontana a way not seen in non-revolutionary or fixed revolutionary polities. The forms of politics which transpire during normal politics seem to be more absent in Israeli politics. There might be less ability to compromise on fundamental principles because elites and voters always care about first principles. Even if elites cared about fundamental principles more or less at different times, they are accountable to a citizenry which perpetually cares about those issues. Compromise is difficult because constitutional politics are perpetual. Other systems have constitutional moments which last more than a moment – they might last years or decades. Joanne Freeman has helpfully described the earlier years of the American Constitution as precisely this: The framers of the Constitution may have been looking forward to the detached indifference of ‘normal politics’ with a sense of foreboding . . . but the ongoing constitutional moment of the 1790s – when the rules of the game were yet to be fleshed out – was no picnic. Uncertainty ruled the day – uncertainty about the structure of the Constitution and the durability of the Union, the impact and implications of popular politics, the new republic’s place on the international stage, and the larger significance of a national two-party war. All of these controversies had constitutional implications, provoking ongoing anxieties about the need for constitutional change. In essence, during the 1790s, there was no ‘normal politics’ – no long-term disengagement from the national political process among populace and politicians alike.7
The constitutional moment persisted past the creation of the Constitution, to the early years of the United States of America. But even Freeman recognises that this period ended, even if it lasted longer than some might realise. We might say the same things about other countries (the period in South Africa in the 1990s leading to the creation of a permanent Constitution, for instance). These moments might be longer than immediate moments, but they still function like a play, with a beginning, middle, and an end. Likewise, there is a discussion about whether the creation of the Basic Laws in 1992 represented a ‘Constitutional Revolution’.8 But implicit in these discussions as well is the sense that constitutional development in Israel has an ending, or periods where there is far less interest in basic constitutional principles. III. AN APPLICATION TO THE CHAPTERS
This understanding of the nature of constitutional moments is what can help us understand the Hostovsky Brandes9 and Weintal10 chapters. I do not know the specifics of these chapters, but it struck me that the reality of perpetual constitutional moments is behind much of what they are discussing. As regards Professor Weintal’s fascinating chapter, it is hard to think of a ‘three-track democracy’ existing anything more than in normative theory in Israel given the political realities. Weintal describes some of these tracks in ways which might always be true, but seem less true in Israel. For instance, he
7 JB Freeman, ‘The Election of 1800: A Study in the Logic of Political Change’ (1999) 108 Yale Law Journal 1959, 1967–68. 8 I assume this debate is familiar to participants, so I will forego the many citations which could be supplied here. 9 Ch 18 in this volume. 10 Ch 19 in this volume.
A Reply to Hostovsky Brandes and Weintal 307
argues that the parliamentary process fails in involving ‘the people directly’.11 This might always be true when systems are not direct democracies in any meaningful way, but less true in Israel because of the intense interest in basic principles – perhaps people are only concerned about Parliament when it deals with important issues. These same features seem to be even truer for Professor Hostovsky Brandes’s chapter. It is true that every system has debates about interpretation, but many of the features of the debate highlighted in Professor Hostovsky Brandes’s chapter seem to result from a system of perpetual constitutional moments. It is part of the reason why ‘no clear test exists today for the recognition of unnamed rights’.12
ibid text accompanying n 12. Hostovsky Brandes (n 9) part I.
11 12
21 In Search of the Right to Health in Israeli Constitutional Law AEYAL GROSS*
I. INTRODUCTION: ISRAELI VERSUS ISRAELI AND OTHER TALES OF INEQUALITY
I
N 2006, AMIR Israeli petitioned the Israeli Supreme Court (sitting as the High Court of Justice) contesting the decision of the Council for Cable and Satellite Broadcasting which allowed the broadcasting of the World Cup soccer games within the framework of the ‘pay-per-view’ system. He argued that the high fee demanded for watching the games caused severe damage to him and to sports fans in Israel.1 The Supreme Court rejected the petition, finding that as a result of the Council’s instructions the cable and satellite companies had been required to revise their proposed pay-per-view programme and that the revised package as presented to the Court included the broadcasting of the major games (including the final), in the public commercial channels as well as a free daily magazine which covered each day’s games. Additionally, against payment of 492 NIS (New Israeli Shekel) (about $100) viewers were able to purchase a package which would allow them to view another 50 games. It was also submitted to the Supreme Court that the package’s price was later reduced and was liable to be reduced even further. The Supreme Court thus found that the revised plan complied with the Council’s policy of guaranteeing free viewing of events of public importance, and its decision to allow the broadcasting of only the less important games for a fee, was therefore within the boundaries of its discretion. Furthermore, given the further reduction in the price of the package, the Supreme Court held that it was no longer possible to claim that severe damage was caused to the petitioner. Less than two weeks after giving its judgment in the Israeli (World Cup) decision, the Supreme Court gave a judgment in the case of Victoria (Vicky) Israeli.2 Ms Israeli was in the process of losing her hearing, but this was a process that could be reversed were she to undergo a cochlear implant operation. This operation was included * I am grateful to Sharon Bassan for her research work and to Michael Prawer for his editing work. Thanks also to the Cegla Center for Interdisciplinary Research of the Law in Tel-Aviv University for its support of the research conducted for this chapter. 1 HCJ 699/06 Israeli v Ministry of Communication (30 May 2006), Nevo Legal Database (by subscription) (in Hebrew). 2 HCJ 2974/06 Israeli v Committee for the Expansion of the Health Basket (11 June 2006), Nevo Legal Database (by subscription) (in Hebrew).
312 Aeyal Gross in the Health Services Basket (HSB) as anchored in the Israeli National Health Insurance Law (NHIL)3 but entitlement to the operation under the law was conditional upon a self-contribution of 70 per cent, which, in this case, amounted to 70,000 NIS (about $15,500). This sum was far in excess of the economic capacity of the petitioner, a teacher and single mother, and she based her petition not only on the right to health but also on the right to equality, the latter having been recognised as part of the right to human dignity under Israel’s Basic Law: Human Liberty and Dignity (hereinafter: the Basic Law). Although the Court expressed concern as to how a 70 per cent self-contribution could be regarded a ‘contribution’, it refused to intervene, arguing that the question of deciding on the services to be included in the HSB and under what conditions, as well as the establishing of priorities, rests first and foremost with the Committee authorised to recommend the scope of the HSB and to establish priorities, and ultimately with the Government. The Supreme Court held that the need to set priorities is unavoidable given the rapid technological and medical changes on the one hand, and the high costs of technologies and medications on the other. While expressing its hope that the issue would be revisited by the HSB Committee in the future, the Supreme Court did not intervene in the matter but rather wished the petitioner a full recovery, and expressed its hope that she would find comfort in the fact that she had raised the problem and given it a voice. The latter hope was fulfilled: the litigation was part of a process that drew attention to the problem of the high self-contribution requirement for cochlear implant operations – a requirement that was cancelled by the Government two months later.4 These two Israeli judgments, Israeli (World Cup) and Israeli (Cochlear Implant) tell a more general story about Israeli constitutional law. In both cases a complex process of litigation and advocacy produced changes that lowered the access threshold to public resources. But whereas in the Israeli (World Cup) case the petition was rejected by the Supreme Court only after it was determined that access to the core of the resource – in that case the major games – would be free and that payment for access to further games would not be excessive, in the Israeli (Cochlear Implant) case the Supreme Court rejected the petition before the governmental intervention. This meant the retention of a situation in which access to an essential medical operation was denied by reason of the applicant’s inability to pay a prohibitively high fee. While the Israeli (World Cup) case was not framed as a constitutional case, but as an administrative law case concerning the discretion of the regulator, two other cases which were discussed from a constitutional perspective illustrate the neglect of judicial consideration of access to health care in Israeli constitutional law when compared to other issues of access to public resources, and the still existing gap between the constitutional protection of social rights on the one hand, and civil and political rights on the other hand. In the first of these cases, heard six months after the decision in the Israeli (Cochlear Implant) case, the Supreme Court held in Majority Camp5 that conditioning a demonstration permit on the organisers’ financing of the expenses of the demonstration’s security is National Health Insurance Law, 5754-1994. Government Decision 406, taken on 27 August 2006: www.pmo.gov.il/Secretary/GovDecisions/2006/Pages/ des406.aspx (in Hebrew). 5 HCJ 2557/05 Majority Camp v Israel Police 62(1) PD 200 [2006] (in Hebrew). 3 4
The Right to Health 313
illegal as it violates the right to freedom of speech. The Supreme Court held that given the importance of the right to freedom of speech and demonstration and the role of the state and the police in protecting this right and the possibility of exercising it, the police do not have the authority to condition a licence to demonstrate on the demonstrators’ financing of the security for the demonstration. In this case the Supreme Court emphasised that the state has a twofold duty to protect the freedom of speech and demonstration. First, the state is duty bound not to violate a person’s right to freedom of speech and demonstration by imposing a prohibition on the person’s ability to exercise the right, a right anchored in section 2 of the Basic Law which prohibits the violation of a person’s life, bodily integrity or dignity. Second, the state has a positive duty, within reason and taking into consideration the means available to it and according to its own scale of priorities, to allocate the required resources needed in order to enable the realization of the right to freedom of speech and demonstration, the latter being grounded in section 4 of the Basic Law, which determines that every person has the right to a protection of his or her life, bodily integrity and dignity.6 According to the Supreme Court it is inconceivable that the police should be able to impose a financial burden on those seeking to exercise their right to freedom of speech and demonstration, and thereby place a ‘price tag’ on the exercise of a right, for this would imply a violation of the rights for those who cannot afford it.7 The rule is that when considering the allocation of resources the police may condition the demonstrations upon compliance with requirements pertaining to time, place and manner, but the Supreme Court held that saving resources is not in and of itself legitimate grounds for a police refusal to provide security for a demonstration: the protection of human rights costs money and a society that respects human rights must be willing to bear the burden.8 As in the Israeli (Cochlear Implant) case, Majority Camp likewise involved a governmental fee imposed on people wishing to exercise a right. The difference between them, however, was that in the context of the right to freedom of speech, such a fee was considered as illegally violating this right, with the Court emphasizing that doing so would condition the right on the ability to pay for it and deny access to those without means, whereas the Court refused to make a parallel determination in the context of the right to health. Contrasting the two cases demonstrates the discrepancy between the respective rulings and the judicially authored hierarchy of civil rights and social rights which is both artificial and ideological, given that both categories of rights may require public funding to be exercised equally. This however is the practice of the Supreme Court: placing a ‘price tag’ on the exercise of freedom of speech was considered a prohibited violation of the rights of those who cannot afford it – a consideration that Ms Israeli did not merit when it came to the price tag put upon the cochlear implant she needed. While the Majority Camp decision was framed in terms of freedom of speech, it could alternatively have been considered as raising the issues of equality: in previous case law in a different context the Supreme Court held that requiring high payments that deny accessibility to important public services may constitute a violation of the right to equality. In Hupert v Yad Vashem9 the Supreme Court addressed the legality of a project of ibid para 14. ibid para 16. 8 ibid para 17. 9 HCJ 5394/92 Hupert v Yad Vashem 48(3) PD 353 [1994] (in Hebrew). 6 7
314 Aeyal Gross Yad Vashem – the national Holocaust Memorial in Israel – enabling each person, against payment, to place a memorial stone carrying the name of a person who died in the Holocaust. The petitioner argued that this amounted to discrimination between relatives of Holocaust victims who could afford the payment and those who could not. The Supreme Court determined that there is a duty, also anchored in the Basic Law (interpreted as including the right to equality), to treat all people equally in this context. As such, a requirement of payment even if levied equally upon all, may constitute discrimination if its result is that people lacking in means are either unable or barely able to afford it and if the benefit or service for which the payments are required are essential or important to an extent that precludes their being conditional upon the individual’s ability to pay.10 The Supreme Court emphasised that its tendency to define the payment requirement as prohibited discrimination would increase in direct proportion to the increase in the rate of the payment and the importance of the service or the benefit.11 In the Hupert case the Supreme Court chose not to intervene, given that payments amounted to a few hundred dollars, and did not impose a high or unreasonable burden, and insofar as Yad Vashem events memorialised all Holocaust victims and the service in question concerned an additional memorial. While the Majority Camp decision rested on freedom of speech and the Hupert decision rested on the right to equality, they both determined that imposing burdensome payments on access to basic essential services or basic rights may be illegal. The differences between the cases may lie in the fact that the issue discussed in Majority Camp – freedom of speech – was considered as being within the scope of constitutional rights guaranteed by the Basic Law, whereas the issue discussed in Hupert – the creation of memorials for Holocaust victims – was not considered as a basic right per se. Hence, the need in the latter case to base the ruling on the right to equality, and the lack of an absolute prohibition on payment in this case, which compelled the formulation of the tests of: (a) rate of payment and the burden it imposes; and (b) the essentiality of the service. In this regard the Hupert case resembles the Israeli (World Cup) case, even if the latter did not include a constitutional analysis. However, all three cases are distinct from the Israeli (Cochlear Implant) case, which, based on either of their rationales, could and should have been decided differently. Under the rationale of Israeli (World Cup) and Hupert the issue of the cochlear implant should have been considered as concerning access to an essential medical operation within the public health system, thus constituting an essential service, and the excessively high payment requested for that essential service should have been viewed as giving rise to unjustified inequality (‘price tag’) based on the ability to pay. Under the Majority Camp rationale it could – and should – have been decided that the right to health proscribes the imposition of payments that impede access to it, especially when it involves essential medical treatments that are part of the right to life and bodily integrity explicitly recognised in the Basic Law, and that the state is obligated to guarantee its exercise without putting a price tag on it that makes it accessible only to those who can afford it. It bears note that while freedom of demonstration, memorials for Holocaust victims, and the viewing of soccer games were all viewed as rights on which a price tag (or at least an excessive price tag) cannot be imposed, health was treated by the Supreme Court ibid para 13. ibid.
10 11
The Right to Health 315
as a commodity on which a price tag can be imposed, turning it into something available only to those who can afford it.12 In order to understand the gap between Israeli (Cochlear Implant) and the three other decisions, it is necessary to understand the framework under which the right to health is relegated to second class status in Israeli constitutional law, being viewed as part of the perceived ‘distinct’ category of ‘social and economic rights’. Part II that follows takes up this issue. Part III offers a short description of the Israeli health system and its legal framework, as an introduction to part IV, which revisits the issue of the right to health within Israeli constitutional law, focusing on litigation. Finally, part V looks at basic dilemmas posed by the concept of health rights, and addresses the role that the concept of human rights can play within public health systems such as the Israeli one. While the right to health is not limited to issues of access to health care, this chapter will focus exclusively on that component. II. THE BARE MINIMUM: SOCIAL RIGHTS IN ISRAELI CONSTITUTIONAL LAW
The discrepancy between the Israeli (Cochlear Implant) case and the other cases discussed in the previous part, points to what I argue is an ideological schism between so-called ‘civil and political rights’ and so-called ‘social and economic rights’.13 As the cases discussed clearly illustrated, the Israeli Supreme Court does not recoil from imposing positive duties upon the Government when it comes to issues which it views as touching on civil rights, and from restricting the imposition of barriers on public resources and services, even where there is no recognised ‘right’ to those resources and services, when such barriers are created through the imposition of payments (‘price tag’). This schism is reflected in the prevailing interpretation given to the Basic Law, one of the two Basic Laws on human rights legislated in 1992.14 Much of the debate about ‘social and economic rights’ focuses on the interpretation of the right to human dignity guaranteed under the Basic Law, based on the proposition that this is an ‘umbrella’ right from which numerous, more specific rights may derive. The debate over the more specific rights that derive from the right and specifically, whether these include social and economic rights, continues to rage.15 The dominant position is that in the context of the social right to health the right to human dignity means nothing more than the right to a minimal material existence, including ‘access to rudimentary medical services’,16 and it can only be asserted in cases of a material 12 See for this distinction, A Gross, ‘Health in Israel: Between A Right and a Commodity’ in Y Rabin and Y Shani (eds), Economic, Social and Cultural Rights in Israel (Tel Aviv, Ramot Publishing, 2004) 437, 459 (in Hebrew). 13 For more on this ideological divide, see D Barak-Erez and A Gross, ‘Do We Need Social Rights? Questions in the Era of Globalization, Privatization, and the Diminished Welfare State’ in D Barak-Erez and A Gross (eds), Exploring Social Rights: Theory and Practice (Oxford, Hart Publishing, 2007) 1, 7. 14 On the Basic Laws and their background, see: A Gross, ‘The Politics of Rights in Israeli Constitutional Law’ (1998) 3 Israel Studies 80, 81–83; D Barak-Erez, ‘From an Unwritten to a Written Constitution: The Israeli Challenge in American Perspective’ (1995) 26 Columbia Human Rights Law Review 309, 312–13. 15 A Gross and D Barak-Erez, ‘Social Citizenship: The Neglected Aspect of Israeli Constitutional Law’ in D Barak-Erez and A Gross (eds), Exploring Social Rights: Theory and Practice (Oxford, Hart Publishing, 2007) 243, 260. 16 LCA 4905/98 Gamzu v Yeshayahu 55(3) PD 360 [2001] para 20 (in Hebrew); HCJ 366/03 Commitment to Peace and Social Justice Association v Minister of Finance 60(3) PD 464 [2005] para 15 (in Hebrew).
316 Aeyal Gross deficiency that prevents a person from subsisting.17 This minimal approach to social rights casts a shadow over the possibility of judicial review of legislation in the name of the right to health. At the same time, the Israeli legal system has a longstanding tradition of recognizing non-written rights in a way that at the very least allows judicial review of administrative actions even if it balks at interfering in primary legislation.18 Moreover, the Basic Law does include explicit references to the rights to life and bodily integrity, both of which can directly impact on issues of access to healthcare.19 In the recent Hassan decision,20 for the first time, the Supreme Court struck down a statute based on its determination that it violated the right to exist in human dignity. The Supreme Court held that the statutory determination that a person who owns or uses a car will be denied income support by the state violates the right to exist in human dignity. Writing for the Court in this case Court President Beinisch accepted the argument that there is no distinction between civil and political rights on the one hand and social rights on the other hand, pointing to the ‘positive’ and ‘negative’ elements required for the protection of both.21 However, while this decision could signal a breakthrough in the Court’s case law on social rights, the scope of the Court’s recognition of the right was limited to minimal (rather than adequate) conditions of existence in dignity and its implications on further case law have yet to be seen. In the following part I will briefly discuss the Israeli health system and the legislative framework that regulates it, after which I will explore the ways in which the minimalist approach to social rights affects both the Supreme Court decisions on access to health care and those of the lower courts as well. III. THE STRUCTURE AND THE LEGISLATIVE FRAMEWORK OF THE ISRAELI HEALTH SYSTEM
Constitutional discussion of health rights in Israel is conducted against the background of the structure and the legal framework of the health system, now to be briefly described.22 The current Israeli health system is the product of a reform entrenched in the NHIL. Section 1 of the NHIL reads: National health insurance in accordance with this law, will be based upon the principles of justice, equality and mutual assistance. 17 Commitment to Peace (n 16) para 15. For a more detailed discussion of this case see Barak-Erez and Gross, ‘Social Citizenship’ (n 15) 250. One area of law in which the Supreme Court has shown relative willingness to accept petitions was that of the right to education, when petitions invoking social rights required the judicial review of administrative, rather than legislative, actions. For further discussion, see ibid 254. 18 See Barak-Erez and Gross, ‘Social Citizenship’ (n 15) 254. 19 In this context I do not address other possible sources for the right to health in Israeli law that derive from international law, most importantly the International Covenant on Economic, Social and Cultural Rights (adopted 16 December 1966, entered into force 3 January 1976) 993 UNTS 3 (ICESCR). 20 HCJ 10662/04 Hassan v National Insurance Institute (28 February 2012), Nevo Legal Database (by subscription) (in Hebrew). 21 ibid para 28. 22 For reasons of space the description here is succinct and does not analyze specific provisions of the NHIL. For a broader discussion, which includes references to the specific clauses in the relevant legislation as well as a more detailed discussion of some of the issues addressed here, see A Gross, ‘Health Right in Israel Between Solidarity and Neo-Liberalism’ in C Flood and A Gross (eds), The Right to Health at the Public/Private Divide: A Global Comparative Study (Cambridge, Cambridge University Press, forthcoming) ch 3.
The Right to Health 317
Under the NHIL, the system under which Israelis were insured by Sick Funds which also provided most of the health services, was replaced by a universal national health system in which the insurance is national, and a mandatory progressive health tax is paid to the state. The state assumes financial responsibility for the HSB, while most health services continue to be provided by the Sick Funds. The NHIL created a system based on a trilateral relationship, in which the state bears the responsibility for funding of the HSB, the Sick Funds remain responsible for the provision of most health services, and residents are entitled to health services. In theory, this structure distinguishes between the financial relationship between the resident and the state (to which he or she pays a health tax), and the clinical relationship between the resident and the Sick Funds which provide him or her with health services. A financial relationship also exists between the Sick Funds and the state, with the latter being obligated by the NHIL to finance the health basket. This concept is one that, if maintained, would have treated health as a right, as it would create equal access to health services according to need and not according to the ability to pay.23 However, the promises and premises of justice and solidarity as well as universality which were articulated in the NHIL were already undermined to a certain degree by the statute itself, and even more dramatically by a series of legislative amendments introduced shortly after its enactment. These amendments included significant changes in the financing of the health system in a manner that reduced the state’s responsibility and shifted some of the burden to patients in the form of direct out-of-pocket payments.24 Unlike the health tax, these payments are regressive: they are uniform and with the exception of certain exemptions, are levied from all patients regardless of their financial capacity. Apart from certain ceilings and exemptions, user’s fees have actually become a major bar to the equal access of all people to health care.25 Systematic research of a number of groups has shown that a significant part of the population has been forced to waive certain forms of medical treatment or prescription drugs because of their cost.26 More generally, the imposition of these payments reflects a larger shift in the financing of the health system. The shift to a greater role for private finance in Israel is discernible on the macro level: whereas in the 1980s the ratio of public–private financing for health care in Israel was 75:25,27 and in 1996 it was 69:31,28 by 2011 it was 58:42. This contrasts 23 On egalitarian theories of distributive justice as requiring that access to health care services be predicated on need as opposed to ability to pay, see C Flood, International Health Care Reform: A Legal, Economic and Political Analysis (Oxon, Routledge, 2003) 26–28. 24 For background of the law and for a discussion of the reform it created, see S Asiskovitch, Price Tag for Life: The Political Economy of the National Health Insurance Act in Israel (Jerusalem, Magnes Publication, 2011) (in Hebrew) 200–03.The author discusses how the law itself, while promising health coverage to all Israeli residents, transferred much of the decision-making on healthcare to the bureaucracy of the Ministry of Finance, thus making health care dependent on budgetary decisions. This is particularly significant when considering the Ministry of Finance’s role as a leading agent of neo-liberalism in Israel. For an analysis of the legislative reforms which created this change see Gross, ‘Health in Israel’ (n 12) 474. 25 See NHIL, s 8, regarding the legislative framework for co-payments. See also NHIL, s 13(5) and the changes it underwent. 26 For research which points to the effects of co-payments on patients who give up on prescription medicines and other services because of it, see: R Gross, S Brammli-Greenberg and B Rosen, ‘Co-Payments: The Implications for Service Accessibility and Equity’ (2007) 6 Law and Business 197, 213–20 (in Hebrew). See also G Ben-Nun, Y Berlovitz and M Shani, The Health System in Israel, 2nd edn (Tel Aviv, Am Oved Publishing, 2010) 247 (in Hebrew). 27 Ben-Nun, Berlovitz, and Shani (n 26) 125. 28 D Arieli, T Horev and N Keidar (eds), ‘National Health Security Insurance: Statistics 1995–2011’ 51, chart 37, available at: www.old.health.gov.il/Download/pages/stat2011_1995.pdf (2012) (in Hebrew). 2.6% of the figure given for private expenditure originated in the contributions from outside Israel.
318 Aeyal Gross sharply with the average OECD ratio of 72:28 in favour of public expenses.29 This data is augmented by research that has shown that between the years 1995–2003 there was a decrease of 33 per cent in real terms in the budget allocated by the state to the HSB.30 IV. IN SEARCH OF THE RIGHT TO HEALTH IN LITIGATION
The aforementioned provides the background for health rights litigation in Israel. In what follows I will address four main areas of litigation. Due to space constraints my focus will be on cases that concerned issues of financial access to the health care system and its affect on equality among Israeli residents covered by the NHIL. I will not address litigation concerning other issues of no less importance such as discrimination among residents (especially concerning the Bedouin population living in ‘unrecognised villages’) and the perils of populations not covered by the NHIL (migrant workers, asylum seekers and Palestinian residents of the Occupied Territories) whose right to health is often denied.31 A. Financing/Updating and Structural Issues The NHIL contains no fixed mechanism for proper updating of the HSB, having only a limited health cost index. This legislative vacuum generated a series of cases, all of them originating in the fact that the Health Council, a statutory advisory body, originally recommended the establishment of an automatic updating system for the HSB. This update mechanism would reflect the real costs of the Israeli health system and take into account the growth and ageing of the population as well as technological advancement. After the Ministers of Health and Finance failed to adopt this recommendation, the Sick Funds petitioned the Supreme Court, requesting it to order them to act in accordance with the recommendation. The Supreme Court held that the Ministers should consider the recommendation, draw conclusions and then give their reasoned decision.32 When the Ministers ended up rejecting the recommendation, an additional petition was launched against their decision. The Supreme Court rejected the petition, having been persuaded that the Ministers had duly considered the recommendations, and pointing out that the Health Council is only an advisory body. In this judgment the Supreme Court refrained from addressing the issue on its merits, emphasizing that discretion rests with the Ministers.33 It would take another decade for the Supreme Court to change its approach. When the issue again came before the Supreme Court it was stressed that the limited updating mechanism within the NHIL does not include the changing price of 29 OECD, Health at a Glance 2011: OECD Indicators (OECD Publishing, 2011) 157, chart 7.5.1, available at: www.oecd.org/dataoecd/6/28/49105858.pdf. 30 L Achdut, A Shmueli and M Sabag-Andbler, ‘The Financing of the Services Basket in the First Decade of the Operation of the Law – Issues and Tendencies’ in G Ben-Nun and G Ofer (eds), A Decade Since the National Health Insurance Law 1995–2005 (Tel Aviv, The National Institute for Health Services Research and Health Policy, 2006) 219, 226–27 (in Hebrew). 31 For discussion of these populations and their exclusion from or within the Israeli health system, see D Filc, Circles of Exclusion: The Politics of Healthcare in Israel (Ithaca, Cornell University Press, 2009). 32 HCJ 2344/98 Maccabi Healthcare Services v Minister of Finance 54(5) PD 729, 778 [2000] (in Hebrew). 33 HCJ 9163/01 Clalit Health Services v Minister of Health 56(5) PD 521, 528 [2002] (in Hebrew).
The Right to Health 319
hospital days and the changing wages in the public sector. The result was that one decade later, in 2012, the Supreme Court finally held that the Government had acted unreasonably in its protracted failure to address the recognised need for an update of the ‘health cost index’. This index determines the cost of the HSB and by extension, the adjustment of the budget allocated to it. The Supreme Court noted that the erosion of the budget of the Sick Fund effectively emptied the right to health of its content and ordered the Ministers of Health and of Finance to expedite – within six months – the adoption of a mechanism for updating the health cost index.34 This decision is of import ance given its recognition of the connection between the question of the right to health and the health cost index mechanism, and its determination – the practical implications of which are still unknown at the time of writing – that the Government must introduce some changes in this regard. B. Access to Drugs and Services Litigation concerning the scope of the HSB takes place in Israel in two formats. In the first format, insuree’s claims are brought to the Labour Courts against Sick Funds by Israeli residents covered under the NHIL, in cases in which individual patients request a treatment or medicine not provided within the HSB. The second format comprises petitions brought to the Supreme Court, in which petitioners seek judicial review of governmental decisions not to provide certain medication or treatments. As we will see below while some of the first kinds of claims have succeeded, claims of the second type have all failed. i. Labour Courts: Insurers’ Claims against the Sick Funds Numerous petitions are brought before the Israeli Labour Courts, which are the courts with jurisdiction under the NHIL in suits against Sick Funds concerning denial of coverage for prescribed medications and services. The challenges faced by the courts can be classified under a few major headings: the question of the scope of the treatments covered under the HSB (eg how many sessions of physical therapy is a patient entitled to); eligibility for treatments or medications that are included in the HSB but for a different indication than the one for which it was prescribed; medications and services completely excluded from the HSB, but which patients nonetheless request the Sick Funds to provide; medications excluded from coverage but which are required in cases of emergency; and medications and services that are excluded from the HSB but which the Sick Funds in fact provided to certain patients, thus exposing themselves to discrimination based claims.35 The rulings of the Labour Courts on these matters give rise to a number of conclusions.36 Generally, the Labour Courts’ position has been that the health basket is limited and does not purport to include all of the medical services required by a person. The 34 HCJ 8730/03 Maccabi Healthcare Services v Minister of Finance (21 June 2012), Nevo Legal Database (by subscription) para 54 (in Hebrew). 35 For an elaborate discussion of the case law, see Gross, ‘Health in Israel: Between a Right and Commodity’ (n 12) 502–28. For a discussion of litigation concerning health rights in Israel, see also C Shalev and D Chinitz, ‘Joe Public v The General Public: The Role of the Courts in Israeli Health Care Policy’ (2005) 33 Journal of Law, Medicine and Ethics 650, 652. 36 Only a few of these decisions were appealed to the Supreme Court and in even fewer cases did the Court actually issue a substantive decision.
320 Aeyal Gross health services included in the basket, which the Sick Funds must provide, are the basic, essential services. The Sick Funds can, however, offer additional services or medications beyond those prescribed by law, in accordance with their financial resources and in light of their commitment to fulfillment of the goals and principles of the NHIL and their status as a public organ.37 The lower courts have occasionally relied on the lack of recognition of the right to health as a constitutional right to reject claims made by patients to anything beyond the basic basket components.38 On the other hand, in those cases in which courts have ruled in favour of the petitioners, they have relied on four different forms of legal justification. All of them may be regarded as strategies for bypassing the ‘minimum’ approach to the right to health: 1. The duty to provide emergency care unconditionally: based on section 3(b) of the Patient’s Rights Law,39 which determines that ‘in a medical emergency a person is entitled to emergency medical care unconditionally’.40
2. Dynamic interpretation of the health basket: cases holding that the basket must be interpreted in a flexible and dynamic way.41 3. Judicial review of decisions to exclude services from the health basket: in one case the Labour Court reviewed a Government decision not to include a certain device in the health basket. The Labour Court held that the Government’s decision was defective by reason of its failure to take into account the various considerations in favour of including it, in addition to the financial considerations, and that the Government had a duty to occasionally revisit the matter, giving due consideration to matters of human dignity as mandated by the Basic Law and the NHIL.42 4. The duty to exercise discretion in specific cases: the Labour Courts held that Sick Funds can and should exercise discretion for the provision of services beyond those prescribed by law. Later cases which resorted to this form of legal justification focused on the Sick Funds’ ‘Exceptions Committees’. In this context the National Labour Court held that the right to health services is a social human right – a penumbral right of the Basic Law,43 and is a legal social right with constitutional features, based on the inclusion of the right to life and bodily integrity within the Basic Law.44 The National Labour Court rejected the position taken by a Sick Fund, that the ‘Exceptions Committee’ must only consider medical exceptions. It held that the Exceptions Committee should examine ‘hard cases’ and determine whether, given the recognition of the right to health and the right to life, it is possible to provide the patient with the required treatment in cases where there is no solution within the HSB.45 NLC 5-7/97 Madzini v Clalit Health Services 33 PDA 193, 203 [1999] (in Hebrew). DLC 22/99 Isaac v Minister of Health (10 December 1999) (unpublished). 39 Patient’s Rights Law, 5756-1996. 40 DLC 14-339/99 Grundstein v Clalit Health Services (24 March 1999) para 7 (unpublished). 41 LA 1557/04 Clalit Health Services v Kaftsan (29 December 2005) para 5 of Judge Rabinovich’s opinion (unpublished). A Petition to the Supreme Court against this judgment ended in a settlement where the principle of flexible interpretation was accepted, subject to budgetary considerations. See HCJ 3723/06 Clalit Health Services v Ministry of Health (24 July 2006), Nevo Legal Database (by subscription) (in Hebrew). 42 DLC 5360/01 Dekel v Klalit Health Services (1 August 2002) para 106 (unpublished). It was later decided again, by the HSB Committee, not to include the device in the basket of health services. 43 LA 1091/00 Shitrit v Meuchedet Health Services 35 PDA 5, 23 [2000] (in Hebrew). 44 LA 575/09 Maccabi v Dehan (6 January 2011) para 24 (unpublished). 45 ibid para 32. The case is currently under review in the Supreme Court. See HCJ 5438/11 Maccabi v Dehan (pending). 37 38
The Right to Health 321
This case law requires the Committees to examine the data and evidence regarding technologies not included in the HSB both generally, and in regard to specific patients. This may reduce the risk of new technologies being made available exclusively by reason of pressure exerted by interested parties such as pharmaceutical companies. At the same time it ensures that specific circumstances relating to ‘concrete’ patients are taken into account. For example in one case involving a cancer patient who for medical reasons could not undergo chemotherapy, the Sick Fund argued that a drug indicated in the HSB as a second line drug should be provided only if chemotherapy failed, given that the relevant medical indication stated ‘failure in previous treatment’. The Exceptions Committee accepted this position and rejected the patient’s petition. However the Court rejected this position holding that a proven medical situation which prevents the first line treatment amounts to ‘failure’ even if the first line treatment was not tried on the patient. Accordingly the Court ordered the provision of the drug in question.46 The developments described here indicate that whereas in some cases judges regard the health basket and its specific composition as an immutable reality which cannot be tampered with, leading to their rejection of petitioners’ arguments, in other more recent cases they have been willing to engage in rights analysis, be it the right to life and bodily integrity explicitly recognised in the Basic Law, the right to health as a penumbral right, or the right to emergency health care guaranteed under the Patient’s Rights Law.47 Rights analysis, in conjunction with the principle of flexible and dynamic interpretation, has paved the way to judicial intervention in favour of patients in a way that arguably may broaden access to health care. ii. The Supreme Court: Petitions against the Government While the Labour Courts have occasionally shown willingness to intervene in cases involving insuree’s claims against the Sick Funds, the Israeli Supreme Court has not demonstrated the same willingness regarding litigants’ petitions to the Supreme Court requesting direct judicial review of the Government’s decisions concerning the scope of the HSB. In fact it has consistently refused to intervene in matters concerning the scope of the HSB. The most extensive judicial discussion of the scope of the HSB was conducted by the Supreme Court in the Louzon case, decided in 2008.48 The dispute in this case focused on the Erbitux drug, which was prescribed for certain patients (who ultimately petitioned the Supreme Court) but was not rated sufficiently high in the prioritization of the HSB Committee and was hence excluded from the HSB even after an additional budgetary allocation by the Government.49 The petitioners requested that the Supreme Court order the inclusion of the Erbitux drug in the HSB with an indication for colon cancer treatment. They argued, inter alia, that the exclusion of this drug violated the right to health which is part of the right to life and bodily integrity, as well as the right to human dignity, all of which are anchored in LA 45021-05-10 Eliav v Clalit Health Services (12 July 2010) para 26 (unpublished). Gross, ‘Health in Israel: Between a Right and Commodity’ (n 12) 502; Gross, ‘The Right to Health in Israel between Solidarity and Neo-Liberalism’ (n 22). 48 HCJ 3071/05 Louzon v Government of Israel (28 July 2008), Nevo Legal Database (by subscription) (in Hebrew). Similar reasoning can be found in a few Supreme Court cases given before and after Louzon. See HCJ 4004/07 Tronishvili v Ministry of Health (19 July 2007), Pador Legal Database (by subscription) (in Hebrew); HCJ 434/09 Davidov v Minister of Health (3 May 2009), Pador Legal Database (by subscription) (in Hebrew). 49 Louzon (n 48) paras 2–4. 46 47
322 Aeyal Gross the Basic Law. They also argued that the HSB Committee’s decision did not give sufficient weight to the value of human life, and discriminated them vis-à-vis other patients whose required medications were included in the HSB.50 The judgment’s main determinations may be summarised as follows: 1. Constitutional status of the right to health: Regarding the constitutional argument, the Supreme Court determined that the scope of the right to health is difficult to define as prima facie it encompasses a very broad range of rights, it may be regarded as a general name for a cluster of rights related to human health, some of which enjoy constitutional protection in Israel. Given the many aspects of the right, Court President Beinisch held that there is no point in examining the constitutional status of the right as a whole, but rather that it is preferable to evaluate the underlying reasons for the different rights and interests they protect, in terms of their relative social importance and the intensity of their link to the constitutional rights enumerated in the Basic Law.51 The right to health care, and specifically to publicly funded health care, is not explicitly mentioned in the Basic Law, which raises the difficult question of whether it amounts to a constitutional right, especially to the extent that its recognition imposes a positive duty on the state that encompasses its responsibility for financing of health services in Israel.52 The Supreme Court held that constitutional rights in the Basic Law may include elements touching on the area of social welfare and security, including the sphere of health, but it is unclear whether this means interpreting the Basic Law as including a constitutional right to health services which is broader than the basic level necessary for human existence within society. The dilemma is between the centrality of health to human life and the ability to realise all other human rights on the one hand, and the reality in which the right to public funded health services entails the imposition of a positive duty, and hence involves general distributional questions, on the other hand. While acknowledging that the state’s duties in the sphere of civil rights may also involve a ‘positive’ duty which involves the allocation of resources, Beinisch opined that the right to publicly funded health services typically involves a dominant ‘positive’ dimension, which places issues of allocation of social resources based on national priorities in a higher register.53 Thus, given the need of the Supreme Court to be wary of shaping economic policies and interfering with national priorities, it is doubtful whether one should read into the rights in the Basic Law a right which would impose a correlative duty to provide public health services on a level higher than the minimum required for human existence in society.54 Moreover, even under the assumption that the right is included within the constitutional rights, like all other rights it is not absolute and would have to be balanced against competing rights and interests.55 For the purposes of the case in question, Beinisch ruled that it sufficed to determine that given that the drug in question is a novel medicine for colon cancer, and there is no consensus regarding its ability to save or prolong life, it would seem that the medi ibid para 5. ibid para 9. 52 ibid para 10. 53 ibid. For a critique of such a distinction between civil and social rights, see Barak-Erez and Gross, ‘Do We Need Social Rights’ (n 13) 8. 54 Louzon (n 48) para 10. 55 ibid. 50 51
The Right to Health 323
cine in question as well as other new and experimental medications, are not part of the basic health services required for minimal human existence in society, and that given the limited public resources, it is doubtful if the demand for public funding of new medications, may be anchored in the core of the Basic Law’s protected constitutional rights.56 Furthermore, even in accordance with an interpretive position which would expand the constitutional dimension of the right to human dignity beyond the necessary minimum, only in exceptional circumstances would the state be constitutionally obligated to fund a specific drug – out of numerous drugs for which demands for public funding are raised.57 2. Scope of the statutory right to health services: After addressing the constitutional issue, the Supreme Court emphasised that the right to public health services does exist as a statutory right, regardless of the constitutional question, and this warrants an examination of whether this statutory right was violated.58 The right itself is anchored in the Patient’s Rights Law which determines that every person is entitled to appropriate health care in accordance with the Law and in accordance with the conditions and arrangements valid at the time in the Israeli health system,59 as well as by the NHIL.60 Beinisch analyzed the structure of the NHIL, determining that the Law is based on the principle of solidarity, its aim being for each insured person to pay in accordance with his ability and to receive according to his needs, having consideration for the weaker members of society. Court President Beinisch noted, however,61 that over the years a series of amendments to the NHIL had eroded the principle according to which the right to receive medical services is an independent right, detached from the ability to pay for the service. After addressing these issues, as well as the mechanism for updating the cost of the HSB and the widespread criticism thereof, and the ensuing litigation, Beinisch concluded that the Government has wide discretion in determining the annual addenda to the cost of the HSB.62 Based on the statutes, Beinisch ruled that while the Israeli legal system recognised a statutory right to health services which is broader than the core minimum of basic health services required for human existence in society, it is nonetheless clear that the HSB does not purport to include the full range of possible medical services that an individual may require. The statutory right, it was held, includes a core and a penumbra. The core of the statutory right to basic health services includes the health services that the state is committed to fund. In accordance with the NHIL this includes the ‘basic services basket’ that was in force in Clalit, the largest Sick Fund, before the NHIL entered into force on 1 January 1994,63 and which served as the baseline under the statute, and the funding of which within the HSB is a statutory duty. At the penumbra of the statutory right are the health services not included in this baseline, and the right to expansion within the HSB beyond the baseline is a budget-dependent right, which derives
ibid para 12. ibid. 58 ibid para 13. 59 ibid para 14. 60 ibid para 15. 61 ibid, citing Gross, ‘Health in Israel: Between a Right and Commodity’ (n 12) 495. 62 Louzon (n 48) para 15. 63 This is according to the NHIL which, upon coming into effect, established that basket as the starting point for the HSB. 56 57
324 Aeyal Gross from policy as determined in the annual budget law.64 Thus, the scope of the statutory right to public health services beyond the ‘basic basket’ derives from the annual budget law. Having said that, Beinisch added one caveat: the budget limit is not ‘an impenetrable roof’: international law indicates the duty to act for the progressive realization of the right, and when the state takes retrogressive measures in relation to the right, it assumes the onus of showing that these measures are necessary given the maximum resources available to it: [T]he question that is likely to arise in our legal system is whether a serious reduction in the funding of the health services basket – including by way of significant cumulative erosion of the funding of the basket in the absence of a substantive mechanism for a real adjustment of its cost – transfers the burden to the State to show that this reduction is indeed justified and dictated by reality.65
This interpretive question, Beinisch added, would necessitate an examination of whether the statutory right for the expansion of the HSB, while budget-dependent, should be interpreted in accordance with international law and the principle of progressive realization. Beinisch noted, however, that this is not the question before the Supreme Court in the present case and should be left open.66 This final comment relied on the fact that the petitioners in Louzon did not challenge the budgetary framework allocated by the state for the expansion of the HSB but rather attacked the exercise of discretion by the basket committee. Given this analysis Beinisch could conclude that since Erbitux was not part of the baseline basket under the NHIL, access to it is not a statutory, but rather a budget-dependent right. 3. The determination of the basket and the discretion of the basket committee: In this part of the judgment Beinisch addressed the procedure for adding new technologies and described the process in which on an annual basis the Ministry of Health issues a call for requests to include new medications and technologies in the HSB. This is followed by an evaluation process of the various proposals, conducted by professional bodies in the Ministry which then pass the information to the HSB Committee. The HSB Committee prioritises the new medications and technologies, and presents its recommendations to the Health Council, a statutory body authorised to advise the Minister of Health. If the Minister accepts the recommendations, then further consent is needed from the Minster of Finance, to be followed by the consent of the Government, which must also approve the financial resource to expand the HSB.67 The Committee, it was held, enjoys a wide range of discretion and the Supreme Court will not intervene in its discretion as long as its recommendations were formulated in a proper procedure and as long as it does not substantially deviate from the relevant considerations which it is obligated to take into account or clearly deviates from striking an appropriate balance between them.68 Louzon (n 48) para 16. ibid para 17. In this context Beinisch drew on the ICESCR and on UN Committee on Economic, Social and Cultural Rights, General Comment No 14: The Right to the Highest Attainable Standard of Health (Art 12 of the Covenant) (11 August 2000) UN Doc E/C 12/2000/4. 66 Louzon (n 48) para 17. 67 ibid para 20. 68 ibid paras 21–23. For a discussion of the relevant considerations as laid out by the Committee before the court, see ibid para 26. 64 65
The Right to Health 325
The Supreme Court did not find any grounds for intervention, noting that the question of how to set priorities in the allocation of public resources is a controversial one, admitting of differing opinions, and that it is not the Court’s role to choose one prioritization system over another.69 Having perused the protocols of the HSB Committee’s discussion, the Supreme Court ruled that the scientific evidence about the drug in question was still elementary and non-conclusive. Accordingly, the Committee’s decision to give it a rating that was lower than the rating of proven life-saving technologies is not unreasonable to an extent that warrants judicial intervention.70 The Supreme Court’s analysis in Louzon enables the following observations: regarding the constitutional status of the right, the Supreme Court adhered to the minimalistic approach, maintaining the traditional yet problematic division between civil and polit ical rights on the one hand and social and economic rights on the other hand. It remains to be seen whether the more recent Hassan case mentioned earlier, which rejected this distinction, will give rise to changes in future right to health cases. But no less problematic is the Supreme Court’s holding concerning the scope of the statutory right. The division between the core and the penumbra may indeed have its parallel in the recognition of ‘core obligations’ in the right to health.71 However, the Supreme Court chose to identify the core obligations with the 1994 baseline HSB, so that anything beyond it is within the scope of budget-dependent penumbra. This particular choice neglects a substantive test, which requires consideration to be given to the changing meanings of what is at the ‘core’ in accordance with technological developments, thus confusing the ‘is’ with the ‘ought’. The combination of the Supreme Court’s holdings in the three aspects of its judgment seems to close the door on the possibility of judicial review of decisions concerning the inclusion of new technologies. Even if new technologies or drugs are more established and consensual than Erbitux, the Supreme Court’s analysis precludes their inclusion within the scope of the statutory right and it would seem practically impossible to make claims on a constitutional or administrative level. Indeed, the state has allocated budgets for new technologies since 1998, but this budget has been declared to be insufficient for purposes of covering all essential and life-saving drugs. It relies upon the Government’s annual budgetary decision and is not guaranteed, and hence cannot be trusted, especially given the lack of an adequate regular updating mechanism as described above. As for the third aspect of the judgment, the criterion for intervention set by the Supreme Court may be a sound one as indeed the Supreme Court cannot be expected to replace the Committee’s discretion with its own discretion. Nonetheless, Louzon still reads like a missed opportunity: consider the paragraphs in which Beinisch suggests that significant and cumulative erosion of the financing of the basket and the lack of a substantial mechanism for a realistic update of its cost may amount to a severe violation of the financing of the HSB (‘retrogressive measures’)72 which imposes a burden upon the ibid paras 27–28. ibid para 29. 71 See, A Gross, ‘The Right To Health in an Era of Privatisation and Globalisation: National and International Perspectives’ in D Barak-Erez and A Gross (eds), Exploring Social Rights: Theory and Practice (Oxford, Hart Publishing, 2007) 289, 303; L Forman, ‘What Future for the Minimum Core? Contextualizing the Implications of the South African Socioeconomic Rights Jurisprudence for the International Human Rights to Health’ in J Harrington and M Stuttaford (eds), Global Health and Human Rights (London, Routledge, 2010) 62, 67–70. 72 Louzon (n 48) para 17. 69 70
326 Aeyal Gross state to demonstrate that the said violation is indeed justified and necessary. Indeed, Beinisch determined that this was not the issue confronting the Supreme Court, and the petitioner did not challenge the budgetary allocation for health care but rather its determination of priorities therein. Nonetheless, it is arguable that on a factual level it was indeed the erosion of the financing of the health basket, as described, and the lack of a substantial mechanism for realistic update of its cost as discussed above, that formed the background to the litigation in Louzon. The reason is that these factors caused the exclusion from the HSB of technologies that do not rank sufficiently high on the Committee’s list, despite their life-saving or life-extending potential. Indeed the Committee should have wide discretion and as a rule it is difficult to justify intervention in the priorities it sets. However, the state’s policies on health care may indeed justify a transfer of the onus to the state to explain and justify any significant reduction of public funding for health care. In that scenario, judicial intervention would be justified if the explanation was inadequate. Accordingly, while the petitioners in this case did not challenge the budget allocated for the health care basket, but rather the way in which it was allocated, the Supreme Court could have seized on the opportunity to transfer the burden of proof and determine that the state’s neglect of this issue created a presumption against the propriety of its decisions. Litigation on access to medications in Israel should be viewed within the context of the specific issues facing the Israeli health system and rights litigation, and while it cannot replace policy making, in cases such as Louzon it may arguably serve as a catalyst for a review of the policy from a rights perspective.73 C. Co-Payments While the question of the scope of the services included in the HSB generated extensive litigation, the issue of co-payments has merited far less direct judicial attention, despite the fact that as described above, the expansion of co-payments after the 1998 legislation created bars to accessibility and made co-payments a major funding source for the health system, and in a way that substantially increased inequalities. As mentioned, in the Israeli (Cochlear Implant) case, the Supreme Court rejected arguments that this type of payment infringes on the rights to health and equality. D. The Mixture of Public/Private Programmes The term ‘Private Health Services’ (PHS)74 is used to refer to a programme that enables patients to choose their doctor – specifically surgeons – within public hospitals, for an additional out-of-pocket payment by the patient. These programmes have been promoted in government hospitals since 1996, by way of private companies that entered into agreements with the hospitals’ research funds.75 In 2002, the Attorney General pronounced the 73 Erbitux, by the way, was added to the HSB in 2012. Director General Directive 1/12, ‘Expansion of the Health Services Basket for 2012’ (10 January 2012) (in Hebrew): www.health.gov.il/hozer/mk01_2012.pdf (Ministry of Health) (in Hebrew). 74 The Hebrew acronym is ‘SHARAP’. 75 For a discussion, see Ben-Nun, Berlovitz and Shani (n 26) 131–37 (in Hebrew); Y Shuval and O Hanson, Ha’Ikar HaBriut (Most Importantly, Health) (Jerusalem, Magnes, 2000) 307–14 (in Hebrew).
The Right to Health 327
practice illegal. Petitions against his decision were filed in the Supreme Court by a group of doctors and a group of citizens and in 2009 the Supreme Court decided to reject the petitions and uphold the Attorney General’s decision concerning the illegality of the PHS (the Kiryati ruling). In addressing the matter, Justice Berliner noted that in reality, the PHS was limited to the selection through payment of surgeons in government hospitals. Nonetheless, the issue before the Supreme Court actually has far broader implications, because permitting PHS would also legitimate the provision of a wide array of health services by means of a payment on top of the health tax. This could have significant implications for the character of the public health system in Israel, which she described as national and public, and as one which is subsidised by the state, with each citizen paying a fixed monthly sum from his or her salary regardless of his or her medical condition, and receiving treatment in accordance with his or her needs.76 Berliner contrasted this system to the PHS regime under which the patient can choose the surgeon based on a private out-of-pocket payment, which is usually used to select the more senior doctors who naturally have greater expertise in their respective fields.77 Health services in the hospitals should be provided exclusively in accordance with medical considerations, and under no circumstances should they be based on enhanced ability to pay.78 The service provided for citizens within state hospitals is a public service given for free, and any ability to buy an improved service within the public service which is normally given for free, must be authorised by the legislature.79 The petitioners in this case attempted to enlist rights arguments in making the case for allowing PHS. It was argued that whether desired or not, the health system in Israel does not treat all citizens equally being based rather on equality of opportunity, and as such, all people are entitled to buy the PHS services. The prohibition of PHS violates the rights enumerated in the Patient’s Rights Law as well as in the Basic Law, which protects the autonomy of the individual, including the individual’s right to select the doctor and the hospital in which he or she is to be treated.80 Justice Berliner rejected these arguments, relying on the previous Supreme Court rulings which recognised the right to dignity as including a minimum human existence but not a broader right to health. Accordingly, a right to choose a physician, is not part of the core services required for the maintenance of human existence in dignity.81 Additionally, if the right to choose a physician were to be considered as a part of the personal autonomy recognised by the Basic Law, it would mean allowing every patient to select a physician without payment, as part of the HSB. Yet the protection of autonomy does not entitle the citizen to the realization of each and every one of his desires, just because he can pay for it.82 Justice Berliner’s ruling in Kiryati demonstrates a commitment to the principles of the NHIL, and to a public health system in which medical services are provided in accordance with need and not just the ability to pay. Her rejection of the rights arguments made by the petitioners demonstrates a commitment to an egalitarian concept of health which is provided in accordance with need and not in accordance with ability to pay, 76 HCJ 4253/02 Kiryati v Attorney General (17 March 2009), Pador Legal Database (by subscription) para 2 (in Hebrew). 77 ibid para 3. 78 ibid paras 34, 41, 48. 79 ibid para 36. Justice Berliner based her decision also on a few statutory provisions (see paras 26–30). 80 ibid para 24. 81 ibid para 52. 82 ibid para 57.
328 Aeyal Gross and she justly held that if autonomy means the right to choose one’s doctor, than it should be provided to all patients, and not just to those who can pay for it. This ruling is in contrast with the Canadian Supreme Court decision in Chaouli 83 which protected access to private health insurance – in the name of the protection of the right to the security of the person – for those who can pay for it. Interestingly, in reaching her conclusion, Justice Berliner relied on the narrow, minimal recognition of the right to health as part of the right to human dignity as recognised in Israeli law. In this case it was a minimal construction of the right to health that served to justify equal access to health care rather than a broader construction. All the same, given Justice Berliner’s rejection of the argument from autonomy, she could have rejected an argument made from the right to health even if the right had been given a broader construction in Israeli law. That is to say – the rationale for rejecting the right to autonomy claim could also have been used for rejecting the claim based on the right to health. Both of them are rights to which all persons should have equal access to regardless of ability to pay. Rejecting the argument made from the right to health on these grounds would have been a better response than the one given in the judgment. From reading Kiryati it is evident that rights discourse can be, and was in fact, used by the petitioners for the promotion of privatization within the health care system in a manner that creates inequality. However, a concept of rights grounded in a substantial concept of equality84 – which was adopted in this case by the Supreme Court – should reject these claims and point to the fact that for a rights argument to be justified, it must be shown that access to health care is equal and not dependent on the ability to pay. At the same time, and as the petitioners justly argued, the principle of equality is already undermined by many other factors. It will be recalled that while Kiryati prohibited discrimination between patients within the public hospital in regard to choice of surgeon based on the ability to pay, the Israeli (Cochlear Implant) decision upheld discrimination within the public hospital, based on the ability to pay, even where that ability determined the very possibility of undergoing an essential operation. This gap points to the limits of the Supreme Court’s discourse. It may in fact indicate that it was easier for the Supreme Court to protect equality in Kiryati, where its decision was limited to the upholding of policy and did not require intervention. On the other hand, and effectively retreating somewhat from the Kiryati ruling, the Supreme Court rejected a later petition which challenged the legality of PHS in publicly funded hospitals which provide public services, but are operated by private companies and are not government hospitals in the same way as the hospitals addressed in Kiryati were. Writing for the Court, Court President Grunis based his reasoning mostly on the laches doctrine, holding that the petition was submitted after a significant delay, but also distinguishing the case from Kiryati based on the difference between the types of hospitals involved. Addressing the question of whether because of the nature of the rights involved the Supreme Court should consider the merits of the case notwithstanding the delay, the Court noted that even if the right to equal access to health care is to be recognised as derivative of the constitutional right to human dignity, then the right is at the
83 Chaoulli v Quebec (AG) [2005] 1 SCR 791 (Can). See A Gross, ‘Is There a Human Right to Private Health Care?’ (2013) 41 Journal of Law, Medicine and Ethics 138. 84 See on this, M Horwitz, ‘Rights’ (1988) 23 Harvard Civil Rights–Civil Liberties Law Review 393.
The Right to Health 329
periphery and not at the core of the constitutional right.85 This statement further points to the weak status of the right to health in Israeli constitutional law and to a position which is at odds with the need to integrate the concept of equal access into its analysis. In the absence of such a concept, health’s status as a right will be further eroded, increasingly becoming a commodity, the availability of which is based on the ability to pay and not based on need. V. CONCLUSION: WHAT ROLE FOR THE RIGHT TO HEALTH?
The question of the role of the right to health and related litigation has been debated extensively in recent years, raising a few related questions and dilemmas: what is the scope of the right to health, and more pointedly, what should its contents be (if at all) as a judicially enforced right? Is it possible, and if so is it desirable, for the courts to order the allocation of health services which national authorities decided not to provide? Would not judicial intervention in the area of health undermine the work of the bodies authorised to prioritise in their capacity as bodies that see the ‘whole picture’ rather than ‘just’ the plight of an individual patient? How can rights discourse with its individualist tilt address public health, an issue involving collective decision-making?86 Furthermore, it has been argued that the focus in right to health litigation on access to health care rather than on other elements of the right to health, moves the discussion, and possibly the resources, away from the social determinants of health and away from preventive medicine, towards curative medicine.87 It has also been suggested that right to health litigation may benefit the middle class at the expense of the poor, and powerful arguments have been made in favour of further research of the question of who litigates, who benefits, and how health rights litigation affects the overall equity of the system.88 All of these questions merit discussion. Indeed health rights, and specifically health rights litigation, cannot replace policy and prioritization decisions. However, they can serve as a tool for reviewing whether a chosen policy adheres to principles of rights. Introducing the notion of the right to health into public health systems, especially at times of privatization of financing and of provision of services, may reinstate public values such as equality. At the same time it may reinforce privatization by allowing demands for privatised medicine to be dressed up as rights claims. The Israeli experience reflects both of these, even if the Supreme Court thus far has rejected attempts to achieve the latter. 85 HCJ 2114/12 Association for Civil Rights in Israel v Government of Israel (15 August 2012), Nevo Legal Database (by subscription) (in Hebrew). 86 See Gross, ‘The Right to Health in an Era of Privatisation and Globalisation’ (n 71) 337–38 in A Clapham and S Marks, International Human Rights Lexicon (Oxford, Oxford University Press, 2005) 197. 87 See, eg BM Meier, ‘The World Health Organization, Human Rights, and the Failure to Achieve Health for All’ in J Harrington and M Stuttaford (eds), Global Health and Human Rights (London, Routledge, 2010) 163, 182. 88 See, eg S Gloppen, ‘Litigation as a Strategy to Hold Governments Accountable for Implementing the Right to Health’ (2008) 10 Health and Human Rights 21. For comparative studies on the role of law and litigation in this regard, see V Gauri and DM Brinks, Courting Social Justice: Judicial Enforcement of Social and Economic Rights in the Developing World (Cambridge, Cambridge University Press, 2008); AE Yamin and S Gloppen (eds), Litigating Health Rights: Can Courts Bring More Justice to Health? (Cambridge, Harvard University Press, 2011); C Flood and A Gross, The Right to Health at the Public/Private Divide: A Global Comparative Study (Cambridge, Cambridge University Press, forthcoming).
330 Aeyal Gross An understanding of the right to health grounded in a distributive and substantive concept of equality may help avoid the danger of rights discourse being invoked to reinforce privatization and make it possible to use rights analysis to scrutinise policy decisions according to the principles of progressive realization and equal access. In the Israeli context, this would mean that while health rights litigation cannot replace properly conducted prioritization, it may prompt judicial review of whether the policies taken by the Government advance an egalitarian concept of the right to health. Do the policies reflect a commitment to broadening accessibility and to a growing commitment to health on the part of the state, or a withdrawal from such commitments? Does the state show a commitment to progressive realization of the right to the highest attainable level of health? Do the health policy decisions advance equal access, or do they actually promote inequality? Is the health care system financed on an egalitarian basis, or rather in a way that imposes a disproportionate burden on the poor? Asking these questions might have produced different results, or at the very least, might have lead to lines of reasoning and consideration of factors that do not appear in some of the Israeli cases mentioned here. Based on the record of the Israeli Supreme Court, we can see that it actually refrained from intervening in any of the major right-to-health cases brought before it. The right to health claim was ineffective as a sword for those who sought to advance equal access, but at the same time attempts to argue for privatization using right to health arguments also failed. It was actually in the latter cases, especially in Kiryati, that the Supreme Court developed an egalitarian concept of the right to health that served as a shield against attempts to expand privatization of the system, when such privatization was already restricted by the Government. The reasoning employed by the Supreme Court when it used the rights analysis as a shield against arguments based on the right to privatization, if taken seriously, could have also been used as a sword against privatization and against withdrawal from commitment to health. However, treading the latter path would have required the Supreme Court to enter the arena and intervene in the form of accepting petitions, something it has thus far been reluctant to do in this area. The second PHS case, points to this reluctance, but also to the Supreme Court’s possible withdrawal from the egalitarian perspective taken in Kiryati. Arguably, with the exception of the most recent petition, in the series of cases concerning the health cost index update, non-intervention is the uniting theme of the Supreme Court’s jurisprudence on the right to health – more than any other aspect of this right. Unlike the Supreme Court cases, in some of the Labour Court cases, the courts chose to intervene in favour of patients. Decision-making in individual cases of the type that come before the Labour Courts may indeed obscure the general picture. But on the other hand, given its ability to see the individual coming before it, the Labour Courts may be ideally positioned to address concrete cases which ‘fall between the cracks’ and do not find a solution in the HSB. Judicial review of the existence and the operation of committees within the Sick Funds that look into such cases is a welcome development. Take the case in which a certain medication is included in the HSB but only as a second line medicine, and because of a certain condition or a conflicting drug the patient cannot take the first line medicine. These committees (and in their absence, or in the event that they fail to exercise reasonable discretion, the Labour Courts) have the ability to provide the patient with the required remedy, given the need to see the ‘concrete’
The Right to Health 331
person before it and not just ‘abstract’ other,89 who may not find a solution within the HSB. From this perspective, the decision of the Israeli Labour Courts that required the Sick Funds to exercise discretion in cases not covered in the health basket, along with the judicial review of the decisions made by the Sick Funds’ Exceptions Committees, are examples of what rights analysis may achieve in the context of health. The approach advocated here does not ignore the budget issues or the need to set priorities. However, when a right exists, it is the state’s duty to find the appropriate resources, and it cannot waive its responsibility based on budgetary arguments alone. The state can set priorities within the existing health system and can prefer one treatment over another if it is more essential, effective, or cheap, provided that it upholds the core of the right to health, the principle of accessibility (including economic accessibility), and equality, and provided that it demonstrates a commitment to health and progressive realization of the right. Beyond the individual cases decided by the Labour Courts, such a commitment dictates a departure from the minimalistic approach adopted by the Supreme Court in the cases that come before it, towards a constitutional approach grounded in equity and substantive equality that will scrutinise decisions based on these principles. Access to health is at least as important, if not more, than access to watching soccer games.
89 I use these terms following Seyla Benhabib’s distinction in S Benhabib, Situating the Self (London, Routledge, 1992) 102–48, which in turn draws on Carol Gilliagn’s discussion of ethics of justice and ethics of care: C Gilligan, In a Different Voice (Cambridge, Harvard University Press, 1994). See in more detail Gross, ‘The Right To Health in an Era of Privatisation and Globalisation’ (n 71) 329–30.
22 The Fiscal Objection to Social Welfare Rights: A Closer Look AMIR PAZ-FUCHS*
I. INTRODUCTION
T
HOUGH IT HAS been repeatedly asserted that theoretical arguments against recognition of social welfare (SW) rights have been effectively answered,1 they tend to resurface in implicit fashion in legal doctrine. This chapter argues that at the core of contemporary manifestations of the traditional objections to SW rights is their unique relationship with money. The chapter sheds light on the structure of the current version of the old objections by reviewing judicial rulings in courts in general and in Israeli courts in the fields of health, education and welfare in particular. Focusing on this one particular justification – that objections to social and economic rights collapse into reservations regarding judicial abilities to deal with fiscal issues – carries the potential to highlight the ambiguities and inconsistencies in the judgments and in the arguments that underlie them. But before doing so, some preliminaries must be addressed. First, as always: terminology. But in this case, terminology is also substantive. Scholars writing on the theoretical and pragmatic objections to judicial enforcement of rights to education, health, housing and the like, tend to use the term ‘social and economic rights’.2 This chapter, however, will follow the lead suggested by Mark Tushnet and use the term ‘social welfare rights’.3 As Herman Schwartz and others have recently noted, ‘the central issue is not really about social and economic rights, but primarily about social rights. More precisely, it is
* I would like to thank Mark Tushnet, Aeyal Gross, Neta Ziv and other participants in the conference for thoughtful and helpful comments. Unless otherwise stated, all translations are mine. 1 J Waldron, ‘Introduction’ in J Waldron (ed), Theories of Rights (Oxford, Oxford University Press, 1984) 1, 11. 2 There are good reasons for doing so. After all, international documents bind social and economic rights together. See, eg International Covenant on Economic, Social and Cultural Rights (adopted 16 December 1966, entered into force 3 January 1976) 993 UNTS 3; UN Commission on Human Rights, Note verbale dated 5 December 1986 from the Permanent Mission of the Netherlands to the United Nations Office at Geneva addressed to the Centre for Human Rights (‘Limburg Principles’) (8 January 1987) UN Doc E/ CN4/1987/17; Maastricht Guidelines on Violations of Economic, Social and Cultural Rights (22–26 January 1997). 3 M Tushnet, ‘Social Welfare Rights and the Forms of Judicial Review’ (2004) 82 Texas Law Review 1895, 1895.
334 Amir Paz-Fuchs about social and certain economic rights’.4 The reason for excluding paradigmatic economic rights such as property and contract from this discussion lies not only in the fact that there is little controversy over the legitimacy of their judicial enforcement, but also because judicial deference to economic rights such as property and contract conflicts with the redistributive interests that lie at the heart of social welfare rights.5 The second preliminary note comes in the form of a qualification: the chapter does not seek to confront the issue that has occupied many American constitutional scholars, to wit: whether social and economic rights belong in a constitution.6 The normative focus of this chapter, with its affinities to legal realism, agrees that ‘there is no reason to think that it is the constitutionalization of these rights that is crucial’.7 Courts may have profound impact on social policies without resorting to judicial review of legislation. First, because a significant amount of policy decisions, on the micro and macro level, are made at the sub-legislative, administrative, level.8 Second, even where legislation is involved, courts can enforce rights through other methods, such as the interpretation of statutes.9 Turning to the substantive focus of the chapter, it is necessary to ask: what is the advantage of analyzing the judicial enforcement of social and economic rights through the perspective of money? The fact that the perspective has largely been swept away as ‘an offense to polite manners’10 is a good enough reason to finally do so. But it is far more than that. One may say that the same, well-worn theoretical objections to SW rights that were, as noted, effectively answered, are being redressed in fiscal clothes. Three such objections are noted: first, that the judicial enforcement of SW rights endangers proper separation of powers; second, that the courts lack the expertise and competence to deal with the complex issues that SW rights raise; and third, that vexing enforcement problems related to SW rights will reflect poorly on rights in general, and will endanger respect for first generation, civil and political (CP) rights. Now, while these objections were formulated, at first, in a straightforward fashion, today they almost always seek reinforcement from one or more of the usual concepts: budget, funding, resources, and the like. But, of course, the substantive rebuttal to the objections still remains, and should be articulated. 4 H Schwartz, ‘Do Economic and Social Rights Belong in a Constitution?’ (1995) 10 American University Journal of International Law and Policy 1233, 1235; also D Barak-Erez and A Gross, ‘Social Rights and the Struggle for Social Citizenship’ in S Almog and Y Rotem (eds), Dorner Book (Srigim, Nevo, 2009) 189, 190. 5 M Tushnet, Weak Courts, Strong Rights: Judicial Review and Social Welfare Rights in Comparative Constitutional Law (Princeton, Princeton University Press, 2008) 170–94. For the impact of social welfare rights on redistribution, see D Barak-Erez and A Gross, ‘Introduction: Do We Need Social Rights – Questions in the Era of Globalization, Privatization and the Diminished Welfare State’ in D Barak-Erez and A Gross (eds), Exploring Social Rights: Between Theory and Practice (Oxford, Hart Publishing, 2007) 1, 16. 6 F Michelman, ‘The Constitution, Social Rights and Liberal Political Justification’ in Barak-Erez and Gross, Exploring Social Rights (n 5) 21; C Fabre, Social Rights Under the Constitution (Oxford, Oxford University Press, 2000); H Schwartz, ‘Economic and Social Rights’ (1993) 8 American University Journal of International Law and Policy 551. 7 Schwartz, ‘Economic and Social Rights’ (n 4) 1243. 8 R Titmuss, ‘Welfare “Rights”, Law and Discretion’ (1971) 42 The Political Quarterly 113, 120–21, 124 (‘public assistance in the United States is almost entirely discretionary’); M Diller, ‘The Revolution in Welfare Administration: Rules, Discretion and Entrepreneurial Government’ (2000) 75 New York University Law Review 1121, 1147; J Handler, ‘Discretion in Social Welfare: The Uneasy Position in the Rule of Law’ (1983) 92 Yale Law Journal 1270, 1276; D Barak-Erez, ‘The Israeli Welfare State – Between Legislation and Bureaucracy‘ (2002) 9 Labor, Society and Law 175, 177–78 (in Hebrew). 9 Tushnet, ‘Social Welfare Rights’ (n 3) 1898; Tushnet, Weak Courts, Strong Rights (n 5) 224. 10 S Holmes and C Sunstein, The Cost of Rights: Why Liberty Depends on Taxes (New York, WW Norton, 1999) 24.
The Fiscal Objection to Social Welfare Rights 335
In what follows, I will briefly outline the three arguments, show each one’s modern association with budgetary concerns, and note some instances that reveal how this association was expressed in Israeli case law. Part II will discuss the argument concerning separation of powers; part III will focus on the argument concerning the courts’ competency to deal with SW rights; and part IV will discuss the objection to SW that is concerned with lack of applicability and the impact that ‘futile’ efforts to implement SW rights supposedly has on ‘real’ (ie CP) rights. I then turn, in part V, to a focused rebuttal of the fiscal objection. The chapter concludes, in part VI, with a suggestion as to why this particular objection has remained so attractive to courts and to scholars alike. I believe that the latter analysis suggests a theoretical reconstruction that has implications not only to the subject at hand, but also to the way SW rights, and rights in general, are reconstructed in contemporary judicial discourse. II. SEPARATION OF POWERS
According to this argument, the judiciary should refrain from enforcing SW rights because acting otherwise would require the courts to adjudicate policies, priorities and, ultimately – to distribute and redistribute funds. Such a result is tantamount to the unconstitutional, illegitimate and undemocratic transfer of authority from Parliament and government to the courts.11 Timothy Macklem’s exposition of the argument reveals its strong ties to matters of resources and funding. He argues that allowing the courts power to enforce social welfare rights would grant them the role to decide the level of funding that health care should receive from the government, and . . . how that funding should be distributed . . . It would be for the courts to set the direction for the economy, to establish the curriculum for the schools, to determine environmental policy – in short, to govern. Clearly this would be undesirable, for it would have the effect of transferring virtually all democratic authority from the people’s present representatives . . . to the courts whose duty it is to interpret and enforce that Constitution.12
Frank Cross drives the point home, employing the timeless phrase and stating that ‘courts understand that requiring legislatures to provide minimal levels of subsistence for all Americans encroaches upon the jealously guarded “power of the purse”’.13 This, indeed, fits well with the US Supreme Court’s self-awareness, remarking as it did that ‘the intractable economic, social, and even philosophical problems presented by public welfare assistance programs are not the business of this Court’.14 Across the ocean, the House of Lords has been characterised as holding ‘a deeply embedded judicial conviction that matters of public finance are the preserve of the elected branches of government and not of courts’.15 And the Irish Supreme Court concluded that ‘it is not the function of the courts to make an assessment of the validity of the many competing claims on national resources’.16 11 See the discussion in G Davidov, ‘Constitutional Review of Budgetary Matters’ (2007) 49 The Lawyer 345, 348–49 (in Hebrew). 12 T Macklem, ‘Vriend v Alberta: Making the Private Public’ (1999) 44 McGill Law Journal 197, 210. 13 F Cross, ‘The Error of Positive Rights’ (2001) 48 UCLA Law Review 857, 890. 14 Dandridge v Williams 397 US 471, 487 (1970); Mathews v Eldridge 424 US 319, 348 (1976). 15 ED Palmer, ‘Resource Allocation, Welfare Rights’ (2000) 20 OJLS 63, 74. 16 TD v Minister for Education [2001] 4 IR 259 (Ir).
336 Amir Paz-Fuchs And yet, on a normative level, it is simply not true that the courts can refrain from intervening in questions of distribution and redistribution. Not intervening, in such cases, would usually mean embracing the highly controversial baseline of the market, or even accepting the economic status quo that, in itself, depends on public institutions and state action.17 The decision not to alter the ‘background rules’ that inform the current social and economic state of affairs is a decision in its own right. Even the distinction between ‘commission’ and ‘omission’ is not relevant in this case. For, in constitutional democracies, courts quite often protected rights to property and contract, at the expense of other social and legal interests (the Lochner era being an obvious example).18 In Professor Tushnet’s words: ‘The state is complicit in creating the distribution of wealth in society whether it “acts” affirmatively or whether it does nothing but enforce the background rules of property and contract law’.19 However, as shown by the example of Israeli health policy, the courts tend to accept the bond between fiscal considerations and the threat to separation of powers. Some background is warranted: the National Health Insurance Law, 5754-1994, establishes a basket of health services to which citizens are entitled, and the process for updating the basket. The Law grants the Ministers of Health and Finance the power to update the basket, following a recommendation by the Health Council.20 In Maccabi Health Services v Minister of Finance,21 the health fund challenged the Minister’s refusal to update the health index, despite a unanimous recommendation to that effect by the Health Council, which was buttressed by a similar assertion by a parliamentary committee of enquiry. The Minister recognised the objective need to update the health index, but replied that budgetary constraints restrained him from acting accordingly. The Israeli Supreme Court (ISC) dismissed this argument, mocking it as an attempt to devise a new theory of ubi remedium ibi jus – where there is a remedy there is a right22 – instead of the other way around. The Court clarifies that where a statutory obligation is recognised, the state cannot argue that fiscal resources prevent it from living up to its duties. The Court also noted that ignoring the Health Council’s opinion, undermines the Council’s statutory standing as an expert advisor in the process.23 However, after all this
A Sen, Inequality Reexamined (Oxford, Clarendon Press, 1992) 1–100; Holmes and Sunstein (n 10) 61. Lochner v New York 198 US 45, 45 (1905) (striking down a New York law that limited the work day to 10 hours, and the work week to 60 hours, for being an ‘unreasonable, unnecessary and arbitrary interference with the right and liberty of the individual to contract’); Adkins v Children’s Hospital 261 US 525 (1923) (striking down a law providing minimum wages for women and children); Adair v United States 208 US 161 (1908) (upholding ‘yellow dog’ contracts that forbid workers from joining unions); Tyson & Brother v Banton 273 US 418 (1927) (striking down a law regulating the price of theatre tickets); Weaver v Palmer Bro Co 270 US 402 (1926) (striking down a public health and safety regulation concerning the use of fabrics as a violation of the due process clause). See C Sunstein, ‘Lochner’s Legacy’ (1987) 87 Columbia Law Review 873, 883 (arguing that the ‘central problem of the Lochner Court had to do with its conceptions of neutrality and inaction and its choice of appropriate baseline’). 19 Tushnet, Weak Courts, Strong Rights (n 5) 189; for Tushnet’s own discussion of the Lochner era see ibid 172–74. 20 The Health Council is chaired by the Minister of Health, and consists of 46 members from government ministries, health funds, Israel Medical Association, academia, trade unions, employers and municipalities – National Health Insurance Law, 5754-1994, s 49. 21 HCJ 2344/98 Maccabi Healthcare Services v Minister of Finance 54(5) PD 729 [2000] (in Hebrew). 22 See also Holmes and Sunstein (n 10) 43 (explaining that individuals enjoy rights in a legal, as opposed to a moral, sense, ‘only if the wrongs they suffer are fairly and predictably redressed by their government’). 23 Maccabi (n 21) 761–62. 17 18
The Fiscal Objection to Social Welfare Rights 337
‘lofty talk’,24 the Court concluded that it cannot order the Government to spend the sums that would result from accepting the petition: ‘We have never instructed the state to pay anyone out of its budget amounts of such magnitude as those that the health funds are asking for in the present matter’.25 III. COMPETENCE
Critiques of judicial enforcement of SW rights suggest that the courts, which hear cases on an individual basis, are not equipped to deal with issues that require a broader view. Moreover, the complex nature of social and economic issues bars the courts from truly assessing the whole environment from which the case stems and which is affected by the decision.26 Continuing the analysis of the judicial role with respect to Israeli health policy, Chinitz and Shalev note that ‘It is not surprising that the Court is hesitant to intervene in such technical and obscure fiscal matters . . . The Court is not necessarily well equipped to deal with accounting’.27 Moreover, even appeals to courts to overturn health fund decisions are ‘unlikely to succeed, unless the letter of the law has been ignored. The courts acknowledge budgetary constraints and accept standards of evidence-based medicine as benchmarks for public funding’.28 However, as the Maccabi case, described above, suggests, it is often the political decision that is not based on ‘evidence-based medicine’, while the judicial one may rely on substantial expert opinion (such as that of the Health Council, in that case). Even then, extreme judicial restraint is observed. Though courts should not be dismissive of the fact that the individual case they are addressing may have repercussions for others whose claim is not heard, it should also be recalled that courts deal, on a daily basis, with intricate economic problems that have serious ramifications in trade, business, monopolies, taxes, and similar realms. This is done, of course, with no qualms regarding competence.29 Moreover, an institutional analysis of legislatures and courts reveals no important differences regarding the pressures to which they are subject, the considerations they take into account and the level of generality they aim for.30 An analysis of post-Communist socio-legal developments suggests that even where constitutional courts intervened in a manner that moderated the transition to market economies, no ‘disaster’ has come to pass, despite warnings to that effect.31 Moreover, in some cases (such as Hungary) the transition may have been improved by the Constitutional Court’s oversight and regulation. The charge that the 24 ibid 750–51; see also D Chinitz and C Shalev, ‘Joe Public v The General Public: The Role of the Courts in Israeli Health Care Policy’ (2005) 33 Journal of Law, Medicine & Ethics 650, 654. 25 Maccabi (n 21) 752. 26 See Holmes and Sunstein (n 10) 18–19. 27 Chinitz and Shalev (n 24) 653. 28 ibid 655. 29 F Michelman, ‘In Pursuit of Constitutional Welfare Rights: One View of Rawls’ Theory of Justice’ (1973) 121 Pennsylvania Law Review 962, 1006. 30 J Deutsch, ‘Neutrality, Legitimacy and the Supreme Court: Some Intersections between Law and Political Science’ (1968) 20 Stanford Law Review 169, 183–84. 31 KL Scheppele, ‘A Realpolitik Defense of Social Rights’ (2004) 82 Texas Law Review 1921, 1948; Tushnet, Weak Courts, Strong Rights (n 5) 235–36.
338 Amir Paz-Fuchs court lacks expertise to deal with social and economic issues, therefore, seems to rely on shaky evidence. The conclusion should be that even where budgetary issues are concerned, the court’s particular expertise justifies expanding ‘the normal conception of the role of the courts in a democratic society to include the role of “policy partner” in ongoing bargaining about how a state should use its scarce resources’.32 A more nuanced rebuttal to this objection stems from the fact that not all reviews of social policies are alike. An important distinction should be made at the level of remedy, between demands that the court itself erects a policy or programme, on the one hand, and charges to administer the policy in a fair and equal fashion, on the other hand.33 An example of the latter case is the Canadian Government’s refusal to provide sign language interpreters to deaf patients on the basis that it would put a ‘severe strain on the fiscal sustainability of the health care system’.34 The Court reasoned that ‘The Legislature, upon defining its objective as guaranteeing access to a range of medical services, cannot evade its obligations . . . to provide those services without discrimination by appointing hospitals to carry out that objective’.35 The Court may well be hesitant to dictate the erection of a particular medical service, but once the Government decided to enter the field, ‘it must provide all the services within the genre’36 to all the relevant recipients. We find, once again, that the ability to link such SW challenges to the duty to treat all citizens with equal concern and respect increases their appeal.37 And yet, while this approach seems quite appealing, it also raises a particular difficulty: the fact that court decisions that give effect to SW rights are deemed justifiable only when a given policy is in effect suggests that it is the principle of non-discrimination, rather than social and economic rights, that is doing ‘the work’. The result is twofold: first, as a principled matter, rather than undermining the ‘liberal’ (in the European sense) tendencies of the courts, this approach reveals a willingness to entertain SW challenges only when they can be broken down to an individual’s (preferably – nameable individual’s) interests, but not when they are presented as collective harms, however ill-conceived, unprofessional and harmful to constitutional rights. Secondly, the concrete effects of the approach are noticeable, insofar as the remedy is concerned: instead of a generalised (or even generalizable) injunction, the court will solve a particular problem for a particular individual. In doing so, it will also praise the importance of SW rights, and mock the irrelevance of budgetary concerns, which will be relatively minimal in the case at hand. We find, then, that this approach does not only trace the limits of the ability to challenge social and economic policies in a successful manner (ie when discrimination is established). It also reinforces, rather than undermines, the traditional objection to the court’s ability to reach decisions that have complex, and perhaps unpredictable, consequences, an objection that is especially visible where SW rights are concerned. A few Israeli cases from the field of education seem particularly relevant: in Botzer, the ISC ruled that a school must make the necessary physical arrangements to grant a Scheppele (n 31) 1935. Davidov (n 11) 351. 34 Eldridge v British Columbia (Attorney General) [1997] 3 SCR 624 (Can). 35 ibid para 51; see Tushnet, Weak Courts, Strong Rights (n 5) 205 (asserting that, under Eldridge, once a government decides to provide some social welfare services, it must do so without discrimination). 36 ibid (n 5) 205. 37 See, eg D Barak-Erez and A Gross, ‘Social Citizenship: The Neglected Aspect of Israeli Constitutional Law’ in Barak-Erez and Gross, Exploring Social Rights (n 5) 243, 252–53; F Michelman, ‘Foreword: On Protecting the Poor through the Fourteenth Amendment’ (1969) 83 Harvard Law Review 7, 11. 32 33
The Fiscal Objection to Social Welfare Rights 339
child who suffers from multiple sclerosis and is bound to a wheelchair, easy access to and from his classroom.38 The Court did not miss the opportunity to assert that ‘ensuring equal opportunity . . . costs money’ and that the constitutional right to dignity includes the right to equal opportunities in education.39 Needless to say, the judgment did not require all schools in Israel to make similar modifications to allow children with disabilities to be integrated in the general school system. Indeed, the very same year the ISC rejected a petition that challenged a ministerial (and not statutory) budgetary decision to cut a programme that assists the children suffering from environmental dis advantage and helps them integrate, socially and intellectually, into the general school system.40 The Court distinguished its decision from Botzer, stating that ‘the principle of equal opportunity does not stand alone. It cannot be severed from the general social context. The realization of the principle requires resource allocation. The financial ability of the government authority should be balanced against the needs’.41 A similar fate was in store for a petition to force the Ministry to apply provisions of the Special Education Law 5748-1988 that require the integration of children with special needs in the general education system to children aged three to four.42 In a brief judgment, the Court acknowledged that the professional consensus is that the integration of children with disabilities enhances their social and intellectual ability, and that time is of critical importance.43 But it rejected the charge, stating that the limited budgetary abilities require prioritization, and that is left to the Government.44 Such budgetary considerations were left irrelevant when the ISC considered a demand that a municipality shoulder the costs of transferring a child who suffered severe social problems in her original school.45 The ISC relied on the constitutional right to education that should be supplied free of charge, including in cases where particular solutions are necessary. A pattern emerges: when a general social problem is laid down for the court, budgetary concerns are raised. When a very similar social issue (eg integration of children into schools) is broken down – the court is much more receptive. As Tushnet’s summary of Irish cases reveals, the same pattern is observed in other countries.46
IV. APPLICABILITY AND SPILLOVER EFFECTS
This argument suggests that, because of their budgetary implications, SW rights simply cannot be enforced. It would seem surprising to find a court stating such a rationale explicitly and, indeed, no such sentiment appears in ISC (or other court’s) judgments that I am aware of. However, the centrality of this argument as a wedge to distinguish
HCJ 7081/93 Botzer v Municipal Council of Maccabim-Reut 50(1) PD 19 [1996] (in Hebrew). ibid para 27. 40 HCJ 1554/95 Shoharei Gilat Society v Minister of Education 50(3) PD 2 [1996] (in Hebrew). 41 ibid para 41. 42 HCJ 5597/07 Alut – The National Association for Children with Autism v Minister of Education (21 August 2007), Nevo Legal Database (by subscription) (in Hebrew). 43 ibid para 4. 44 ibid para 6. 45 HCJ 7374/01 John Does v Director of the Ministry of Education (10 September 2003), Nevo Legal Database (by subscription) (in Hebrew). 46 Tushnet, ‘Social Welfare Rights’ (n 3) 1899–1900. 38 39
340 Amir Paz-Fuchs CP from SW rights in academic writings,47 and its strong relation to budgetary concerns, merits some discussion here. Frank Cross explains that courts ‘avoid involving themselves in matters fundamental to the enforcement of positive rights’.48 The argument is very similar to that of Maurice Cranston’s, almost 50 years ago.49 Though they do not state the argument explicitly in these terms, it can be understood from the thrust of Cranston’s and Cross’s argument, that the recognition of one type of right (social or economic) to which one cannot give effect may lead to a derogation of the status of rights in general and to the state’s commitment to the protection of rights. Cass Sunstein articulated this concern, stating that [i]f positive rights are not enforceable, the constitution itself may seem like a mere piece of paper; there could be adverse consequences for other rights as well . . . the existence of unenforceable rights will in turn tend to destroy the negative rights – freedom of speech, freedom of religion, and so forth – that might otherwise be genuine ones. If some rights are shown to be unenforceable, it is likely that other rights will be unenforceable as well.50
But this worry seems to rest on the fallacious assumption that if a right is recognised, it should always trump other interests. However, there is no reason to assume that advocates of the right assume that it is an absolute one, or that the state is under a duty to invest all its resources in one right or another. As Joseph Raz notes, the tendency to portray support for certain rights as espousing a position in favour of absolute rights amounts to a ‘simple mistake’ which is more common than what would be expected: The fact that a given right can be overridden by moral considerations, just like the fact that it can be overridden by another legal right, shows nothing except that it is not an absolute right which defeats all contrary considerations.51
Of course, if a full realization of the right to health and education would require that everyone be able to enjoy free comprehensive health services on demand and free education to her heart’s content, it may well be that no country will ever have the financial capabilities necessary to hold up to such a standard. This does not imply, however, that within the domestic legal system these rights carry no weight, and surely does not demand the conclusion that they are not rights at all. Despite attempts to dismiss rights that correspond to ‘imperfect obligations’ (used here to refer to obligations that cannot be addressed in full),52 the feasibility of enforcement of rights is not a criterion (or at least not an overwhelming criterion) when dealing with the question of recognition of rights. Since rights cost money, the decision to realise civil and political rights by policing political demonstrations that people find offensive, for example, may divert funds 47 See, eg Y Dotan, ‘The Supreme Court as Defender of Social Rights’ in Y Rabin and Y Shani (eds), Economic, Social and Cultural Rights in Israel (Tel-Aviv, Ramot, 2004) 69, 88–97. 48 Cross, ‘The Error of Positive Rights’ (2001) (n 13) 889, 904–05. 49 M Cranston, ‘Human Rights, Real and Supposed’ in DD Raphael (ed), Political Theory and the Rights of Man (Indiana, Indiana University Press, 1967) 50. 50 C Sunstein, ‘Against Positive Rights’ in A Sajo (ed), Western Rights, Postcommunist Application (Hogue, Kluwer Law International, 1996) 225, 229. For a similar argument, see LW Sumner, The Moral Foundations of Rights (Oxford, Clarendon Press, 1989) 15; R Epstein, ‘Classical Liberalism Meets the New Constitutional Order: A Comment on Mark Tushnet’ (2002) 3 Chicago Journal of International Law 455, 464. 51 J Raz, ‘Legal Rights’ (1984) 4 OJLS 1, 19; Holmes and Sunstein also acknowledge this truism (n 10) 97. 52 R Epstein, ‘The Uncertain Quest for Welfare Rights’ (1985) Brigham Young University Law Review 201, 204.
The Fiscal Objection to Social Welfare Rights 341
away from building hospitals. Such a decision, in other words, is a political and not a conceptual one.53 V. THE FISCAL OBJECTION REVEALED AND DISCUSSED
In the discussion above, I briefly addressed each of the objections noted above in their new, ‘fiscal’, dressing. However, as noted, all these arguments collapse, in their contemporary versions, into objections that have money at the centre of their interest.54 Further evidence that the arguments discussed are strongly linked to fiscal constraints in contemporary discourse is revealed in the negative. In the (justifiably) celebrated South African Treatment Action Campaign (TAC) case,55 the Constitutional Court ordered the Government to distribute a drug, Nevirapine, that reduces the transmission of HIV/AIDS from pregnant mothers to children. The interesting point, for present purposes, is that the drug’s manufacturer was willing to supply as much as was needed at no cost. TAC presented a complex case that required proper professional and bureaucratic expertise, where it could be expected that bona fide questions of competence could be raised. This matter was indeed acknowledged by the South African Constitutional Court.56 And yet, instead of deferring to the Government’s discretion, the Constitutional Court addressed and dismissed the relevant arguments (efficacy, safety, biomedical resistance, cultural reservations to bottle feeding, absence of clean water) one by one.57 Similarly, the Constitutional Court rejected the Government’s position that it is not in a position to distribute the drug across the country as ‘not relevant to the question’ at hand.58 The reason for such an outright rejection of the Government’s arguments in such a complex case seems clear. Where the fiscal impact of compliance is small, the normative objections seem to wither away.59 What made the decision ‘a relatively easy one for the Court’, argues Denis Davis, was the lack of ‘any sustainable argument concerning distributional questions’.60 Indeed, when the ISC discussed a parallel case, the ISC distinguished the matter from TAC by noting (and emphasizing) that, in the latter case, ‘the drug was distributed free of charge’.61 It makes sense, therefore, to focus on the argument that the fact that enforcement of SW rights requires public funds is, in and of itself, sufficient to bar them from judicial enforcement. The surprisingly trivial refutation of this argument is that while SW rights do indeed require funds, they are not unique in doing so.62 However, where CP rights (including R Dworkin, Sovereign Virtue (London, Harvard University Press, 2000) 137. J Kuthari, ‘Social Rights Litigation in India’ in Barak-Erez and Gross, Exploring Social Rights (n 5) 171, 174. 55 Minister of Health v Treatment Action Campaign 2002 (10) BCLR 1075 (CC) (S Afr). 56 ibid para 38. 57 ibid paras 56–64. 58 ibid paras 65–66. 59 See also, Tushnet, Weak Courts, Strong Rights (n 5) 247. 60 D Davis, ‘Socio-Economic Rights: The Promise and Limitations: The South African Experience’ in BarakErez and Gross, Exploring Social Rights (n 5) 193, 199. 61 HCJ 3071/05 Louzon v Government of Israel (28 July 2008), Nevo Legal Database (by subscription) para 12 (in Hebrew). 62 See, eg Michelman, ‘The Constitution’ (n 6) 24; Davidov (n 11) 354–55. 53 54
342 Amir Paz-Fuchs equality) are concerned, the courts are quite impatient with such arguments. Thus, the Canadian Supreme Court accepted a claim that a benefit policy was discriminatory, and was undeterred by the fiscal implications, noting that ‘any remedy granted by a court will have some budgetary repercussions whether it be a saving of money or an expenditure of money’.63 And in Singh,64 the same Court rejected the Government’s position that refugees can be denied an oral hearing because of the unreasonable budgetary burden that granting such a right would entail.65 Similarly, the South African Constitutional Court ruled that prisoners may not be denied the right to vote, even though granting them such rights may have serious logistic and financial implications.66 In Israel, the Supreme Court stated that it is ‘natural’ that implementing equality (in this case – to women in the air force) would have financial consequences.67 In fact, where equality is concerned, the ISC repeated the simple statement that ‘human rights cost money’ on several occasions: when accepting the charge that forcing airline cabin attendants to retire at 60 constitutes age discrimination;68 when accepting the claim that different early retirement arrangements for men and women cannot be justified by budgetary concerns;69 when stating that such concerns cannot justify discrimination between groups entitled to benefits and subsidies;70 when ordering a deeper drill for a pipe line to avoid harming graves, at the cost of 500,000 NIS (New Israeli Shekel);71 and when striking down a law that permits holding accused soldiers for extensive periods before seeing a judge.72 As trivial as this seems, a testament to the power of the original argument discussed in this part is the fact that two prominent constitutional scholars, Stephen Holmes and Cass Sunstein, dedicated a book to refuting it. The main argument in The Cost of Rights73 is clearly stated in the introduction: To the obvious truth that rights depend on government must be added a logical corollary, one rich with implications: rights cost money . . . The right to freedom of contract has public costs no less than the right to health care, the right to freedom of speech no less than the right to decent housing. All rights make claims upon the public treasury.74
Contemporary antagonists of SW rights have understood the force of this reply, and have revised their position, arguing that the distinction between SW rights and tradi Schachter v Canada [1992] 2 SCR 679, 709 (Can). Singh v Minister of Employment and Immigration [1985] 1 SCR 177 (Can). 65 ibid para 70. 66 Minister of Home Affairs v NICRO 2004 (5) BCLR 445 (CC) para 48 (S Afr) (‘Resources cannot be ignored in assessing whether reasonable arrangements have been made for enabling citizens to vote. There is a difference, however, between a decision by Parliament or the Commission as to what is reasonable in that regard and legislation that effectively disenfranchises a category of citizens’). 67 HCJ 4541/94 Miller v Minister of Defence 49(4) PD 94, 142 [1995] (in Hebrew). 68 HCJFH 4191/97 Recanat v National Labour Court 54(5) PD 330, 366 [2000] (in Hebrew). 69 HCJ 6845/00 Niv v National Labour Court 56(6) PD 663, 697 [2002] (in Hebrew) (‘we find it difficult to understand how budgetary considerations could justify discrimination of women’). 70 HCJ 5496/97 Mardi v Minister of Agriculture 55(4) PD 540, 568 [2001] (in Hebrew) (even when the budget for the relevant needs has been spent in its entirety, ‘remedying discrimination . . . might cost money . . . [and there is] an obligation to find the way to pay the necessary sums’). 71 HCJ 4638/07 Al-Aktza Almobarak Company Ltd v Israel Electric Co (29 October 2007), Nevo Legal Database (by subscription) para 11 (in Hebrew). 72 HCJ 6055/95 Zemach v Minister of Defence 53(5) PD 241, 281 [1997] (in Hebrew) (‘defending human rights often has costs’). 73 Holmes and Sunstein (n 10). 74 ibid 15. 63 64
The Fiscal Objection to Social Welfare Rights 343
tional, CP rights, is one of quantity, not quality. In other words, it is indeed true that both ‘types’ cost money, but the degree of SW expenditure is of such a different magnitude that it demands a conceptually different treatment.75 This position, positively noted by the ISC,76 suggests that, because SW rights require massive distributions of funds, they raise questions of competence and democratic legitimacy in a way that minor expenditures that result from decisions concerning CP rights do not. There are two different responses to this renewed version of the traditional argument. The first relates to the above discussion, concerning the importance of the remedy sought when assessing the legal challenge.77 The reason that the remedy comes to the fore is that, since almost all modern nations have in place governmental health care, education, subsistence support and similar programmes, demands that the court institute a new social programme ‘from scratch’ are extremely rare.78 Much more common – and potentially costly – are demands that existing social policies should not discriminate against defined groups (women, single parents, single sex couples, immigrants, etc). And yet, even those advancing the objection under review do not suggest that the courts should refrain from scrutinizing claims of discrimination, even against the background of SW policy.79 A second response to this objection takes it at face value: even if there is a clear way of distinguishing CP claims from SW claims (as was made clear, at least equality blurs the line), the argument relies on empirical evidence that is not offered. In fact, there may well be evidence to the contrary. The Costs of Rights, in fact, includes an appendix entitled ‘Some Numbers on Rights and Their Costs’,80 exemplifying the astounding cost of CP rights enforcement in the US. For example, the Federal prison system costs $2.465 billion; the Federal court system costs almost $1.5 billion; the joint cost of animal and plan inspection and food and safety inspection is close to $1 billion; Defence Department obligations exceed $20 billion; environmental protection (clear air, hazardous waste, natural resources, and water quality) costs over $1.3 billion; and Census Bureau activities, government publications, postal services and national archives cost over $500 million. In Israel, several cases can be noted to make the same point. While finding that the ‘Gaza Disengagement plan’ is constitutional, the ISC struck down four compensation mechanisms to (former) settlers in the region.81 A senior Ministry of Finance official estimated the cost of the judgment at 500 million NIS, raising the total cost to 7.5 billion NIS,82 while a subsequent report issued by the Knesset (Israeli Parliament) Research and Information Center assessed the cost of the judgment at 10.8 billion NIS (about $3 billion).83 Following the Disengagement, the ISC did not hesitate ordering the Government to 75 Tushnet, ‘Social Welfare Rights’ (n 4) 1896; M Cohen-Eliya, ‘The Blues, the Reds, and the Judge in Between: A Procedural Approach with Bites for Judicial Review on Social Rights’ (2009) 12 Law and Government 407, 410–15, 420–21 (in Hebrew). 76 Louzon (n 61) para 10. 77 Text to nn 33–37. 78 Schwartz, ‘Economic and Social Rights’ (n 4) 1238. 79 Cohen-Eliya (n 75) 429. 80 Holmes and Sunstein (n 10) 233–36. 81 HCJ 1661/05 Hof Aza Regional Council v Knesset 59(2) PD 481 [2005] (in Hebrew). 82 Z Klein, ‘Haber: The HCJ Judgment will increase the cost of the Disengagement Plan by 500 million NIS’, Globes, Rishon-Lezion, 16 June 2005 (in Hebrew) www.globes.co.il/news/article.aspx?did=924114. 83 I Biton, ‘The Disengagement Plan from the Gaza Strip and Areas in Northern Samaria’ Knesset Research and Information Center (31 January 2010) (in Hebrew), available at: www.knesset.gov.il/mmm/data/pdf/ m02539.pdf.
344 Amir Paz-Fuchs complete the necessary defence for classrooms in the Gaza vicinity, despite warnings that the costs would reach ‘hundreds of millions of NIS in the short run, and probably several billion NIS in the long run’.84 The overall cost of the ISC’s orders to dismantle (and re-erect) certain segments of the separation barrier85 is estimated at 1.5 billion NIS ($416 million).86 The ISC’s unprecedented decision to strike down a law that would authorise the establishment of a private prison87 has been estimated to cost the tax payer up to 637 million NIS ($177 million).88 Another case merits particular mention, as it combines exceptional, clear financial consequences with the Court’s refusal to give them any consideration. In addition, the case manifests the Court’s complete disregard of the Government’s legitimate priorities and rationales. In the case, Dan Area Revenue Service v Perry,89 the ISC denied an appeal on the Tel-Aviv District Court’s decision, according to which child care should be recognised as exemptions for tax purposes. The decision was made despite the Ministry of Finance’s assertion that it would cost the public treasury over 3 billion NIS ($900 million) per year.90 Where the tax interests of successful lawyers are concerned, it would seem, fiscal concerns lose their appeal. When SW claims appear before the Court, particularly in their ‘naked’ form which highlights the fiscal aspect, the Court’s worst fears are met with an unmitigated reply. In Manor91 and in Commitment to Peace92 claims were made that social benefit cuts (to old age pensions and to subsistence benefits, respectively) infringe an impoverished individual’s ability to a life of dignity and undermine the right to social security.93 In light of the above, the fact that the Court ruled against the claimants is not surprising. The surprise, or disappointment, stems from ISC’s logic, which is a testament to its complete deference to government fiscal policy. The Israeli constitutional method that has developed over the past two decades involves a three-stage analysis.94 First, the court assesses whether a constitutional right has been violated. If so, it continues to assess whether the infringement is by law, proportionate to the benefit, and cannot be achieved by less harmful means. The third stage involves a discussion of the constitutional remedy, if necessary. In the vast majority of cases, the court is very lenient insofar as the first stage is concerned, and focuses its attention on the second stage. The court, in other words, is ready to recognise an infringement of constitutional right even when it is patently clear that the infringement is justified, and marginal when compared to the beneficial objective HCJ 8397/06 Wasser v Minister of Defence 62(2) PD 198 [2007] para 8 (in Hebrew). HCJ 2056/04 Beit Sourik Village Council v Government of Israel 58(5) PD 807 [2004] (in Hebrew); HCJ 7957/04 Mara’abe v Prime Minister of Israel 60(2) PD 477 [2005] (in Hebrew). 86 S Arieli and M Sfard, The Wall of Folly (Tel-Aviv, Yedi’ot Sfarim, 2008) 225. 87 HCJ 2605/05 Academic Center of Law and Business v Minister of Finance (19 November 2009), Nevo Legal Database (by subscription) (in Hebrew). 88 B Arad, The Cost of Judicial Activism (Jerusalem, Jerusalem Center for Market Research, 2010). 89 CA 4243/08 Dan Area Revenue Service v Perry (30 April 2009), Nevo Legal Database (by subscription) (in Hebrew). 90 H Kanna, ‘Tax claims worth billions are swamping the courts’, Calcalist, Tel-Aviv, 29 June 2009 (in Hebrew) www.calcalist.co.il/local/articles/0,7340,L-3275640,00.html. The judgment was overturned by statute. 91 HCJ 5578/02 Manor v Minister of Finance 59(1) PD 729 [2004] (in Hebrew). 92 HCJ 366/03 Commitment to Peace and Social Justice Association v Minister of Finance 60(3) PD 464 [2005] (in Hebrew). 93 See Barak-Erez and Gross, ‘Social Citizenship’ (n 37) 250–52. 94 CA 6821/93 United Mizrahi Bank Ltd v Migdal Cooperative Village 49(4) PD 221, 428 [1995] (in Hebrew). 84 85
The Fiscal Objection to Social Welfare Rights 345
that the law in question achieves.95 The sizable majority of petitions are denied, therefore, only after the proportionality of the violation is assessed, along with a cost-benefit analysis and an enquiry to see if the objective can be achieved at a lesser harm to the right. The ISC’s analysis in Manor and Commitment to Peace was quite different. Despite the significant cut in benefits (in some cases – up to 30 per cent), the Court was unwilling to rule that the laws in question infringed the constitutionally recognised (at least in rhetoric) right to social security at all.96 We find, then, that while the Court expands its understanding of the scope of CP rights and, consequently, the occasions where a violation is exhibited, its construction of SW rights is much more narrow, and its willingness to acknowledge a violation – much more limited. A very similar path was taken, in a series of cases, by the European Court of Human Rights, which struck down claims for subsistence benefits at the preliminary stage. Even though the claimant lacked any independent means of support, the Court ruled that the claim is ‘manifestly ill-founded’ since it was not clear that there were no alternative support systems.97 Similarly, the Canadian Supreme Court rejected a claim to equalise general assistance benefits to young adults (under 30), stating that there is no legal support to the claim that a reduced benefit (one-third) constitutes a substantive constitutional harm.98 Finally, an issue that posits the tension rights and money in their pure form has the potential to occupy the courts time and time again in the future. The matter, which may be termed colloquially as ‘chipping in’, concerns the conditioning of rights on payment. Here again we find a very different judicial approach where CP and SW are assessed. In the case of Fordyce County99 the US Supreme Court ruled that the Government cannot charge the immediate users of freedom of speech, such as protesters in a public park, for the expenses for speech related activities. And in a very similar Israeli case, the ISC recognised the police’s power to subject a protest license to certain conditions that are relevant to matters of public order, but ruled that ‘setting a price for the implementation of a right means violating the rights of those who cannot afford it’.100 But this sentiment is somewhat more attenuated when SW issues are concerned. As emerges from Eva Brems’s portrayal of several copayment cases in Europe, the courts are unwilling to allow even a preliminary judicial review of the position that substantial copayments violate the right to health and to equality.101 Similarly, in Israel, a woman who required surgery so as not to lose her hearing was asked to pay 70 per cent of the cost.102 Being a teacher and a single parent, she could not afford the ‘bill’, and 95 HCJ 9333/03 Kaniel v Government of Israel (16 May 2005), Nevo Legal Database (by subscription) (in Hebrew) (any new tax infringes on the right to property); HCJ 5026/04 Design 22 Shark Deluxe Furniture Ltd v Rosenzweig 60(1) PD 38 [2005] (in Hebrew) (law dictating a mandatory day of rest bears on the constitutional right to freedom of occupation). 96 Commitment to Peace (n 92) para 20. 97 E Brems, ‘Indirect Protection of Social Rights by the European Court of Human Rights’ in Barak-Erez and Gross, Exploring Social Rights (n 5) 135, 155–56. 98 Gosselin v Quebec [2002] 4 SCR 429 (Can); P Macklem, ‘Social Rights in Canada’ in Barak-Erez and Gross, Exploring Social Rights (n 5) 213, 237–38. 99 Fordyce County, Georgia v Nationalist Movement 505 US 123 (1992). 100 HCJ 2557/05 Majority Camp v Israel Police 62(1) PD 200 [2006] para 16 (in Hebrew). 101 Brems (n 97) 141. 102 HCJ 2974/06 Israeli v Committee for the Expansion of the Health Basket (11 June 2006), Nevo Legal Database (by subscription) para 28 (in Hebrew).
346 Amir Paz-Fuchs challenged the onerous copayment requirement. In rhetoric that is strikingly different from the one employed in the free speech cases, the ISC stated that when priorities are to be set, the judiciary should defer its judgment to professional committees, such as the one that deliberated the question in this case.103 In conclusion, it would be worthwhile to refer to Ruth Gavison’s suggestion that the Court’s most marked deference to governmental priorities is visible in cases where ‘human and social interaction’104 is necessary. Contrariwise, we find that such deference is actually most notable in social policy cases where the budgetary requirements are pure and clear. VI. HOW DID IT COME TO THIS? RIGHTS AS SOCIAL BARGAINS
The discussion to this point has shown how objections to SW rights, proven unsuccessful on their own terms, have been transformed to fiscal arguments, and have been remarkably triumphant in doing so. Moreover, as the preceding part has shown, this resurgence has taken place despite the fact that arguments focused on budgetary matters fail to distinguish convincingly between SW matters and traditional rights litigation. I end this chapter with the suggestion that this specific transformation, and its success, is not coincidental. In fact, the fiscal redressing of the arguments against SW rights is part of a larger phenomenon, one that has deeper meanings and wider consequences. Legal discourse is in the process of restructuring rights as ‘conditional’105 or, in the terminology suggested by Sunstein and Holmes (importantly – in the closing sections of a book arguing against fiscal objections to SW rights) – as ‘social bargains’. These are the makings of ‘consumer-citizenship’,106 one that is ‘reconceived in terms of consumption and participation in markets’.107 The importance of this trend has been noted: if the welfare state was once understood to reflect ‘the subordination of market price to social justice, the replacement of the free bargain by the declaration of rights’,108 it is now urged that ‘that trajectory has been reversed’.109 What are the jurisprudential consequences of the central role that market discourse has captured in policymaking? Borrowing a phrase from a different jurisprudential debate, it may be suggested that the fiscal paradigm has the potential to lead to a situation whereby ‘rights become illusory’ since they would ‘extend only up to the point where our actions ceased to make a net contribution to the collective project’.110 103 For a detailed discussion of the case and its aftermath see Barak-Erez and Gross, ‘Social Citizenship’ (n 37) 253–54; A Gross, ‘The Right to Health in an Era of Privatization and Globalization’ in Barak-Erez and Gross, Exploring Social Rights (n 5) 289, 323. 104 R Gavison, ‘On the Relationship Between Civil and Political Rights and Social and Economic Rights’ in JM Coicaud, MW Doyle and AM Gardner (eds), The Globalization of Human Rights (United Nations University Press, 2003) 23, 37. 105 A Paz-Fuchs, Welfare to Work: Conditional Rights in Social Policy (Oxford, Oxford University Press) 53–64. 106 N Harris, ‘The Welfare State, Social Security and Social Citizenship Rights’ in N Harris (ed), Social Security Law in Context (Oxford, Oxford University Press, 2000) 3, 27. 107 K Rittich, ‘Social Rights and Social Policy’ in Barak-Erez and Gross, Exploring Social Rights (n 5) 107, 111; similarly, JP Euben, ‘Walzer’s Obligation’ (1972) 1 Philosophy and Public Affairs 438, 444. 108 TH Marshall, ‘Citizenship and Social Class’ (1963) 115. 109 M Katz, The Price of Citizenship: Redefining the American Welfare State (Pennsylvania, University of Pennsylvania, 2008) 1. 110 N Simmonds, ‘Rights at the Cutting Edge’ in M Kramer, N Simmonds and H Steiner (eds), A Debate Over Rights:Philosophical Enquiries (Oxford, Clarendon Press, 1998) 113, 145.
The Fiscal Objection to Social Welfare Rights 347
Curiously, while the above quote was meant to criticise a dominant theory of rights,111 it is almost an exact translation of some ISC (and other court) pronouncements dealing with SW rights. Thus, in Manor, the Court stated that the objective of ‘healing the market’ is ‘important to preserve the social structure, which in turn protects human rights’.112 With respect to the rights of manpower workers, Court President Barak stated that ‘it is worth harming human rights to preserve a social structure that protects human rights’.113 The relevance to our current discussion is made explicit by the fact that the ISC supported the above statement with a reference to another case – Kontram v Ministry of Treasury114 – which, somewhat surprisingly, ‘is considered an important contribution to the emergence of social rights in the [Israeli] Supreme Court’.115 The case involved an administrative matter of licensing, quite unrelated to social rights. Zamir J does suggest that ‘human rights are, indeed, of supreme importance’.116 But he continues: But that is just one role [of the government] . . . In fulfilling our role, each of us must accept a system of responsibilities, not only towards other individuals, but also towards society in general . . . That is how I perceive the social contract.117
The quote is important, I believe, because the language placing ‘rights besides duties’; ‘no rights without responsibilities’;118 and rights as ‘social bargains’ or ‘community assets’119 forms and informs the judicial manifestation of the social contract version that is currently in vogue.120 In particular, positing rights against general welfare, and rights against social justice, allows even those who express support for SW rights to subject them to general interest. The following quote, from Court President Barak, is part of that particular agenda: ‘the normative status and scope of the right to social security . . . is derived from the nature of the economic and social regime that governs a certain society . . . It expresses the economic strength of its market’.121 It would seem difficult to offer a better example to support Simmonds’ warning that rights may end up serving merely as place holders for forms of protection, that are ‘balanced against countervailing considerations, so that the protections that finally result are the outcome of this calculus of conflicting reasons’.122 The natural result, per Simmonds and others, is that ‘rights or the interests of the rightholders would play no strategic role in moral and political affairs’.123 While this proposition seems somewhat extreme as a general assessment, it has merit insofar as SW rights discourse in Israel (and elsewhere) is concerned. 111 I refer to the ‘interest theory of rights’. See, eg J Raz, Ethics in the Public Domain: Essays in the Morality of Law and Politics (Oxford, Clarendon Press, 1995) 44–55. 112 Manor (n 107) 740. The passage was quoted approvingly in HCJ 4947/03 Be’er Sheva Municipality v Government of Israel (10 May 2006), Nevo Legal Database (by subscription) para 11 (in Hebrew) (confirming a law abolishing tax benefits to cities in the south of Israel). 113 HCJ 450/97 Tnufa v Minister of Labour and Welfare 52(2) PD 433, 441 [1998] (in Hebrew). 114 HCJ 164/97 Kontram v Ministry of Finance 52(1) PD 289 [1998] (in Hebrew). 115 Barak-Erez and Gross, ‘Social Citizenship’ (n 37) 247. 116 Kontram (n 114) 340. 117 ibid. 118 A Giddens, The Third Way and its Critics (Cambridge, Polity Press, 1998) 65–66; A Gewirth, The Community of Rights (Chicago, University of Chicago Press, 1996) 42, 223, 231–35. 119 Holmes and Sunstein (n 10) 217. 120 See, eg Michelman, ‘The Constitution’ (n 6) 26–30. 121 HCJ 494/03 Physicians for Human Rights v Minister of Finance 59(3) PD 322, 333–34 [2004] (in Hebrew). 122 Simmonds (n 110) 160. 123 J Chan, ‘Raz on Liberal Rights and Common Goods’ (1995) 15 OJLS 15, 29.
348 Amir Paz-Fuchs The important point here is the conceptual one. It is expressed in the ISC’s succinct statement that, in some cases, ‘where there is no budget, there is no right’.124 It should be made clear that the Court does not express the obvious truth that budgetary considerations may limit the implementation of the right. Rather, it suggests that they are, as Dennis Davis argues when discussing South African cases, ‘in effect . . . defined in terms of availability of resources’.125 He refers to Chaskalson P’s conclusion, according to which: ‘access to housing, health care, food, water and social security are dependent upon the resources available for such purposes, and that the corresponding rights themselves are limited by reason of the lack of resources’.126 VII. CONCLUSION
Despite assertions to the contrary, objections to SW rights have not disappeared. In fact, the current divide between SW rights and traditional, CP rights, seems wider than ever. The explanation for this curious situation, this chapter argued, lies in the fact that while the original objections were found unconvincing on their face, their contemporary versions capture the sign of the times: the market, consumer–citizenship paradigm. Costs, budgets and economic criteria are naturally central to this paradigm and thus re-enforce arguments that were on the verge of extinction. But history had a different idea in mind, and not for the first time. Over 200 years ago, Jeremy Bentham wrote of the social contract: ‘I was in hope . . . that this chimera had been effectively demolished by Mr Hume. I think we hear not so much of it now as formerly’.127 So now, insofar as paradigms supporting social policy are concerned, we hear of little else other than the social contract. And as the previous part of this chapter hypothesised, this paradigm, along with its derivatives, are playing a role in forming the new challenge for advocates of SW rights.
Louzon (n 61) para 16. Davis (n 60) 197; also Gross (n 105) 316; D Davis, P Macklem and G Mundlak, ‘Social Rights, Social Citizenship and Transformative Constitutionalism: A Comparative Assessment’ in J Conaghan, M Fischl and K Klare (eds), Labour Law in an Era of Globalization: Transformative Practices and Possibilities (Oxford, Oxford University Press, 2002) 511, 523–24. 126 Soobramoney v Minister of KwaZalu Natal 1997(12) BCLR 1696 (CC) 8, para 11. 127 J Bentham, ‘A Fragment on Government or a Comment on the Commentaries’ in JH Burns and HLA Hart (eds), Collected Works of Jeremy Bentham (London, Athlone Press, 1977) 393, 439. 124 125
23 Constitutional Review of ‘Eligibility Conditions’ in Social Rights Litigation NETA ZIV
I. INTRODUCTION
I
N FEBRUARY 2012 the Israeli Supreme Court handed down a decision – Hassan v National Insurance Institute1 – invalidating section 9A(b) of Israel’s Income Supplement Law.2 It was the first time the Supreme Court had struck down a law due to its infringement of a social right – in this case the right to human subsistence, derived from the basic right to human dignity. The unconstitutional section defined an eligibility condition for welfare as lack of ownership or regular use of a car. The Court decided that the requirements within this condition constituted a violation of the basic right to human subsistence, as they did not meet the proportionality ‘means–end’ test, required from a law violating a basic human right. Judicial review of eligibility conditions, common to social rights legislation, is the focus of this chapter. The broader context of this ruling was the constitutional standard for review of social rights. Prior to the Hassan decision the state had argued that due to their positive nature and their budgetary implications on social policy, judicial review of the scope and content of social rights ought to be performed primarily at the first stage of analysis, that is, when we enquire whether a violation of a basic social right had occurred. It also asserted that the burden of persuading that a violation had occurred – which lies upon the law’s challengers – is heavy and could be met only if the norm under review bears no relationship to the purpose of the legislation. In the Hassan case the state extended this argument to the review of eligibility conditions within social rights legislation. However, the Court rejected this proposition, ruling that the burden imposed on challengers of a social right is not heavier than in cases of civil and political rights. Accordingly, review of the eligibility conditions will be performed at the second stage of constitutional analysis, following a determination that a social right had been infringed upon. 1 HCJ 10662/04, 3282/05, 7804/05 Hassan v National Insurance Institute (28 February 2012), Nevo Legal Database (by subscription) (in Hebrew). The author took part in this litigation and represented petitioners in one of the cases, HCJ 3282/05 Merav Bin-Nun and others. Part of the documents upon which analysis in this chapter is conducted was obtained by the author during the course of representation. 2 Income Supplement Law, 5741-1980 (hereinafter: Income Supplement Law). S 9A added to this law in: Income Supplement Law (Amendment No 15), 5761-2001, SH No 1772, p 122.
350 Neta Ziv The goal of this chapter is to demarcate the boundaries of this ruling by analyzing the nature of eligibility conditions within social rights legislation and to assess its potential impact on constitutional protection of social rights. Part II reviews the connection between social rights and eligibility conditions from three perspectives: social policy and theory, human rights and constitutionality. Part III distinguishes between substantive eligibility conditions and verification methods and then discusses two other types of conditions: citizenship and residency requirements, and discriminatory conditions. Although they too impose qualifications that limit entitlement, I differentiate between them and direct eligibility conditions, which are analyzed in the remaining part of this chapter. Part IV describes the development of judicial review of social rights in Israel and part V focuses on judicial review of eligibility conditions, by analyzing the Hassan case. I conclude by assessing the impact of this decision on future constitutional litigation of social rights. II. SOCIAL RIGHTS AND ELIGIBILITY CONDITIONS
The literature discussing social rights as human rights attempts to delineate their uniqueness and underscore the commonalities between social-economic rights and civil and political rights.3 Despite the growing recognition of the inalienability of all human rights, there are characteristics that epitomise social rights. Rabin and Shani emphasise the elements typifying most social rights as being:4 (a) dominant positive characteristics – social rights often involve considerable obligations of public provision, which may entail significant costs;5 (b) non-affluent constituency: social rights recipients are often less well-off individuals who cannot obtain these resources otherwise;6 (c) historical context – social rights have developed within the context of the welfare state.7 Notwithstanding the recognition of social rights as human rights, entitlement to social resources is oftentimes selective, rather than universal. By universal I mean the right of every person – without individual determination of particular circumstances – to obtain certain resources. Such as, for example, the right to education and the right to receive emergency medical services in most countries, including Israel.8 In Israel child allowances and old age allowances are also universal, and do not require means testing or other eligibility conditions.9 3 On the nature of social rights as human rights and their constitutionality, see D Barak-Erez and A Gross, ‘Introduction: Do We Need Social Rights? Questions in the Era of Globalization, Privatization, and the Diminished Welfare State’ in D Barak-Erez and A Gross (eds), Exploring Social Rights: Between Theory and Practice (Oxford, Hart Publishing, 2007) 1. 4 Y Rabin and Y Shani, ‘The Case for Judicial Review over Social Rights: Israeli Perspectives’ (2008) 4 Israeli Affairs 681. 5 Obviously this distinction is not accurate as many civil and political rights are costly in their implementation. On this point, see S Holmes and CR Sunstein, The Cost of Rights: Why Liberty Depends on Taxes (New York, WW Norton, 1999) 35, 39, 44–48. 6 See generally, D Bilchitz, Poverty and Fundamental Rights: The Justification and Enforcement of SocioEconomic Rights (Oxford, Oxford University Press, 2007). 7 See also, J Tweedy and A Hunt, ‘The Future of the Welfare State and Social Rights: Reflections on Habermas’ (1994) 21 Journal of Law and Society 288, 288–90. 8 Mandatory Education Law, 5709-1949, ss 1–2. See also, Charter of Fundamental Rights of the European Union (18 December 2000) [2000] OJ C364/1, Art 31(1): ‘Every worker has the right to working conditions which respect his or her health, safety and dignity’. 9 AC Lewin and H Stier, ‘Who Benefits the Most? The Unequal Allocation of Transfers in the Israeli Welfare State’ (2002) 83 Social Science Quarterly 488.
‘Eligibility Conditions’ in Social Rights Litigation 351
Indeed, states can fulfill social rights obligations in more than one way. Legislation may spell out particular conditions one must meet to become eligible for ‘social rights’ in the sense of direct public provision.10 For example, a statute may determine that welfare entitlement is conditional upon the recipient’s acceptance of work offered to her by the state; regulation may condition public housing assistance upon the recipient never having owned a dwelling; a law may state that a welfare recipient using a car is not eligible for public assistance. In cases of this sort entitlement is not universal; one needs to meet certain criteria to become eligible for public provision. Even countries that have a strong universal system may impose eligibility criteria in the form of means testing in some contexts.11 This type of conditioning is rarely found in civil and political rights; their application is for the most part universal and not conditional. As will be discussed below, this difference between the two types of rights bears upon their constitutional analysis. My concern is when these kinds of qualifications appear problematic: what if the job offered to the welfare recipient stands in contrast to her religious beliefs? What if the person with no housing resources had lost his home due to a natural disaster? What if the vehicle is lent to the welfare recipient by a family member so she can take her child to school? How should we assess the legitimacy of such requirements? Eligibility conditions of this sort can be discussed from different angles. In the following sections I will briefly address the topic from the perspectives of social theory and public policy, human rights and constitutionality. A. Social Policy As a matter of public social policy states may choose to implement a universal social system, in all, or some areas of life such as welfare, employment, education or health. Accordingly, each person would be entitled to receive benefits regardless of her personal behaviour, condition or means. But many countries, including Israel, prefer selective arrangements in some contexts, and institute entitlement upon fulfilling certain conditions. Eligibility conditions are geared to ensure resources are allocated in accordance with the rationale underlying provision. Hence if the public purpose is to ascertain that the neediest will receive resources, means testing will be applied;12 if it is to encourage participation in the workforce, behavioural requirements such as work-oriented activities (‘workfare’) will be established;13 if the goal is to prevent waste of a scarce resource 10 W Korpi and J Palme, ‘The Paradox of Redistribution and Strategies of Equality: Welfare State Institutions, Inequality, and Poverty in the Western Countries’ (1998) 63 American Sociological Review 661. The authors studied different systems adopted by capitalist democracies and their distributive affects. Their claim is that particular institutional factors – rather than meta-approach to social welfare – are extremely significant in leading to different distributive outcomes. 11 A Doron and J Gal, ‘The Israeli Income Support Program in a Comparative Perspective’ (2000) 58 Social Security 5, 6–7 (in Hebrew). 12 Since one of the justifications for economic and social human rights is human need, it is arguable that need ought to be individually determined: FI Michelman, ‘Foreword: On Protecting the Poor Through the Fourteenth Amendment’ (1969) 83 Harvard Law Review 7. 13 On the workfare globally, see J Peck, ‘Workfare: A Geopolitical Etymology’ (1998) 16 Environment and Planning Society and Space 133; on the Israeli experience, see A Beinisch, ‘Contracting Out from the Perspective of Public Law: Privatized “Welfare to Work” and the Limits of Regulations and Oversight’ (2008) 38 Mishpatim 283.
352 Neta Ziv (water, for example) cap usage may be imposed;14 if it is to enable young children to enjoy parental care, parents to young children will be exempt from work requirement, etc. The most common eligibility conditions are ‘targeted programmes’, directed at the poor, distinguishing between those in most need and those in less need.15 Such programmes are the least interventionist in market distribution, as they are geared towards reducing poverty rather than narrowing inequality.16 Korpi and Palme have acknow ledged, that although ‘support for targeting [social programmes] has decreased among social scientists, it has increased among policymakers’. They are also supported by international institutions such as the International Monetary Fund and the World Bank.17 I assume that in various circumstances imposing eligibility conditions – in principle – is appropriate while in others it becomes more problematic. However, the appropriateness of eligibility conditions in principle does not release us from examining the legitimacy of particular conditions. B. Human Rights Theory and Doctrine Notwithstanding controversies between policymakers (and academics) about the effectiveness of a universal versus selective social system, imposing eligibility criteria for social assistance raises questions regarding the theoretical basis underlying social rights as human rights, in terms of their universality. Human rights – both social rights and civil rights – are universal in the sense that every person is entitled to their respect, protection and fulfillment.18 How can we reconcile universal entitlement with a selective system of eligibility? The answer is based on the distinction between two understandings of social rights: the first views them as the unconditional right to receive a resource; the second as the unconditional right of reasonable access to a resource.19 According to White, both notions can serve as a theoretical basis for social rights: The notion of a social right can quite intelligibly be understood in the second way as well as in the first: as an unconditional right of reasonable access to a given resource, rather than as a right to be given this same resource unconditionally.20
14 See, eg the South African case City of Johannesburg v Mazibuko 2009 (8) BCLR 791 (SCA) (S Afr). In this case the South African Constitutional Court reviewed the legality of a city policy under which usage over a certain quantity of water resulted in a water cut-off. 15 On the targeted model as one social policy model, see Korpi and Palme (n 10) 667–69. 16 According to Esping-Andersen’s model these programmes represent more strongly the conservative state. G Esping-Andersen, The Three Worlds of Welfare Capitalism (Princeton, Princeton University Press, 1990). These programmes also correspond with a ‘minimum welfare state’ model (in contrast to a regulatory or distributive welfare state). On this point, see H Kliemt, ‘On Justifying a Minimum Welfare State’ (1993) 4 Constitutional Political Economy 159. 17 Korpi and Palme (n 10) 663. 18 M Green, ‘What We Talk About When We Talk About Indicators? Current Approaches to Human Rights Measurement’ (2001) 23 Human Rights Quarterly 1062, 1071; TH Marshall, ‘Citizenship and Social Class’ in Marshall (ed), Citizenship and Social Class (Cambridge, Cambridge University Press, 1950) 148; JW Nickel, ‘How Human Rights Generate Duties to Protect and Provide’ (1993) 15 Human Rights Quarterly 77. 19 S White, ‘Review Article: Social Rights and the Social Contract – Political Theory and the New Welfare Politics’ (2000) 30 British Journal of Political Science 507. 20 ibid 510.
‘Eligibility Conditions’ in Social Rights Litigation 353
White grounds his claim in social contract theory, in particular TH Marshall’s approach, under which the state has a duty to ensure that citizens have reasonable access to certain vital resources.21 However, he explains, ‘a right of reasonable access to a decent minimum of income does not necessarily have to take the form of a universal right to be given a minimum income unconditionally’.22 Accordingly, since a person with adequate personal wealth has reasonable access to subsistence, conditioning provision with need does not violate one’s human rights. Considering social rights in this manner, ‘reasonable access’ is an internal qualification within the right, in the sense that we cannot define the right to provision without identifying the criteria and conditions for entitlement. Thus eligibility conditions of this sort are not external factors to any ‘pre-existing right’; rather they are components of the right, delineating its scope and range. ‘Reasonable access’ is also closely related to the public purpose underlying the social arrangement. In order to determine whether a particular eligibility condition meets the reasonable access rationale, reasonableness has to be ‘broken down’ into its components. These include: (a) defining the purpose of the selective arrangement (fulfillment of basic needs, distributive justice, participation in the workforce, etc); (b) applying a means–end test, that is, enquiring whether the restrictive condition indeed furthers the purpose, whether it is proportionate in terms of costs versus benefit, etc. From a positivistic perspective, international human rights law recognises states’ leeway to impose conditions upon social provision. General Comment 19 of the Committee on Economic, Social and Cultural Rights on the right to social security, for example, states that ‘qualifying conditions for benefits must be reasonable, proportionate and transparent’.23 The comment continues to clarify that States parties are also obliged to provide the right to social security when individuals or a group are unable, on grounds reasonably considered to be beyond their control, to realise that right themselves, within the existing social security system with the means at their disposal (emphasis added).24
In other words, under the Covenant States may condition provision on individual determination of eligibility, as long as these conditions meet certain criteria: reasonableness, proportionality, transparency, and when individuals do not have the means to provide for themselves. C. Constitutionality The constitutionality of eligibility conditions entails a different analytical enquiry than the one made from the perspective of social policy and human rights. It is accepted that states may adopt a selective system as appropriate social policy, and that they can fulfill their international human rights obligations by instituting selective social rights arrangements. However, constitutional analysis provides a framework to review particular conditions in terms of their adherence to the duty to protect a human right. Marshall (n 18). White (n 19) 511. 23 UN Committee on Economic, Social and Cultural Rights, General Comment No 19: The Right to Social Security (Art 9 of the Covenant) (4 February 2008) UN Doc E/C12/GC/19, Art 24. 24 ibid Art 50. 21 22
354 Neta Ziv What are the constitutional implications of the discussion above? In statutes that incorporate eligibility conditions, entitlement is not born – that is, the right does not come into existence – until the designated criterion is met. It seems that from a jurisprudential perspective, once a selective mechanism has been put in place, all its elements – including substantive eligibility conditions – constitute and form an inherent part of the (social) right. In this sense, eligibility conditions are not ‘external prerequisites’ which lay outside the scope of full constitutional review. Rather, constitutional oversight of an eligibility condition is embedded in the process of judicial review of a law’s constitutionality. Constitutional judicial review (in Israeli law as well as other jurisdictions) often includes two stages.25 First a court must determine if a violation of a constitutional right has occurred, that is, that a constitutionally protected interest has been infringed. The burden of proof at this stage lies upon the party challenging constitutionality.26 At the second stage, conducted under the Limitation Clause,27 we examine if the restriction of a right is justified. First, we examine whether the limitation furthers a proper purpose, and if so, we continue to enquire if it meets the three-tiered proportionality test: (a) the violating measure bears a rational relationship to the purpose; (b) the purpose underlying the statute could not be achieved using measures that infringe less upon the right (the least restrictive requirement); and (c) overall the benefits of the violation outweigh its disadvantages.28 At this second stage the burden of proof to show that these requirements are met lays upon the party that asks to uphold the challenged arrangement.29 Where do eligibility conditions ‘fit’ within this constitutional analysis? At what stage should we examine their adherence to constitutional standards? To what extent does constitutional review of eligibility conditions differ from review of other qualifying conditions for provision? These questions lay at the heart of the Hassan case. In order to examine them, the next part maps out various schemes for qualifying resource allocation, and differentiates them from direct eligibility conditions – the main topic of this chapter. III. TYPES OF QUALIFYING CONDITIONS
A. Substantive Eligibility Conditions versus Verification Methods Eligibility conditions differ from methods for verifying entitlement for social provision or for revoking it. Oftentimes statutes include obligations imposed upon recipients as a 25 On the stages of constitutional review in Israeli law see, eg CA 6821/93 United Mizrahi Bank Ltd v Migdal Cooperative Village 49(4) PD 221 [1995] (in Hebrew); HCJ 6055/95 Zemach v Minister of Defence 53(5) PD 241 [1999] (in Hebrew); HCJ 6427/02 Movement for Quality Government v Knesset 61(1) PD 619 [2006] (in Hebrew). 26 See, eg HCJ 366/03 Commitment to Peace and Social Justice Association v Minister of Finance 60(3) PD 464 [2005] para 19 of Court President Barak’s opinion (in Hebrew). 27 The Limitation Clause in Basic Law: Human Dignity and Liberty states: ‘8. There shall be no violation of rights under this Basic Law except by a law befitting the values of the State of Israel, enacted for a proper purpose, and to an extent no greater than is required’, and in Basic Law: Freedom of Occupation: ‘4. There shall be no violation of freedom of occupation except by a law befitting the values of the State of Israel, enacted for a proper purpose, and to an extent no greater than is required, or by regulation enacted by virtue of express authorization in such law’. 28 HCJ 2887/04 Abu-Medigam v Israel Land Administration 62 (2) PD 57 [2007] para 36 of Justice Joubran’s opinion (in Hebrew). 29 United Mizrahi Bank (n 25).
‘Eligibility Conditions’ in Social Rights Litigation 355
condition for entitlement, however these are means employed by the state agency to verify substantive requirements. Such methods may include the duty to consent to entrance into one’s home to enquire about its inhabitants; reporting obligations imposed upon recipients to obtain information about income or family relationships; surveillance, inspections and other measures to acquire data about social assistance recipients. Numerous cases relating to social assistance in the United States have fallen under this category: the legitimacy of ‘midnight raids’ into welfare recipients’ homes,30 summary termination of allowance without a hearing,31 entering one’s home to obtain data about its inhabitants, etc.32 Some of these measures were annulled by the US Supreme Court as violating due process or privacy rights.33 Similarly, the Israeli Supreme Court has reviewed the legality of summary welfare repeal without a hearing and the petition led to the establishment of pre-termination hearings in most circumstances of welfare annulment.34 Verification measures of this sort relate to social rights entitlement, but are not the focus of this chapter. They often involve infringements of the right to privacy, due process, liberty and equality.35 Therefore, they are subject to heightened judicial review. Oftentimes they entail administrative or statutory interpretation rather than constitutional judicial review. Administrative verification methods of this sort should not be conflated with eligibility conditions that constitute ‘signals’ for other requirements. The ‘man in the house rule’ – under which welfare entitlement of single mothers was terminated upon evidence of male presence – is a condition of this sort. The rule, implemented by state agencies until invalidated by the US Supreme Court,36 is still utilised (in some form) by Israeli welfare agencies.37 Underlying this measure is a presumption that the presence of a man in one’s home indicates monetary contribution to the family and thus ‘stands for’ unreported income. Similarly, a rule that disentitles recipients from social allowance if they stay outside the country for more than a designated number of days,38 or own or use a car (as in the Hassan case), treats these conditions as markers of another factor, usually the availability of (concealed) means. For the purpose of this chapter I consider such conditions as substantive eligibility requirements; they will thus be addressed in part IV, and I argue for a strict standard of review of their constitutionality. 30 See CA Reich, ‘Midnight Welfare Searches and the Social Security Act’ (1963) 72 Yale Law Journal 1347. Night raids were abolished by the Federal Court in Parrish v Civil Service Commission 66 Cal 2d 260 (1967). 31 Goldberg v Kelly 397 US 254 (1970). 32 Wyman v James 400 US 309 (1971). 33 In general, see W Forbath, ‘Constitutional Welfare Rights: A History, Critique, and Reconstruction’ (2001) 69 Fordham Law Review 1821. 34 HCJ 1512/04 Hanuka v National Insurance Institute (10 April 2005), Nevo Legal Database (by subscription) (in Hebrew). Based on the settlement agreed upon by the parties the Court dismissed the petition on 10 April 2005. 35 PL Morgan, ‘Public Assistance for the Price of Privacy: Leaving the Door Open on Welfare Home Searches’ (2009) 40 McGeorge Law Review 227, 230–39. In the US advocates had attempted to claim they infringed the right to property: see generally, M Davis, Brutal Need: Lawyers and the Welfare Rights Movement 1960–73 (New Haven, Yale University Press, 1993) 81–89, 104–07. 36 King v Smith 392 US 309 (1968). 37 The definition of ‘partner’ or ‘spouse’ for welfare purposes had changed over time to include not only married couples but also those in some form of cohabitance. On this point, see E Albin, ‘Income Security Benefits for the Alternative Families – the Case of Polygamous Families in Studies in Law, Gender and Feminism’ in D Barak-Erez et al (eds), Studies in Law, Gender and Feminism (Srigim, Nevo, 2007) 617 (in Hebrew). 38 Income Supplementary Law, s 14A.
356 Neta Ziv B. Social Citizenship States often condition social rights eligibility on a requirement that a person maintains a particular association with the country – citizenship or some form of residency. Since state obligations pertain to people within their boundaries, and welfare being ‘closed systems’ – in the sense that the rights provided to some impact the rights of others – in principle such requirements seem sensible.39 More so, in an era of mass migration and open borders, policies relating to ‘social citizenship’ have become prominent topics of debate within destination countries favoured by migrant workers, refugees and other immigrants.40 Alongside the question of the appropriate policy in regards to ‘social citizenship’, in a number of countries the constitutionality of citizenship or residency requirements was adjudicated. The legal issue was whether conditioning eligibility for social rights upon residency or citizenship violated the basic right at stake. In Israel it was the right to social security; in South Africa the right to income support; in Germany the right to unemployment benefits. I will discuss these cases focusing on the constitutional standard of review for these conditions. i. Israel In 2003 an amendment to Israel’s National Insurance Law changed the definition of ‘residency’ for temporary residents (non-citizens) that hold certain visa types, restricting these groups’ entitlement to social security and national health insurance.41 Prior to the amendment the state had determined residency status according to a person’s individual circumstances and her actual association with the country (length of stay, centre of life, family attachments, etc). The amendment created a categorical exclusion, based solely on the formal visa type of the person. In Physicians for Human Rights v Minister of Finance42 the constitutionality of this amendment was challenged. Petitioners argued that the new law infringed the basic right to equality, to social security and to health, all recognised within the right to human dignity. They challenged the categorical, rather than the individual, exclusion of eligibility based on a formal visa status rather than on substantive personal determinates. The state responded that entitlement for social rights can be conditioned upon a determination of some ‘attachment’ to the state. Since de facto visa status is a strong indicator for such attachment, the categorization is legitimate. In addition, the state claimed that the disentitlement exclusion does include emergency medical treatment, certain maternity benefits and work accident insurance, the most pertinent of social rights. 39 In the words of Freeman ‘the preservation of the advantage of the welfare state entails limited access to their benefits’ (GP Freeman, ‘Migration and the Political Economy of the Welfare State’ (1986) 485 Annals of the American Academy of Political and Social Science 51, 52). 40 See, eg T Faist, ‘Boundaries of Welfare States, Immigrants and Social Rights on the National and Supranational Level’ in R Miles and D Thranhardt (eds), Migration and European Integration: The Dynamics of Inclusion and Exclusion (New Jersey, Fairleigh Dickinson University Press, 1995) 177, 179–80. 41 The amendment was enacted as part of the Arrangements Law (Legislative Amendments for Achieving the Budget Goals and the Economic Policy for the 2003 Fiscal Year), 5763-2002, ss 17(2)(a)(1), 17(2)(c), 17(3)(a), 17(11), 17(13). These benefits included, inter alia, income support, unemployment benefits, maternity benefits, old age benefits and health insurance. 42 HCJ 494/03 Physicians for Human Rights v Minister of Finance 59(3) PD 322.
‘Eligibility Conditions’ in Social Rights Litigation 357
The Court rejected the petition and affirmed the constitutionality of the Law, stating that a requirement for an attachment or affinity with the state is a reasonable condition for ‘social citizenship’, and that there exists a typical correlation between one’s visa type and the kind of linkage she maintains to the state. In the words of the Court ‘whoever holds a temporary visa, for the purpose of a visit, work, etc, would typically not retain a significant and ongoing affinity to the country’.43 As to the constitutionality of the social residence requirements, the Court enquired into the ‘relevancy’ of the condition (visa type) to the purpose of the legislation (ascertaining affinity and affiliation with the state), and was satisfied with a showing of a general correlation between a visa type and one’s substantive connection to the country. This means–ends test does not impose a heavy burden upon the state and is relatively easy to meet. It is similar to the ‘rational basis’ standard of review in the US constitutional jurisprudence. All the state has to demonstrate is that the condition is not arbitrary and that there is some correlation between the eligibility condition and the law’s purpose. A more stringent test would be troubled by instances in which individuals with strong and long-lasting connections to the country – though still holding temporary visas – would not be entitled to social security and health insurance. It would ask the state to present detailed information about the number of such cases, to demonstrate that determination of individual entitlement would impose real hardship (not just inconvenience) and that overall the benefits of the exclusion outweighed its disadvantages. My point is not that the case should have inevitably ended differently, but that the lenient standard of review for social citizenship created an analytical framework that could significantly weaken judicial review of social rights. ii. South Africa In 2004 the South African Constitutional Court decided a case relating to eligibility requirements of South Africa’s Social Assistance Act 2004.44 This law conditioned receiving certain social benefits (old age grants, child support grants and dependence care grants) on South African citizenship. The petitioners were citizens of Mozambique who were permanent residents in South Africa (some acquired residence through work, others as refugees). They were all destitute and would have qualified for social benefits but for the citizenship requirement. The challenge to the citizenship requirement was based on section 27 of the South African Constitution, which guarantees the right to healthcare, food, water and social security to ‘everyone’; the right to equality (section 9) the right to life (section 11) and the right to dignity (section 10).45 The Constitutional Court struck down the clause conditioning social assistance upon citizenship. It enquired into the bond between the country and permanent residents, describing them as ‘those who have become part of our society and have made their home in South Africa’.46 The Court relied on specific legislation that required treating
ibid para 13 of Court President Barak’s opinion. Khosa v Minister of Social Development 2004 (6) SA 505 (CC) (S Afr). 45 Insofar as child allowance was at stake the petition was also based on the rights of children, recognized in the South African Constitution, 1996, s 28. 46 Khosa (n 44) para 58. 43 44
358 Neta Ziv permanent residents similarly to citizens,47 and distinguished between permanent and temporary residents.48 The point, however, is not the ruling’s outcome, but the constitutional analytical process, as well as the standard of review applied by the Court. The decision (85 pages long) dwells in minute detail on every argument of the Government, scrutinizing it thoroughly, through the Constitution’s ‘Limitation Clause’ (section 36). The Government claimed that permanent residents may acquire citizenship and therefore excluding residency is legit imate, but the Court stated that such a result was unlikely given South Africa’s immigration policy. The state asserted that its citizens have priority over permanent residents, but the Court rejected this justification, citing the Immigration Act (mentioned above) as proof of their equal status. The state claimed that accepting the petition would impose financial hardship, but the Court concluded that the state did not offer precise figures about the number of potential eligible beneficiaries or anticipated expenses. The state argued that it wanted to promote self-sufficiency rather than dependency of permanent residents, but the Court reasoned that it was the state itself that decided to give these people the status of permanent residents, and cutting them off from basic social assistance ran counter to the state’s ‘constitutional commitment to developing a caring society’.49 More so, the state referred to US constitutional jurisprudence, which affirmed the exclusion of social assistance from non-US citizens, utilizing the ‘rational basis’ standard of review.50 Justice Mokgoro distinguished between the US constitutional standard and that of South Africa: I am prepared to assume that there is a rational connection between the citizenship provisions of the [South African] Act and the immigration policy it is said to support. But that is not the test for determining constitutionality under our Constitution. Section 27(2) . . . sets the stand ard of reasonableness which is a higher standard than rationality.51
The Court concluded that the distinction was discriminatory as it excluded ‘a vulnerable group’, stigmatised inferiority, created dependency and its totality alluded to an infringement of human dignity of people who were ‘part of the South African Community’.52 This review standard of social citizenship conditions is thus stricter than the one adopted by Israeli or US courts. iii. The European Court of Human Rights In Cevat Gaygusuz v Austria the European Court of Human Rights (ECtHR) addressed the legality of an Austrian statute that denied unemployment emergency benefits to nonAustrian citizens.53 The petitioner, a Turkish national, had legally lived and worked in Austria for about 14 years, during which he had paid his insurance contributions and taxes. Due to medical conditions which rendered him unable to work, the petitioner 47 Immigration Act (No 13, 2002), s 25(1) (S Afr) holds that ‘The holder of a permanent residence permit has all the rights, privileges, duties and obligations of a citizen, save for those rights, privileges, duties and obligations which a law or the Constitution explicitly ascribes to citizenship’. 48 Khosa (n 44) para 59: ‘It may be reasonable to exclude from the legislative scheme workers who are citizens of other countries, visitors and illegal residents, who have only a tenuous link with this country’. 49 ibid para 65. 50 City of Chicago v Shalala 189 F 3d 598 (1999), cert denied 529 US 1036 (2000). 51 Khosa (n 44) para 67. 52 ibid paras 68–78. 53 Gaygusuz v Austria (1996) 23 EHRR 364.
‘Eligibility Conditions’ in Social Rights Litigation 359
applied for an advance on his pension in the form of emergency assistance (monetary assistance for those no longer entitled to unemployment benefit, guaranteeing a minimum income). Gaygusuz was denied this benefit, since it was conditional upon Austrian citizenship. Gaygusuz argued before the ECtHR that the citizen-only provision in Austria’s unemployment law contradicted the European Convention on Human Rights (ECHR),54 infringing his right to a fair hearing, respect for private and family life, peaceful enjoyment of his possessions and to non-discriminatory treatment. The ECtHR found the Austrian law in violation of the ECHR. It recognised that the Convention protected the benefit at stake (a pecuniary right enjoyed by the petitioner) and thus addressed the counter arguments raised by the Government defending the distinction. The Court enquired whether the distinction furthered ‘a legitimate aim’ and if there was a ‘reasonable relationship of proportionality’ between the end and the means employed, imposing a heavy burden on the state to fulfill them.55 The Government’s claims – that the state has special responsibilities to its own citizens; that nationality was a relevant criterion in social legislation; that there was no contractual obligation towards Turkish nationals – were dismissed by the Court as not meeting this standard. In sum, the South African Supreme Court and the ECtHR applied a stricter standard of review for social citizenship than the Israeli standard adjudicated in Physicians for Human Rights. This is not to say the Israeli case would or should have ended differently had the court applied the South African or ECtHR’s analysis; rather that the Israeli Court adopted a lenient standard for scrutinizing exclusionary social citizenship conditions for social rights.
C. Discriminatory Conditions Social rights entitlement can also be conditioned upon discriminatory criteria, based on a suspect classification (race, gender, nationality); however their analysis is somewhat different. The prohibition on discrimination is a universal norm in international human rights and constitutional law. Article 2(2) of the International Covenant on Civil and Political Rights obligates states to ensure that social rights will be respected without discrimination of any kind as to race, colour, sex, language, religion, political or other opinion, national or social origin, property, birth or other status.56
Similar provisions can be found in the European Social Charter;57 the Additional Protocol to the American Convention on Human Rights in the Area of Economic, Social 54 Convention for the Protection of Human Rights and Fundamental Freedoms (European Convention on Human Rights, as amended). 55 Gaygusuz (n 53) para 42. 56 International Covenant on Economic, Social and Cultural Rights (adopted 16 December 1966, entered into force 3 January 1976) 993 UNTS 3. 57 European Social Charter (18 October 1961) CETS 35, safeguards, among others, the right to employment, housing, health, social security and protection against poverty, ‘without discrimination on grounds of race, colour, sex, religion, political opinion, national extraction or social origin’ (Preamble, and pt II, Arts 1–19). See similar provisions in the European Social Charter (Revised) (3 May 1996) CETS 163, pt V, Art E and pt I.
360 Neta Ziv and Cultural Rights (Protocol of San Salvador)58 and the African (Banjul) Charter of Human and Peoples’ Rights.59 When the group-based eligibility distinction is explicit – such as a rule that does not recognise the right of women to own or to inherit property – it is discussed and adjudicated within the context of non-discrimination.60 Although it can be considered a form of an eligibility condition, the focus of judicial review is the right to equality, which usually entails a heightened standard of judicial review. Yet some discriminatory conditions are indirect and implicit – they appear neutral but impose a disparate negative impact on a protected group. In the Israeli context military service is a case of this sort. Since a number of social groups do not serve in the army – the Arab citizens of Israel, persons with disabilities, some new immigrants – criteria that condition entitlement of social services upon military service often mask indirect discrimination and can be treated as such.61 From a legal perspective – statutory as well as constitutional – if neutral criteria operate as a cover for unequal treatment, they can be considered as prohibited discrimination. In 2002, the Knesset (Israeli Parliament) amended the section relating to child allowance in the National Insurance Law,62 differentiating between parents who served in the military and those who did not, establishing a preference to those who served.63 A number of individuals and civil society organizations challenged the law in the Supreme Court.64 The petitioners claimed the law was discriminatory, and thus asked for a stringent level of judicial review of criteria disparately impacting a minority group, in particular since social benefits were at stake. The Supreme Court was about to deliver a decision following an en banc hearing (by 13 Justices) however, before the ruling, the Government extracted the amendment and petitions were withdrawn. Hence when social rights are conditional – either explicitly or by way of a non-bona fide neutral qualification – on belonging to a protected/suspect group, constitutional review falls within the equal protection domain, thus subject to heightened scrutiny and imposing a heavier burden upon the state to justify the condition. The issue that remained unresolved, and arose in the Hassan case, was the review standard of eligibil58 Additional Protocol to the American Convention on Human Rights in the Area of Economic, Social and Cultural Rights (Protocol of San Salvador) (entered into force 16 November 1999) OAS Treaty Series No 69 (1988) reprinted in Basic Documents Pertaining to Human Rights in the Inter-American System OEA/Ser L V/ II82 Doc 6 Rev 1 at 67 (1992). The Protocol adds to the protection of economic and social rights recognized in two other human rights instruments in the Americas: American Declaration of the Rights and Duties of Man, OAS Res XXX adopted by the Ninth International Conference of American States (1948) reprinted in Basic Documents Pertaining to Human Rights in the Inter-American System OEA/Ser L V/II82 Doc 6 Rev 1, 17 (1992) and American Convention on Human Rights, ‘Pact of San Jose’, (adopted 22 November 1969, entered into force 18 July 1978). 59 African Charter on Human and Peoples’ Rights (adopted 27 June 1981, entered into force 21 October 1986) (1982) 21 ILM 58. The Charter recognizes certain social rights (health, housing, education and family life) and includes a prohibition on discrimination. 60 An interesting case in point is the decision of the High Court of Swaziland in the case of Mary-Joyce Doo Aphane (Aphane v Registrar of Deeds and others [2010] SZHC 29 (SZ), available at: www.swazilii.org/sz/ judgment/high-court/2010/29) in which a statute restricting married women in common law marriages to have immovable property registered on their name was found to be unconstitutional according to the Swaziland Constitution. 61 On the effect of eligibility criteria of different social groups in Israel, see Stier and Lewin (n 9). 62 National Insurance Law (Consolidated Version), 5755-1995. 63 Economic Emergency Programme Law (Legislative Amendments for Achieving Budgetary Goals and the Economic Policy for the 2002 and 2003 Fiscal Years), 5762-2002. 64 HCJ 4953/02 John Does v Government of Israel (unpublished). The case was dismissed following a change in the Government position.
‘Eligibility Conditions’ in Social Rights Litigation 361
ity requirements that are neither based on social citizenship nor exclude a ‘protected group’. IV. CONSTITUTIONALITY OF SOCIAL RIGHTS IN ISRAEL
The debate on the constitutionality of social rights in Israel emerged following the 1992 enactment of Israel’s Basic Laws (Basic Law: Human Dignity and Liberty, and Basic Law: Freedom of Occupation). At first it was not clear if social and economic rights were protected constitutional rights. This debate is largely over. Israeli jurisprudence has gradually but consistently recognised that the basic right to human dignity includes social and economic rights and they enjoy constitutional protection.65 The Israeli Supreme Court began acknowledging the constitutionality of social and economic rights in 1998, in a decision in which the then Court President Barak declared, as obiter dictum, that human dignity includes a ‘minimum core’ of resources in the area of housing, health and education.66 In the following decade, after petitions filed largely by non-profit Israeli organizations, the court expanded its recognition of social rights, such as the right to education, health, housing, welfare, old age and child allowance – as protected constitutional rights.67 As Barak-Erez and Gross explain, the outcome of these cases was mixed: on the one hand the Supreme Court acknowledged that social and economic rights enjoy constitutional protection and are subject to judicial review; on the other hand they resulted in a narrow protection of their content and scope. The Court ruled that constitutional protection guaranteed a ‘minimum core’ of resources, but no more.68 How much is ‘minimum’? In the context of income support minimal core apparently includes the right to receive only the basic resources for subsistence. In the leading case on judicial review of social rights – Commitment to Peace and Social Justice Association v Minister of Finance69 – petitioners challenged the reduction in income support benefits (welfare), claiming that it amounted to a violation of the right to human subsistence. Rejecting this claim Court President Barak stated that constitutional protection extends only a basic ‘safety net’, ensuring a person has enough food and drink in order to live; a place to live in which he can realise his privacy and his family life and be protected from the elements; tolerable sanitation and medical services, which will ensure him access to the facilities of modern medicine.70
65 Y Rabin and Y Shani, ‘The Israeli Unfinished Constitutional Revolution: Has the Time Come for Protecting Economic and Social Rights?’ (2003–04) 37 Isrrael Law Review 299, 301–08; Barak-Erez and Gross (n 3). 66 LCA 4905/98 Gamzu v Yeshayahu 55(3) PD 360, 375–76 [2001] (in Hebrew). See also HCJ 4128/02 Adam Teva Va Din v Prime Minister of Israel 58(3) PD 503, 518 [2004] (in Hebrew); LCA 5368/01 Yehuda v Teshuva 58(1) PD 214, 220–22 [2003] (in Hebrew). 67 On this process, see G Mundlak, ‘Fifty Years to the Implementation of The National Insurance Act – Parties will Take Place in the Supreme Court’ (2004) 67 Social Security 83, 96–102 (in Hebrew). 68 Barak-Erez and Gross (n 3). See also R Shiv, ‘“Stop the Revolution without Me”: On the Promulgation and Disciplining of the Discourse of Distributive Justice and Social Rights’ (2009) 2 Tel-Aviv University Journal on Law and Social Change, 185, 194–95. 69 Commitment to Peace (n 26). 70 ibid 483–84.
362 Neta Ziv More so, in Commitment to Peace the Court imposed a relatively heavy burden of proof upon challengers of a social right to demonstrate that an infringement of a protected basic right had occurred. Court President Barak stated that petitioners must provide the Court ‘with a complete factual basis’ proving violation of dignity, including ‘details, based on appropriate documentation, of the sources of income and the current and fixed expenditure of that person’ against ‘all the national and other support systems that assist that person and the steps he takes in approaching them in order to exhaust his rights’.71 As the petitioners did not provide the required proof, the Court ended constitutional review at the first stage, and did not require the state to justify the measure. This stringent burden of proof standard set by the Court constitutes a significant impediment for effective judicial review.72 This point cannot be underscored. The question of proof – who has to prove what – is often what lies at the heart of the dispute during judicial review. Imposing an onerous duty on petitioners might limit judicial review significantly. It frees the state from the need to justify the measures it utilises and leaves it with broad leeway to curtail social resources with limited judicial oversight. To what extent does the Commitment to Peace ruling pertain to judicial review of eligibility conditions? This was the legal question in the Hassan case. V. CONSTITUTIONALITY OF ELIGIBILITY CONDITIONS – THE HASSAN CASE
Eligibility conditions vary: they may constitute behavioural requirements, such as participation in work training and placement, refraining from drug abuse, cap for water usage; or direct means testing – level of income and savings, past and present property ownership, number of family members; as well as indirect means testing (ie conditions that ‘mark’ or indicate the existence of income) – frequency of travel abroad, or use/ ownership of a car. I have already suggested that such conditions constitute an inherent element of a social right and that consequentially, their constitutionality ought to be determined through the established judicial model applied to human rights. But what does this mean? At what stage of review and under which standard should eligibility conditions be scrutinised? As discussed above in section IIIB, in the context of social citizenship the Israeli Supreme Court ruled that residency or citizenship requirements are subject to a lenient standard of review – rational basis. But residency and citizenship may be considered a distinctive category. At times they amount to discriminatory practices, and then should be treated within that category. Moreover, social citizenship demarcates the boundaries of community solidarity (given the distributive nature of social rights) and thus are threshold requirements, defining ‘who’ is eligible to a certain resource, rather than ‘when’ and ‘under what circumstances’. Hence the review standard applied to social citizenship need not bind other eligibility requirements.
ibid 485. In contrast, in his dissenting judgment, Justice E Levy ruled differently: ‘I am of the opinion that all of the material that was presented to us allows us to determine that the constitutional human right to live with dignity has been violated as a result of the reduction, or at least that the petitioners have succeeded in raising a real doubt as to whether the recipients of the benefits are able to support themselves with dignity’ (ibid 499–500). 71 72
‘Eligibility Conditions’ in Social Rights Litigation 363
A. Background and Arguments In the Hassan case the Supreme Court was asked to review the constitutionality of section 9A(b) of Israel’s Income Supplement Law,73 which denied welfare to any person – otherwise eligible – who regularly used or owned a car.74 This Law includes an ‘income test’, to verify that only the neediest would receive support. The National Insurance Institute (NII) thus argued that regular use of a car (or ownership thereof) ought to be construed as ‘income’; the amount of this income was to be appraised at the level of an individual’s allowance. Under this formula there need not be any correlation between the extent and frequency of car usage (or value of the owned car) and the actual amount of presumed income. Section 9A created a non-refutable presumption under which income was assumed, and at the level of one’s eligibility. The petitioners – Jews and Arabs together with Israeli non-profit organizations – argued that such a broad and undifferentiated condition was a constitutional violation of the basic right to human dignity, which does not survive the Limitation Clause, with its three-tiered test of proportionality.75 They described the circumstances of the petitioners, who all either owned a car or had used one (borrowed from family members or friends) for shopping, driving to work, taking children to school or visiting the doctor. Their benefits were denied or revoked.76 In response, the state (and the NII) articulated their position on the constitutional review model of social rights, based on the following elements: 1. Constitutionality of the car ownership/usage condition ought to be determined at the first stage of constitutional analysis, that is, when the court examines whether a violation of a protected constitutional interest had occurred. There are valid policy and methodological reasons not to conduct the analysis at the second stage of review, in lieu of the positive nature of social rights and the financial ramifications of review. 2. The protected interest in this case is the right not to be compelled to live in poverty, that is, the right not to live in dire need involuntarily/inevitably. This means that if a 73 During the Court hearings the Income Supplement Law was amended in 2007 and the prohibition on car ownership and usage was narrowed, as follows: a welfare recipient could own or use a car if they worked and received an income not exceeding 25% of the average Israeli income, and providing that the car did not surpass a certain engine size and yearly model: see Income Supplement Law (n 2) s 9A(c). The amendment improved the situation of some welfare recipients but did not alter the arrangement altogether. 74 According to the state, as mentioned in writing and orally during the hearing, regular use of a car included usage once or twice a week, or even less frequently. The NII employs inspectors who conduct surveillance and inspections of welfare recipients, who then report their findings. 75 In Hassan HCJ 10662/04 and HCJ 3282/05 (n 1), petitions were filed directly against the NII challenging s 9A; petitioners were represented by Adalah – The Legal Center for Arab Minority Rights in Israel, and ItachMaaki Women Lawyers for Social Justice. In HCJ 7804/05 the petition was filed against the National Labour Court following an individual rejection of a welfare claim due to the regular use of a car. The petitioner was represented by the State Legal Aid Office. All cases were conjoined and the decision related to the three petitions. 76 This was not the first case in which the legality of eligibility conditions was brought before the Supreme Court; however previous cases did not end in a ruling. In HCJ 1433/03 Bachtin and others v Minister of Finance (3 November 2008), Nevo Legal Database (by subscription) (in Hebrew), the petitioners challenged the constitutionality of a work requirement imposed upon mothers of young children to accept any job offered to them. The petition ended in a consent decree. See D Freibach Hefetz and N Ziv, ‘A Legal-Social-Cultural Struggle of Care Rights and Duties: The Case of Bachtin v The Ministry of Finance’ (2012) 13 Labor, Society and Law 267. The author of this chapter was one of the lawyers representing the petitioners in the Bachtin case.
364 Neta Ziv person has a choice not to live in poverty, she does not meet the compelling need/ involuntary definition of a protected constitutional interest. 3. The eligibility requirement is a precondition to the right itself. As such, we must check if it corresponds with the protected interest (as defined above). Only if the (pre) condition bears no relevance to the protected interest, is it unconstitutional. This analysis is similar to the holding in Physicians for Human Rights; the state thus equated car ownership/usage with social citizenship. 4. Ownership or regular use of a car is either an indicator of unreported income (since driving a car entails high costs), or should be considered as if it was income, provided to the recipient as an ‘in kind’ resource by family and friends. This ‘income’ is grossed up to the level that disentitles the recipient from eligibility. 5. Accordingly, the recipient’s income is calculated to include a surplus, and she is not considered to be in poverty involuntarily.77 There are two key elements to this legal construction. First, the state had added a qualifying condition to the definition of right itself: no longer is it defined as ‘the right not to live in existential want’, or ‘the right to minimum subsistence’. It is all the above, but with a qualification: poverty must be involuntary or inevitable. Second, eligibility is a ‘precondition’ to the right; therefore it needs only to bear a relatively weak connection to the protected interest – as defined above. Accordingly, the condition will survive judicial review if it ‘corresponds’ with the interest, that is, meets the rationality test. Since all one needs to show is some relevancy between the protected interest and the eligibility condition, this burden is quite easy to meet. The petitioners argued, in contrast, that this is a problematic legal construct. They did not contest that eligibility conditions are meant to distinguish between those who can work and those who cannot; between people who are in dire economic need and those who are not. The problem lies, however, with the means–end linkage. Eligibility conditions often lie at the heart of social rights, delineating their scope; thus they are internal to the rights. This ‘structure’ of a right is, for the most part, more common to social and economic rights that are ‘positive’ at the core.78 As such, the petitioners argued that they ought to be scrutinised scrupulously – through the Limitation Clause, with its two substantive stages of review.79 First the state has to articulate and persuade that the purpose of the limitation is proper; then it must pass the three subsets of the proportionality test: 1) the violating measure bears a rational relationship to the purpose; 2) the purpose underlying the statute could not be realised using measures that are less infringing of the right; 3) the overall benefits of the violation outweigh its drawback. These are not easy tests to fulfil, and the burden to persuade the court they are met lies with the party asking to uphold the statute, usually the state. Affidavit on behalf of respondents, November 2009. On file with author. Not all social rights are positive at their core. The right to form a workers’ union, or the right to a family, also recognized as social rights, are liberty-based. 79 Other constitutions contain similar notions but may be phrased differently. See, eg the South African Constitution, 1996, s 36: ‘36. Limitation of rights: 1. The rights in the Bill of Rights may be limited only in terms of law of general application to the extent that the limitation is reasonable and justifiable in an open and democratic society based on human dignity, equality and freedom, taking into account all relevant factors, including: a. the nature of the right; b. the importance of the purpose of the limitation; c. the nature and extent of the limitation; d. the relation between the limitation and its purpose; and e. less restrictive means to achieve the purpose’. 77 78
‘Eligibility Conditions’ in Social Rights Litigation 365
The analysis proposed by the state collapsed the ‘proper purpose’ requirement into the first stage of analysis, and asked to end it there. Under this framework eligibility conditions would hardly ever need to meet the requirements of the Limitation Clause at all and only pass the rationality/relevance test. By applying the rationality test we forgo the two other subtests, imposing a heavy burden on the right challenger to show arbitrariness and irrational exclusion. This is almost impossible to do; thus adoption of this standard would weaken judicial review of social rights significantly. B. Ruling On 28 February 2012 the Supreme Court accepted the petitions, and invalidated section 9A(b) of the Income Supplement Law. Court President Beinisch who wrote the main opinion clarified some questions regarding the constitutionality of social rights, albeit leaving some unresolved. As a general statement Beinisch rejected the claim that there exists an inherent difference between social rights and economic rights on the one hand, and civil and political rights on the other, for the purpose of constitutional analysis. Beinisch discarded the assertion that review of social rights entails unique difficulties due to their vagueness and due to its potential monetary implications.80 Discarding the interpretive model set forth by the state Beinisch stated: [A]doption of the respondents’ approach means imposing the burden of proof upon the petitioners almost in full . . . petitioners will have to prove the infringement of the right as well as the legitimacy of the means chosen in the legislation . . . Changing the rules of scrutiny means relieving respondents from the need to prove the constitutionality of the means chosen by them.81
Following this analysis the Court then continued to enquire whether section 9A infringes the right to human dignity in the sense of minimal human subsistence, and resolved that it does. The purpose of the section, declared Beinisch, was to ensure that only the neediest would receive support and that examination of a recipient’s income is thus legitimate. However, the Court found the flaw in the categorical and irrefutable presumption under which any use (or ownership) of a car was calculated as income at the level of the welfare allowance: [T]he rule infringes upon the right to dignified human subsistence because it stipulates a categorical rule under which anyone who has ownership or usage of a car will not be entitled to income support; this is so with no relation to the individual question if indeed that person has an income at the level that can secure realization of the right to minimal subsistence in dignity.82
The Court was not persuaded that this mechanism indeed verified that only those with the least means would receive welfare. Subsequently, the burden of proof to show the connection between the end of the statute and its means was not met. The Law was thus declared unconstitutional, and the state was given a period of six months to remedy and amend the Law and implement an alternative mechanism. Hassan (n 1) paras 29 and 31 of Court President Beinisch’s opinion. ibid para 30 of Court President Beinisch’s opinion. 82 ibid para 40 of Court President Beinisch’s opinion. 80 81
366 Neta Ziv During the litigation (which lasted over six years) the amount of income support was never discussed or disputed. The case was not about how much the state was providing but to whom and under what circumstances. Therefore the ruling in Commitment to Peace regarding the minimal core was the presumed starting point for deliberation in the Hassan case. Can the Hassan ruling affect the question of level/amount? I will conclude with initial thoughts on this topic. VI. CONCLUSION
In our neo-liberal age Israel’s traditional welfare state, which embodied a relatively high level of universalism, will in all likelihood continue its transformation into a minimum welfare state. Social programmes will increasingly include eligibility requirements and restrictive qualifications for social provision. Given this trend, it is critical to sustain constitutional judicial scrutiny of social rights and to employ a meaningful process for reviewing eligibility conditions within them. In the Hassan case the Supreme Court ruled that review of eligibility conditions ought to be conducted at the second stage of analysis, which requires the state to justify the measures it utilises. Can this ruling affect judicial review of the quantity of resources the state has to provide to fulfill its human rights obligations? On the one hand the Hassan case did not address this issue; therefore a narrow reading of the decision would imply it does not. More so, Court President Beinisch repeated the spirit of the constitutional protection guaranteed by the right to human subsistence, stating that ‘the right to a minimum of human subsistence in dignity is at the heart and core of human dignity. Life in hunger, with no roof over one’s head, in a constant search for help, is not a life in dignity’.83 This is the same minimal standard – the right not to live in ‘existential want’, if you may – that the Court approved in Commitment to Peace. At the same time Court President Beinisch rejected the assumption articulated by the state – one which had previously been articulated in Commitment to Peace as well – regarding the direct correlation between choice and poverty. In both cases the state underscored the presumption regarding the potential of welfare policies to shape individual conduct. It asserted that facilitating access to welfare (by relaxing barriers or increasing the amount of allowance) would lead recipients to adopt unwanted behaviour, namely, over reliance on state provision instead of joining the workforce. This dynamic, known as the ‘poverty trap’ was not rejected by the majority ruling in Commitment to Peace,84 and was repeated by the state in Hassan. By arguing that the protected constitutional interest was the right not to live in poverty involuntarily the state furthered an assumption that use of a car is a choice that can be avoided and that it signals the existence of (concealed or unreported) income. I have already mentioned that the Court rejected the blanket assumption about the correlation between using a car and having an income. But Court President Beinisch also commented about the question of choice, stating that ibid para 35 of Court President Beinisch’s opinion. This assertion was rejected, however, by Levy J dissenting in, Commitment to Peace (n 26) 516: ‘“Poverty traps” are not created only as a result of benefits that are used to supplement income. This approach is erroneous and misleading’. 83 84
‘Eligibility Conditions’ in Social Rights Litigation 367
[a]lthough the inevitability request seems justified in theory, in practice the distinction between choice and inevitability is not simple at all. The question which life circumstances are the outcome of free choice and which life circumstances are the outcome of inevitability and necessity is oftentimes complex. Where does the boundary between free choice and social construction lie? Life choices are influenced, among other things, by the environment in which a person had been raised – from family, economic and social conditions.85
This reasoning represents a viewpoint that incorporates structural aspects and external facets – in additional to personal choice – into the constitutional analysis of social and welfare rights. It rejects the one-dimensional approach that assumes a direct correlation between welfare policy and personal choices. In this sense the Hassan decision embodies extensive implications to social rights adjudication. It may posit an opportunity to reexamine the broad spectrum of social rights in constitutional litigation.
Hassan (n 1) para 51 of Court President Beinisch’s opinion.
85
24 The New Consensus on Enforcing Social Welfare Rights: Comments on Three Papers MARK TUSHNET
I. INTRODUCTION: THE EMERGENCE OF CONSENSUS ON ENFORCEMENT OF SOCIAL WELFARE RIGHTS
A
CCORDING TO NETA Ziv,1 the debate about including social welfare rights in Israeli constitutional law is ‘largely over’ – and not merely in Israeli constitutional law. I believe that the debates over that question around the world are also largely over. Constitution designers and courts have treated such rights as constitutionally guaranteed. The issues for the future are not whether social welfare rights should be constitutionally guaranteed and judicially enforced, but, first, how they should be enforced by courts, and second, what are the conditions leading to effective (and ineffective) enforcement. The papers discussed here represent a new generation of scholarship on constitutionally guaranteed social welfare rights. I begin with some comments on how the emergence of a consensus on the inclusion of social welfare rights in constitutions led to the end of the first generation debates, and then turn to the issues that the new generation of scholarship will be addressing. The apparent consensus on judicial enforcement of second-generation rights has entailed the widespread rejection of several objections to such enforcement that were common before the consensus emerged.2 The objections took the form of asserting that there were categorical differences between ‘traditional’ rights, that constitutions protected and courts enforced, and social welfare rights. The objections rested on principles of separation of powers, expense, and ineffectiveness. Experience and scholarship have established that claims about categorical differences could not be sustained. To the extent that traditional rights and social welfare rights differ, the differences lie in locating specific traditional rights and specific social welfare rights along a spectrum ranging from ‘fully enforceable’ to ‘not likely to be enforceable’. And, sometimes, we will find a specific social welfare right closer to the ‘fully enforceable’ end of the spectrum than some traditional rights.3 Ch 23 in this volume. These paragraphs are based on M Tushnet, ‘Reflections on Judicial Enforcement of Social and Economic Rights in the Twenty-First Century’ (2011) 4 National University of Juridical Sciences Law Review 177, with some modifications. 3 Without addressing other difficulties in this note, I suggest an example: a right to life-sustaining medication may be more easily enforced than a right to be protected against torture (depending on the circumstances 1 2
370 Mark Tushnet
II. THE FADING IMPORTANCE OF SEPARATION OF POWERS OBJECTIONS
The separation of powers objection focused on the programmatic nature of second- generation rights. In one version, the argument was that the term ‘rights’ could not fairly be applied to the interests at stake when these ‘rights’ were discussed. A right, the argument went, required someone or some entity under a corresponding duty, but duties could arise only in connection with relatively well-defined interests. Social and economic ‘rights’ lacked the specificity needed to generate correlative duties. Legislatures could of course pursue programmes that advanced interests in housing and food and the like, but courts could not develop a jurisprudence of rights implicating those interests. A related objection, or perhaps simply a rephrasing of that objection, was that implementing programmes to protect these interests necessarily required a high degree of discretion in the implementer, a much larger degree of discretion than courts enforcing rights could possibly have. Once courts began to try to enforce social welfare rights, these generalised separation of powers objections turned out to be less forceful than it seemed at first. The objections assumed that civil and political rights were categorically different from social welfare rights, for otherwise separation of powers would have blocked the enforcement of civil and political rights, a result no one thought correct. Yet, no such categorical distinction exists. As Amir Paz-Fuchs shows,4 civil and political rights entail governmental duties. Consider the right to vote: the government must provide the facilities for voting, and one can without much difficulty generate substantial arguments that protecting the right to vote requires that the government devote significant resources toward making it possible for people who have the franchise to exercise their right to vote – by making polling places accessible, for example. In addition, courts devised ways of moving from broad statements of the interests at stake to more precisely specified ones, with the effect of allowing them to treat the more precise specifications as ‘rights’ fitting into the traditional category where rights entailed duties. I deal with this in more detail below in connection with Neta Ziv’s chapter. Experience with the enforcement of civil and political rights showed as well that that enforcement also involved a substantial amount of judicial discretion, not only in implementation but even at the rights-specification stage. The emerging consensus was that these experiences showed that separation of powers principles could not support a sharp distinction between first- and second-generation rights. Professor Paz-Fuchs shows in detail how the objections based on fiscal considerations has persisted. But, I believe, they persist only because those who offer them have a mistakenly narrow idea of the remedial forms associated with the judicial protection of rights. It is not merely that enforcing civil and political rights is costly too, as Paz-Fuchs argues. The central image of judicial remedies is the coercive order directing a target – the executive or the legislature – to take (or refrain from taking) specific actions to ensure compliance with constitutional requirements. Sometimes the costs of such coerunder which the torture occurs). The facts relating to the former can be explored easily in ordinary litigation, whereas the facts relating to the latter might be quite difficult to uncover, making it easier to enforce the former than the latter. 4 Ch 22 in this volume.
Comment on Enforcing Social Welfare Rights 371
cive remedies are indeed large. So, for example, a coercive order directing the government to supply housing to all those in need could impose quite substantial budgetary demands, far larger than the costs associated with complying with many other constitutional rights. But, courts have devised different remedial forms, which make the costs of enforcing social welfare rights comparable to those of enforcing civil and political rights. These forms include (mere) planning requirements, in which the relevant bureaucracy is charged with developing a plan to implement the judicially specified right, with the courts only loosely supervising the plan’s implementation.5 Yet, by embedding the right within the bureaucracy, the courts may move the bureaucracy toward an internalised commitment to implementation; lax judicial supervision is then not a bug but a feature. Another form is consultation or engagement, in which the courts insist that government officials interact directly with those who are adversely affected by their resource allocation decisions.6 Again, there is no guarantee that consultation and engagement will result in changes on the ground,7 but here too the hope is that direct human confrontation will induce bureaucrats to modify their plans. Those who raise the fiscal objection might argue that the costs of implementing traditional rights are low relative to the costs of enforcing second-generation rights. That overlooks several points. Exercising constitutionally protected civil and political rights imposes costs on society at large, and those costs can be large. As Professor Gross’s discussion of the Majority Camp decision shows,8 increasing the risk of disorder demonstrations elicits an increase in police deployment. Speech critical of government policy exemplified by the Wikileaks episode also increases social risk and entails costs to prevent repetition. Yet no one contends that the risks and expenses associated with demonstrations and critical speech imply the inappropriateness of judicial enforcement of rights to free expression. The difference between the costs associated with enforcing civil and political rights and social welfare rights is not that the former are small and the latter large, but that the former are generally invisible because they are diffused across the society as a whole without figuring openly in government budgets, while the latter are immediately visible in budget statements. The new remedial forms reduce the immediate, and immediately visible, costs of enforcement, responsive to formulations such as the ‘within available resources’ qualification in the South African Constitution. III. SOME NEW REMEDIAL FORMS
Let me offer somewhat more detailed comments on these new remedial forms. Drawing on a term imported into comparative constitutional law from an important contribution to Canadian constitutional scholarship by Peter Hogg and Alison Bushell, we can call them ‘dialogic’ remedies. So far they come in two variants. 5 In the literature of comparative constitutional law the widely celebrated Grootboom case in South Africa (Government of Republic of South Africa v Grootboom 2000 (11) BCLR 1169 (CC) (S Afr)) is the primary example of a planning remedy. 6 Here the primary example is Occupiers of 51 Olivia Road v City of Johannesburg 2008 (3) SA 208 (CC) (S Afr). 7 For the example usually given of such a failure, see Residents of Joe Slovo Community v Thubelisha Homes 2009 (9) BCLR 847 (CC) (S Afr). 8 See ch 21 in this volume, part I.
372 Mark Tushnet (a) Dialogue between courts and legislatures and executive officials. In the Grootboom case the South African Constitutional Court reviewed the constitutionality of the government’s programme to provide housing for those in need. Mrs Grootboom represented a group of squatters who the government evicted. Rather than challenging the evictions, Mrs Grootboom challenged the government’s overall programme for providing housing to the needy. The government did have such a programme but, the Court held, it was unconstitutionally defective because it did not contain a specific provision aimed at providing housing for those, such as Mrs Grootboom, in desperate need. The Court ordered the government to develop a plan for providing housing to that group. This order was at least implicitly dialogic because it left open the possibility that the lawyers representing Mrs Grootboom could return to Court to challenge the government’s plan as inadequate, and the Court would have to determine whether the plan was constitutionally adequate. Gross’s discussion of the provision of medical care shows how planning remedies fit easily into a constitutional scheme of judicial enforcement of social welfare rights. The health services ‘basket’ he discusses is the government’s plan for the distribution of medical resources, and modifications in the bucket at the urging of courts are no different in principle from other adjustments the government routinely makes based on new information. The dicta Gross cites from the Luzon case are typical of contemporary judicial approaches to the enforcement of social welfare rights. (b) Dialogue between executive officials and plaintiffs. The South African Constitutional Court has pioneered the development of another intriguing remedial possibility. The case again involved squatters: this time they were occupying a building that everyone agreed was unfit for human habitation because it lacked adequate drinking water and was quite vulnerable to fire. The city sought to evict the squatters, not in the service of some property owner’s interest in the building, but to protect the squatters themselves against these real risks. The Constitutional Court held that before going forward with the evictions the city had to engage in a discussion with the residents about whether they could develop some method of providing water and fire protection without evicting the residents.9 The ‘engagement’ resulted in an agreement by the city to leave the residents in place and provide them with water and emergency fire services. IV. THE ISRAELI CASE
The core characteristic of these dialogic remedial forms is that they do not impose financial costs directly or visibly, in the way that coercive enforcement of a social welfare right on behalf of a needy individual might. Gross’s comments suggest that the Israeli Supreme Court has failed to see the possibility of novel remedial forms. And, of course, Paz-Fuchs’s chapter can be read as an extended critique of the assumption that the only remedial form is the coercive order aimed at providing some material good to an identified individual. One outsider’s question about those comments and observations is this: why has the Court, so sophisticated about constitutional law in other dimensions, been seemingly so unsophisticated in this one? 9 Olivia Road (n 6). B Ray, ‘Engagement’s Possibilities and Limits as a Socioeconomic Rights Remedy’ (2010) 10 Washington University Global Studies Law Review 399–425, provides a good analysis of the case and subsequent ones.
Comment on Enforcing Social Welfare Rights 373
I suggest several possibilities, all touched upon in these chapters. First, constitutional language may matter. Drawing on standard international human rights terminology, the South African Constitution qualifies its guarantees of social welfare rights with the phrase ‘within available resources’. That qualification might well suggest to a court that it should search for effective remedies that nonetheless do not require the government to allocate ‘too many’ resources to satisfying constitutional requirements. In contrast, locating the protection of social welfare rights within a constitution’s protection of human dignity or the right to life may push the courts in the direction of preferring coercive remedies, because in most settings ‘human dignity’ and, to a lesser extent, the right to life, are rather unyielding against arguments that social reality requires some accommodation of the right. Courts would be quite uncomfortable in finding what they acknowledged to be torture – understood as a violation of human dignity – permissible in any but the most extraordinary circumstances (though they might strive to characterise the phenomena at issue as ‘harsh interrogation but not torture’).10 Second, but pointing in the other direction, a constitution whose interpretation centres on the doctrine of proportionality might be more open to courts’ finding that some social welfare right has been violated. The first step of proportionality analysis asks whether the impugned policy violates a constitutionally protected right. Such a finding is not in itself decisive or, indeed, consequential in many cases. The reason is that fiscal, separation of powers, and other concerns can be addressed in subsequent stages, when the courts examine whether a violation of rights is justified.11 Finally, social welfare rights may not be seen as free-standing, but either parasitic upon, or at least strongly related to, other constitutional rights. Equality concerns require what is inelegantly known in English as a ‘comparator’ – a class that receives the benefit denied to the claimant in the case at hand. Candidates for comparison include others within the scheme of social provision, such as those who receive treatment for their urgent conditions when the claimant does not receive treatment or her equally urgent condition; or those who are able to obtain the treatment by spending their own resources.12 Gross’s discussion of medical care notes how closely the issues are related to questions of equality. Ziv’s discussion of eligibility conditions strongly implies a connection between problematic conditions and intrusions on a constitutionally protected domain of privacy. I now turn to some additional issues. Proponents of constitutionalised social welfare rights typically assume that the material goods those rights guarantee must be provided by the state. That assumption seems to me to underlie the arguments by both Gross and Ziv. Provision by the state means a scheme of social provision whose adequacy must be assessed in constitutional terms.13 (Or at least so it seems to me). The availability of the 10 Consider here the Israeli Supreme Court’s position in HCJ 5100/94 Public Committee against Torture in Israel v State of Israel 53(4) PD 817 [1999] (in Hebrew), which can be understood as asserting that engaging in torture (in extreme cases) might be impermissible but not subject to legal penalty, or the German Constitutional Court’s decision that abortion, a violation of the right to life, must be condemned by law but may not be penalized under quite a few circumstances (88 BVerfGE 203 (1993) (Ger)). 11 For discussions of proportionality analysis, see, eg R Alexy, A Theory of Constitutional Rights (New York, Oxford University Press, 2002); A Barak, Proportionality (New York, Cambridge University Press, 2011). 12 As to the latter, see Chaoulli v Quebec (AG) [2005] 1 SCR 791 (Can), holding it unconstitutional to bar individuals from spending their own resources to obtain the medical care they believed necessary. 13 Of course social welfare rights are provided by the state in the definitional sense that they are rights inscribed in constitutional law, but that sense is not the one that interests me here.
374 Mark Tushnet private market as a comparator in equality-based cases suggests another possibility. Perhaps constitutionalizing social welfare rights has no significant conceptual implications for the distribution of the source of their provision as between market provision and social provision. That is, we might find universal agreement with the proposition that the constitution guarantees social welfare rights, but still face significant – and, in the jargon, reasonable – disagreement about how much of the provision of those rights should be done by the state, and how much by the market. The resolution of that disagreement will depend on ideology, politics, and history. Once the issue is seen in that light, what is at stake in all the cases Gross and Ziv discuss is the usual and quite ordinary question of constitutional review: what stance should the courts take, what standard of review should they apply, when faced with legislation that rests on a reasonable albeit not conclusive legislative or executive judgment about the best means by which social welfare rights can be guaranteed? One’s answer to that question should be the same as the answer one gives to the same question asked about regulations of freedom of expression, for example. So, Jeremy Waldron and I, court sceptics that we are, would tend to say that courts should adopt a standard akin to ‘mere rationality’ in US constitutional law for legislation that rests on a reasonable, though disputable, understanding of freedom of expression and legislation that rests on a similarly, reasonable though disputable, understanding of the question of distributing social welfare rights protection between market and social provision.14 Court enthusiasts will take a different view. But, as Ziv in particular emphasises, the question as to social welfare rights is the same as that as to civil and political rights. Given this analysis, I think that Ziv’s argument should be slightly reformulated. First, I do not think – and it is not clear to me that Ziv thinks – that anything important turns on whether we describe the rights in question as rights to resources or rights to reasonable access to resources. A right to reasonable access to resources is, I think, simply a description of the case when the legislature chooses market provision as the mechanism by which the right to the resource is guaranteed. Similarly, I would place less weight on the distinction between eligibility conditions that define the contours of the social welfare right and conditions that are in her sense limitations on a social welfare right (such as the rules challenged as discriminatory, and the like). The contours of the right are the boundaries the legislature has chosen to set between market provision and social provision. And, again, the standard of review should be whatever one thinks appropriate for cases of reasonable disagreement over legislative choices. V. CONCLUSION: EFFECTIVE ENFORCEMENT OF SOCIAL WELFARE RIGHTS?
I note one additional point in conclusion: some have questioned whether dialogic remedies are truly effective in moving resources to those truly in need.15 Examined as a whole, 14 See J Waldron, ‘The Core of the Case Against Judicial Review’ (2006) 115 Yale Law Journal 1346; M Tushnet, Taking the Constitution Away from the Courts (Princeton, Princeton University Press, 2000). 15 This criticism is particularly prominent in the literature on the South African cases, in part because the South African Constitutional Court explicitly rejected the proposition that there is some constitutionally protected ‘minimum core’ substantive right, and in part because the dialogic remedies seem to critics inadequately responsive to the social problems the nation faces (although the ‘within available resources’ language in the Constitution rather strongly suggests that effective provision of social welfare rights within available resources does not, as a constitutional matter, require provision to all in need).
Comment on Enforcing Social Welfare Rights 375
the cases in which courts have done something significant in enforcing social welfare rights tend to be cases that primarily benefit not the poorest in a nation but those who are relatively well-off, the middle classes rather than the poor. This is of course clearest in the ‘right to health’ cases involving the provision of specific drugs through individual direct actions.16 Statistics are hard to come by, but it seems clear enough that the primary beneficiaries of these actions are in the middle classes. And, the reason is clear enough as well: to bring such an action a person has to have a lawyer available to him or her, and in general the middle classes have substantially greater access to lawyers than do poorer people. But, the problem arises in connection with other forms of relief too. With respect to dialogic remedies, for example, most people can effectively engage with government officials only if they have some assistance, and, again, lawyers provide that assistance. Of course NGOs can provide assistance, by making lawyers and other resources useful in litigation over second-generation rights available to poor people. But, equally obviously, NGOs are underfunded, available only spottily, and in general cannot be expected to be widely enough available so as to offset the middle class bias of the remedial forms taken as a whole. The US political scientist Charles Epp has provided a systematic explanation for these outcomes. Examining what he calls efforts at rights revolutions in a number of countries, including India, Epp concludes that such revolutions succeed only when they are accompanied by what he calls a ‘support structure’, a set of institutions that supply those seeking constitutional rights with regular access to lawyers, funds to keep litigation going, and favourable publicity.17 Support structures include NGOs, but, intriguingly, Epp finds that some governments provide the necessary resources, even when the lawyers are attacking government policies. Nonetheless, sustaining a support structure is difficult, and national variations will matter a great deal. Future scholarship on the judicial enforcement of second-generation rights will, I think, focus on the question of how support structures can be created and sustained. Perhaps the most important conclusion I draw from these reflections is a much simpler one. The issues scholars in the field need to consider are substantially different from the ones that have preoccupied scholarship on the judicial enforcement of second- generation rights for the past decades. As these articles indicate, a new scholarly agenda is on the table. As I see it, that agenda will move away from concern for separation of powers and fiscal concerns, and toward an analysis of the ideological commitments displayed in discourse about social welfare rights. In addition, it will examine in detail the interactions among individual litigants, the support structures of NGOs, and the courts. The new agenda will, I think, enhance both the scholarly interest in, and excitement about, the constitutionalization of social welfare rights.
That appears to be the conclusion of most studies of the ‘right to medication’ cases in Brazil. CR Epp, The Rights Revolution: Lawyers, Activists, and Supreme Courts in Comparative Perspective (Chicago, University of Chicago Press, 1998). 16 17
25 Constitutional Rights and Private Law* AHARON BARAK
I. THE PROBLEM PRESENTED
A
LL ARE IN agreement that the individual’s constitutional rights1 are directed towards the state. The history of human rights is that of the recognition of individual rights vis-à-vis the state. The texts of the different constitutions have turned this history into a reality. Often, explicit provisions are found in the constitution, according to which the state must respect the rights of individuals determined therein.2 The question before us is: are constitutional rights directed at the state (vertical relationship) alone or are these constitutional rights directed at individuals as well (horizontal relationship)? Is an employee’s (just like any other individual’s) constitutional right to freedom of expression, directed at the state, also an employee’s constitutional right to freedom of expression vis-à-vis the employer? This question is recognised in comparative constitutional law as a one regarding third party’s effects (Drittwirkung). The first party is the individual, the state is the second party, and the third party is made up of another individual. The question is: are constitutional rights, granted by the constitution to the first party (the individual) directed towards the second party (the state) alone or also vis-à-vis the third party (another individual)? The literature on this subject is vast.3 * This chapter is based on my past writing on this subject as well as an additional development of this train of thought. For previous publications, see A Barak, ‘Constitutional Human Rights and Private Law’ in A M Rabello and P Sarcevic (eds), Freedom of Contract and Constitutional Law 105 (Hebrew University of Jerusalem, 1998); A Barak, ‘Constitutional Human Rights and Private Law’ in D Friedmann and D Barak-Erez (eds), Human Rights in Private Law (Oxford, Hart Publishing, 2001) 13; A Barak, ‘Constitutional Human Rights and Private Law’ (1996) 3 Review of Constitutional Studies 218. 1 The term ‘constitutional rights’ is used in this chapter to describe rights expressly or impliedly guaranteed by the constitution. 2 See Israeli Basic Law: Human Dignity and Liberty, s 11. English translation available at: www.knesset.gov. il/laws/special/eng/basic3_eng.htm; Israeli Basic Law: Freedom of Occupation, s 5. English translation available at: www.knesset.gov.il/laws/special/eng/basic4_eng.htm. See also Grundgesetz für die Bundesrepublik Deutschland [Grundgesetz][GG][Basic Law], 23 May 1949, BGBl I, art 1(3) (Ger); South African Constitution, 1996, s 8(1). 3 In the last 10 years there have been many articles published on this subject. The important books published during that period are: AM Rabello and P Sercevic (eds), Freedom of Contracts and Constitutional Law (Jerusalem, Hebrew University of Jerusalem, 1998); R English and P Havers (eds), An Introduction to Human Rights and the Common Law (Oxford, Hart Publishing, 2000); D Friedmann and D Barak-Erez (eds), Human Rights in Private Law (Oxford, Hart Publishing, 2001); A Sajó and R Uitz (eds), The Constitution in Private Relations: Expanding Constitutionalism (Hague, Eleven International Publishing, 2005); T Barkhuysen and SD Lindenbergh (eds), Constitutionalisation of Private Law (Leiden, Martinus Neihoff Publishers, 2006); KS Ziegler (ed), Human Rights and Private Law: Privacy as Autonomy (Oxford, Hart Publishing, 2006); D Oliver and J Fedtke (eds), Human Rights and the Private Sphere: A Comparative Study (London, Routledge-Cavendish,
380 Aharon Barak This dilemma appears to be resolved when the constitution states (expressly or impliedly) a clear stand. In some constitutions there is an express reference to the horizontal effect. Thus, for example, in the Constitution of South Africa there is an explicit provision, that reads as follows:4 A provision of the Bill of Rights binds a natural or a juristic person if, and to the extent that, it is applicable, taking into account the nature of the right and the nature of any duty imposed by the right.
With this provision in mind, seemingly the question in South Africa is the scope of the provision and the circumstances where its conditions are met.5 This conclusion is not free from doubts,6 as the Constitutional Court’s rulings have determined that the application is indirect.7 In a small number of constitutions there are provisions regarding the horizontal application of special constitutional rights.8 The United States Bill of Rights is drafted in a manner that precludes the horizontal effect on most rights. Thus, for example, the First Amendment states:9 Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the Government for a redress of grievances.
This provision is directed at Congress, and burdens it with obligations. This provision has only a vertical effect. Under the assumption that a constitution is not expressly restricted to vertical effect, the question before us is if the constitutional rights apply to horizontal relationships as well. The answer to this question is interpretive.10 What are the options available to the interpreter?11 In principle, one can think of four models: (1) direct application model; (2) non-application model; (3) application to the judiciary model; and (4) indirect application model. In part II those four models will be discussed, and my preference to the fourth model (indirect application) will be explained. Part III will discuss the application of the fourth model in detail generally. Part IV will deal with its application in Israel. Part V will make some general conclusionary remark about the indirect application model and its relationship to the direct application model. 2007); C Mak, Fundamental Rights in European Contract Law: A Comparison of the Impact of Fundamental Rights on Contractual Relationships in Germany, the Netherlands, Italy and England (Alphen an den Rijn, Kluwer Law International, 2008); D Hoffman (ed), The Impact of the UK Human Rights Act on Private Law (Cambridge, Cambridge University Press, 2011). 4 South African Constitution, 1996, s 8(2). 5 See S Woolman, ‘Application’ in S Woolman and M Bishop (eds), Constitutional Law of South Africa, 2nd edn (Pretoria, Juta Law, 2005) ch 31. 6 See H Cheadle, ‘Third Party Effect in the South African Constitution’ in A Sajó and R Uitz (eds), The Constitution in Private Relations: Expanding Constitutionalism (Hague, Eleven International Publishing, 2005) 55. 7 Du Plessis v De Klerk 1996 (3) SA 850 (CC) (S Afr); Bernstein v Bester NO 1996 (2) SA 751 (CC) (S Afr); Carmichele v Minister of Safety and Security 2001 (4) SA 938 (CC) (S Afr); Khumalo v Holomisa 2002 (5) SA 401 (CC) (S Afr); Rail Commuters Action Group v Transnet Ltd 2005 (2) SA 359 (CC) (S Afr). 8 See US Constitution, amend XIII, which bans slavery. This ban applies to interpersonal relationships. See also the Grundgesetz GG (n 2), Art 9(3) which bans agreements between individuals which limit the freedom of association. 9 US Constitution, amend I. 10 See A Barak, Purposive Interpretation in Law (Princeton, Princeton University Press, 2005) 370. 11 See PW Hogg, Constitutional Law of Canada, vol 2, 5th edn (Toronto, Thomson Carswell, 2007) 102.
Constitutional Rights and Private Law 381
II. THE FOUR MODELS
A. The Direct Application Model The first model available to the interpreter is to interpret the constitution’s provisions regarding human rights as being directed at both the state and individuals.12 According to this approach, constitutional rights directly apply to the relationships between individuals. Thus, for example, according to this option, the right to dignity and liberty found in Basic Law: Human Dignity and Liberty also applies to the relationships between individuals. According to this approach, the Basic Law recognises the employer’s con stitutional duty not to limit the employee’s human dignity and the employee’s constitutional right not to limit the employer’s human dignity. Similarly, the freedom of expression anchored in the constitution is not just the freedom of the individual vis-à-vis the state but also vis-à-vis every other individual. Two considerations support this option: first, in the modern reality, the danger to constitutional rights is not only from the state but also from other individuals. At times, these have powers no less than those of the state. Constitutional rights should therefore apply to horizontal relationships as well.13 Second, we are faced with an expansive privatization process.14 State powers, whose use by the governmental authority must respect the constitutional rights of the individual vis-à-vis the state, are transferred over to the private sector. If within this sector the constitutional rights do not continue to apply, the result will be that the privatization will release the state of its constitutional duties without replacing it by imposing such duties on the private sector. Despite the potency of these considerations, I believe that this model is not right.15 The reasons brought for its defence do not support it. Regarding the first consideration (the power of certain individuals as the state) – I believe that the individual’s protection vis-à-vis another individual, no matter how powerful they may be, should be resolved in private law (statute or common law) itself. There is no justification in granting to a weaker individual a constitutional right vis-à-vis a stronger individual. The granting of constitutional status to a right is equal to the protection of said right from the state, a body that can legislate statutes or develop the common law and therefore limit the right. The granting of constitutional status to a right imposes limitations on the state to affect the right. This result is irrelevant to the protection of the weaker individual’s rights visà-vis the stronger individual. As to the relationship between the weaker and stronger 12 For this model, see A Barak, Interpretation in Law: Constitutional Interpretation, vol 3 (Jerusalem, Nevo Publishing, 1994) (in Hebrew). In German literature this approach is referred to as ‘unmittelbare Drittwirkung’. For an analysis of this approach, see R Alexy, A Theory of Constitutional Rights (Oxford, Oxford University Press, 2002) 351; M Kumm and VF Comella, ‘What is So Special About Constitutional Rights in Private Litigation? A Comparative Analysis of the Function of State Action Requirement and Indirect Horizontal Effect’ in, The Constitution in Private Relations: Expanding Constitutionalism (n 3) 241. 13 See Barak, Interpretation in Law (n 12) 656. 14 See I Zamir, Administrative Power, vol 1, 2nd edn (Jerusalem, Nevo Publishing, 2010) 39 (in Hebrew); D Barak-Erez, Administrative Law, vol 1 (Tel Aviv, Israel Bar Association, 2010) 59 (in Hebrew). 15 It is possible that it may be justified in special cases, when the legal system wishes to put up a red flag regarding a certain kind of inappropriate activity. The Indian Constitution grasps as a duty not to discriminate against an obligation as a duty imposed not only vis-à-vis the state but also in interpersonal relationships: see Indian Constitution, Art 15. See also aforementioned US Constitution, amend XIII, which bans slavery. This ban applies to interpersonal relationships.
382 Aharon Barak individual, sub-constitutional law is sufficient. The stronger individual has no legislative power and the Limitation Clause – which applies when the constitutional right is limited by a sub-constitutional norm16 – does not apply. Of course, to the extent that said subconstitutional law does not sufficiently consider the weakness of one side, particularly due to their chronic weakness, the road is open to consider its constitutionality. Indeed, disproportional protection by private law of the weak from the strong may make private law unconstitutional. As such, the weaker party’s constitutional right vis-à-vis the state is sufficient. There is no need to recognise this right as a constitutional right vis-à-vis the stronger side. The vertical effect is characterised by the constitutional right of the individual vis-à-vis the state. The state has no constitutional right vis-à-vis the individual. The second consideration (privatization which results in a release from constitutional obligations) may be relevant when considering the constitutionality of the privatization.17 Thus, for example, the privatization of prisons was declared in Israel unconstitutional due to its disproportional limitation of the prisoner’s right vis-à-vis the state.18 When privatization is constitutional, the law that applies to the relationship between the individuals after the privatization should be the sub-constitutional law, which has been made suitable for privatization.19 It is not fitting that the vertical relationship which applied vis-à-vis the state before the privatization continues to apply, vis-à-vis other individuals after privatization. The direct application model is methodologically problematic. Its application is unsatisfactory. Methodologically when two principle-shaped constitutional rights conflict – whether it be the individual’s constitutional right vis-à-vis a state or one individual’s constitutional right vis-à-vis another individual (according to the direct application model) – the conflict’s solution is not found at the constitutional level.20 The scope of the constitutional right does not change. The conflict’s solution is at the sub-constitutional level, such as at the legislative or common law level. On this level we determine if the law limiting a constitutional right to realise another constitutional right is constitutional. Therefore, even if direct application is recognised, the conflict’s solution regarding the scope of the rights will not be at the constitutional level but rather at the sub-constitutional level, namely in providing the answer to the question of if the limiting law is constitutional. This level must of course fulfill the requirements of the Limitation Clause. Thus, in Israel the limitation of the constitutional right must be ‘to an extent not greater than is required’,21 that is – proportional. Therefore, the recognition of the direct application (horizontal) model is methodologically wrong. The logic at the foundation of the need to recognise the individual’s 16 A typical Limitation Clause is found in s 1 of the Canadian Charter of Rights and Freedoms. For Israel, see Basic Law: Human Dignity and Liberty, s 8. 17 Relevant to this matter is the approach that the Parliament cannot release itself from its duty to determine primary arrangements (the non-delegable duty): HCJ 3267/97 Rubinstein v Minister of Defence 52(5) PD 481, 502 [1998] (in Hebrew); HCJ 11163/03 Supreme Monitoring Committee for Arab Affairs in Israel v Prime Minister )27 February 2006), Nevo Legal Database (by subscription) (in Hebrew); G Sapir, ‘Primary Arrangements’ (2010) 32 Tel-Aviv University Law Review 5 (in Hebrew). 18 See HCJ 2605/05 Academic Center of Law and Business v Minister of Finance (19 November 2009), Nevo Legal Database (by subscription) (in Hebrew). 19 See A Harel, Private Bodies in Administrative Law (Tel-Aviv, Israel Bar Association, 2008) (in Hebrew); D Barak-Erez, ‘Public Corporations’ (1995) 19 Tel-Aviv University Law Review 273 (in Hebrew). 20 See A Barak, Proportionality: Constitutional Rights and their Limitations (Cambridge, Cambridge University Press, 2012) 83. 21 Basic law: Human Dignity and Liberty, s 8.
Constitutional Rights and Private Law 383
constitutional right vis-à-vis the state does not exist regarding the recognition of the constitutional right of one individual vis-à-vis another.22 B. The Non-Application Model The second model provides that constitutional rights apply only vis-à-vis the state (vertical relationship). A constitution is meant to protect the individual from the state and is not aimed at dealing with interpersonal relationships (horizontal), a relationship which has always been regulated by private law. According to this option, anything concerning interpersonal relationships should be regulated in private law, without any influence or penetration of the constitution’s provisions, all of which are part of the public sphere. Of course, the boundary between the public law and private law is not so clear nor is it impenetrable. There are reciprocal ties between public and private law. Therefore, when developing legal doctrines regarding private law, the judge will consider the public law. This consideration reflects the need to observe the entire structure of society, law and the legal system. However, according to the second model, this consideration by the judge is not based on the application – direct or indirect – of the constitution’s provisions regarding human rights in interpersonal relationships. Canadian law essentially takes23 the second option.24 The latter was determined in the Dolphin Delivery case.25 In that case the labour union organised a strike against the employer. The Dolphin Delivery Company was not the employer, but rather supplied products to the employer. Despite this, the labour union decided to picket Dolphin Delivery. The latter applied to the Court to request an injunction, which prevented the labour union from committing a civil wrong of inducing breach of contract. In court the labour union claimed that it should not be handed down an injunction as it would limit its right to freedom of expression which is protected in the Canadian Charter of Rights and Freedoms. The Supreme Court refuted this claim: it ruled that the constitutional rights determined in the Canadian Charter of Rights and Freedoms were directed towards the state and not other individuals. The non-application model is fitting to the extent that it rejects the direct application model. However, it creates too deep a rift between constitutional law and private law. There are closer ties between constitutional law and private law than those presented by the non-application model. At the core of these ties are the duties of private law – and the duty of the legislator and the judges who create and develop private law – to fulfill the constitutional values and principles.26
22 Alexy (n 12) 362: ‘direct horizontal effect cannot mean that rights of the citizens against the state are also rights of the citizen against other citizens . . . direct horizontal effect cannot be achieved simply by exchanging the addressee of the right of a citizen against the state’. 23 It is also possible to see weakened indirect application in it: see S Gardbaum, ‘The “Horizontal Effect” of Constitutional Rights’ (2003) 102 Michigan Law Review 388. 24 See Hogg (n 11) 101. See also Pepsi-Cola Canada Beverages v RWDSU [2002] 1 SCR 156 (Can); Dagenais v CBC [1994] 3 SCR 835 (Can). 25 RWDSU v Dolphin Delivery Ltd [1986] 2 SCR 573 (Can). 26 For this duty, see n 93 below.
384 Aharon Barak C. The Application through the Judiciary Model The third model is the application through the judiciary.27 The starting point is that constitutional rights are directed towards the state alone. However, the judge is a state organ, and therefore the constitutional right is directed towards him. According to this model when the judge speaks, the state speaks; when the judge acts the state acts. The judge is the state and human rights need to be protected vis-à-vis him as well. The meaning of the latter, from the judge’s point of view, is twofold: first, he must develop the common law in a fashion that conforms to his duty not to disproportionally limit constitutional rights. Secondly, he must act within the confines of the specific dispute in a fashion which conforms to the parties’ constitutional rights vis-à-vis the state. This model was developed in American constitutional law, wherein constitutional rights are directed towards the state. In the New York Times Co v Sullivan case28 it was triggered regarding the development of the rules of defamation in tort law. In this case, the discussion concerned a claim of defamation filed by Sullivan (a public figure) against the New York Times. It was decided that the court must develop the common law of defamation according to the First Amendment. In the Shelley v Kraemer case29 it was applied to prevent discriminatory remedies in individual disputes. Here, the Shelley family (an African-American family) had purchased a home. Kraemer lived in that same neighbourhood. He applied to the court in order to prevent the Shelley family from gaining possession of the home as the latter would have contradicted a restrictive covenant applicable on the house, according to which a ‘Negro’ could not purchase the home. The Supreme Court ruled that a judge cannot provide an injunction which would limit the principle of equality found in the Fourteenth Amendment. The third model – so long as it is regarding an individual dispute – is wrong.30 Despite this, the third model is correct when it concerns the common law. According to my point of view, the common law is subject to the general constitutional structure.31 A common law precedent may lawfully limit a constitutional right. To do so, it must fulfill the requirements of the Limitation Clause – it must be proportional.32 Therefore, I believe that the New York Times Co v Sullivan case33 was rightly decided. This is not the case in Shelley v Kraemer.34 When the time comes for the judge to grant an injunction, he must act according to the rules of the common law which is the legal framework where the rules of the restrictive covenant and its enforcement were developed. If these rules allow for the non-granting of a discriminatory injunction – he must act according to these rules. If these rules determine that there is room to grant an injunction order, even if it is discriminatory – he has the duty to lead to a change in these rules (only if he is authorised to do so according to the judicial hierarchy (for example if the matter is brought before the Supreme Court) or according to the rules of judicial power (in the federal law
For this model, see Gardbaum (n 23). New York Times Co v Sullivan 376 US 254 (1964). 29 Shelley v Kraemer 334 US 1 (1948). 30 See the Dolphin Delivery case (n 25) paras 33–35 of Mcintyre J’s opinion. 31 See Barak, Proportionality (n 20) 118. See also n 93 below. 32 Barak, Proportionality (n 20) 121. 33 Sullivan (n 28). 34 Kraemer (n 29). 27 28
Constitutional Rights and Private Law 385
system)). The rules of the common law cannot limit constitutional rights, unless they fulfill the requirements imposed by the legislation limiting the constitutional rights. So long as the common law is not changed – he must act according to the rules of the common law. If not, the constitution’s provisions, which see the right to equality as a right vis-à-vis the state alone and not vis-à-vis other individuals, becomes suddenly a right to equality vis-à-vis another individual. The judge is not empowered to do so. If a constitution’s provision regarding the right to equality applies vis-à-vis the state alone, and not in the relationship between individuals, there is no room for the conclusion that when a dispute between individuals comes to court, the judge must determine the dispute in an equal manner. This conclusion voids the constitutional provision according to which the right to equality applies to the relationships between the individual and the state only. Indeed, Shelley v Kraemer’s ruling was not sustained. Instead, the courts developed the doctrine of state action.35 The application of the constitutional rights included in the Bill of Rights vis-à-vis the judge are conditional on that individual, who according to the claim, limits the constitutional rights in his relationship with others and carries out functions which are traditionally the state’s. A significant amount of case law has been dedicated to this matter and has determined when the nexus between the state and the individual is sufficiently close, so close that it is possible to attribute the behaviour to the state. This doctrine overrules, albeit not explicitly, the ruling in Shelley v Kraemer. In light of the development of this doctrine, American law should no longer be seen as a system which adopts the model of the application through the judiciary. There are those who believe that the American approach adopts the indirect application.36 D. The Indirect Application Model The fourth model is that of indirect application (unmittelbare Drittwirkung).37 According to this model, the individual’s constitutional rights apply directly vis-à-vis the state and indirectly vis-à-vis an individual. The individual should not fill the shoes of the state, which possesses the duty concerning constitutional rights. An individual does not have constitutional rights vis-à-vis another individual. The rights between the individuals are at the sub-constitutional level (statute or common law) and not at the constitutional level. Indeed, ‘the state’s duties to uphold human rights are not the same content-wise as the individual’s duty to uphold human rights’.38 ‘The duty of the state to act equally visà-vis the individual is not the same content-wise as the individual’s duty to act equally vis-à-vis another individual’.39 The individual’s freedom of occupation vis-à-vis the state is different in scope than the freedom of occupation of the individual vis-à-vis another individual. However, the individual’s constitutional rights vis-à-vis the state apply indirectly to the relationships between individuals. This indirect application is based on the 35 See E Chemerinsky, Constitutional Law: Principles and Policies, 3rd edn (New York, Aspen Law & Business, 2006). 36 See Gardbaum (n 23). 37 For an analysis of this approach, see Alexy (n 12) 351; Kumm and Comella (n 12) 246. 38 CA 294/91 Jerusalem Community Burial Society v Kestenbaum 46(2) PD 464, 531 [1992] (in Hebrew) (Barak J’s opinion). 39 HCJFH 4191/97 Recanat v National Labour Court 54(5) PD 330, 362 [2000] (in Hebrew) (Court President Barak’s opinion).
386 Aharon Barak assumption that each of the individual’s constitutional rights vis-à-vis the state have an objective value aspect which is projected itself onto all of the law’s branches. This means that the individual’s rights vis-à-vis the state to dignity, freedom, privacy and property express objective values (but not rights) which shape the relationships between individuals within private law. These values are used for the interpretation of the private law, to complete lacunae in it or develop it. To the extent that the existing private law limits ‘negative’ constitutional rights of the individual vis-à-vis the state, such law (statute or the common law) is constitutional only if it is proportional. To the extent that such existing law does not protect proportionally the ‘positive’ individual rights vis-à-vis the state, the state must develop the private law to make it proportional.40 From the four models I discussed, this model is the proper one. It is accepted in most legal systems where this problem has arisen.41 It applies within the framework of the European Convention on Human Rights.42 Amongst the legal systems which draw on the indirect application are Germany,43 Italy,44 the Netherlands45 and New Zealand.46 The indirect application model recognises the difference between the individual’s constitutional right directed towards the state and the individual’s sub-constitutional (statute or common law) right directed towards another individual. This difference is expressed in the normative status of the right and its scope. Regarding the normative status – while the individual’s right vis-à-vis the state is a right at the constitutional level, the individual’s rights vis-à-vis another individual is at the sub-constitutional level. It is mostly found in private law. As for the scope – the individual’s constitutional right vis-à-vis the state is more expansive than the individual’s right vis-à-vis another individual. There are two reasons for this: first, the objectives of the constitutional right are at full force vis-à-vis the state. Second, in determining the scope of the individual’s constitutional right vis-àvis the state, one should consider the rationale at the right’s basis. No balance should be made on the constitutional level between it and the considerations which justify the right’s limitation (ie the public interest or rights of others). Balances should be made only when determining the constitutionality of the limitation of the constitutional right by a sub-constitutional law.47 With regards to the determination of the scope of the
Regarding this obligation see below, text at section IIIDv. See A Barak, ‘Constitutional Human Rights and Private Law’ in D Friedmann and D Barak-Erez (eds), Human Rights in Private Law (Oxford, Hart Publishing, 2001) 13. 42 Convention for the Protection of Human Rights and Fundamental Freedoms (European Convention on Human Rights, as amended) (ECHR). See C Ovey and RCA White (eds), Jacobs and White: European Convention on Human Rights, 4th edn (Oxford, Oxford University Press, 2002) 31; EA Alkema, ‘The ThirdParty Application of “Drittwirkung” of the European Convention on Human Rights’ in GJ Wiarda, F Matscher and H Petzold (eds), Protecting Human Rights: The European Dimension: Studies in Honour of Gerard J Wiarda (Köln, C Heymann Publishers, 1989) 33; RS Kay, ‘The European Convention on Human Rights and the Control of Private Law’ (2005) 5 European Human Rights Law Review 466 ; A Clapham, ‘The “Drittwirkung” of the Convention’ in Macdonald, Matscher and Petzold (eds), The European System for the Protection of Human Rights (New York, Springer, 1993) 163. For the approach in international law, see JH Knox, ‘Horizontal Human Rights Law’ (2008) 102 American Journal of International Law 1. 43 See Mak (n 3) 57. 44 ibid 104. 45 ibid 83. 46 See Hosking v Runting [2004] 7 HRNZ 301 (NZ). For an analysis of the case law, see A Geddis, ‘The Horizontal Effects of the New Zealand Bill of Rights Act, as Applied in Hosking v Runting’ (2004) New Zealand Law Review 681; P Rishworth, G Huscroft, S Optican and R Mahoney, The New Zealand Bill of Rights (Oxford, Oxford University Press, 2003) 102. 47 See Barak, Proportionality (n 20) 80. 40 41
Constitutional Rights and Private Law 387
individual’s right vis-à-vis another individual, the scope of the right is determined by the sub-constitutional law (such as a statute or common law) and is a product of the balancing between the rationale at the rights basis and the rationale at the basis of the constitutional rights of others or the public interest. The constitutionality of this balance is determined by the Limitation Clause that is, this sub-constitutional law is constitutional only if it is proportional. The indirect application model does not blur the distinction between the individual’s constitutional right vis-à-vis the state and the individual’s right vis-à-vis another individual. According to this model, the rights which exist between individuals are not constitutional rights. They are not a part of the constitution, but rather, they are part of the sub-constitutional law. They are anchored in regular legislation or the common law (such as contract, tort, property and labour law). The rights and remedies for their breach have been shaped by private law over hundreds of years. In this shaping there is significant weight given to the fact that the constitutional law recognises the existence of these rights vis-à-vis the state (vertical relationship). The existence of this relationship affects directly the constitutionality of this sub-constitutional law, and it affects the rights of individuals vis-à-vis the state. Thus sub-constitutional law affects indirectly the rights between individuals (horizontal relationship). This influence is best expressed in the interpretation of the norms in private law, in the completion of the lacunae within it, and the development of new and proportional private law. Hence the difference between this approach and the non-application model. The non-application model creates a barrier between the constitutional rights and private law. This barrier is cracked only when the lack of the non-application model determines that the development of the common law is affected by constitutional rights. The indirect application model takes a much stronger position. The influence of the constitutional rights on common law is extensive and much stronger in the indirect application model than in the non-application model. Private law, made by statute or common law,48 infringes on the constitutional right of each individual vis-à-vis the state, and therefore is subject to the Limitation Clause. The relationship between the indirect application model and the application through the judicial model are complex. On the one hand, both models provide that the constitutional rights apply only vertically. According to both models, the legislator and the judge must act to suit legislation and common law to constitutional rights. For this reason the decision in the New York Times Co v Sullivan49 case befits both models. On the other hand, the indirect model option sees the constitutional rights as objective values working within private law and according to it. This is not true regarding the application to the judiciary. The state action doctrine, as it has been developed in the United States, operates outside of private law. Its operation is within public law, and the constitutional right is directed not only towards the state, but also vis-à-vis individuals carrying out the state’s functions.
ibid 118. Sullivan (n 28).
48 49
388 Aharon Barak
III. THE INDIRECT APPLICATION MODELS IN ACTION
A. Constitutional Rights as Objective Values The starting off point of the indirect application model is in the perception that constitutional rights have both a subjective and objective aspect.50 The subjective aspect is the duty that they impose on the state vis-à-vis the individual. Alongside the subjective aspect there is an objective aspect (objektive Wertordnung). Every subjective constitutional right reflects an objective value. These objective values operate on all of the law’s branches. Thus, for example, human dignity has two elements – subjective and objective. The subjective element is expressed in the constitutional right to human dignity granted to the individual vis-à-vis the state. The objective element is expressed by the fact that human dignity is an objective value of the legal system in general and of private law in particular. This differentiation between the two elements of constitutional rights has been expressed in the Lüth case given by the German Constitutional Court.51 The case examined the legality of Lüth’s behaviour, who asked the public to boycott a film directed by a renowned Nazi. The movie’s producer applied to the Civil Court to prevent Lüth from demanding the film’s boycott. The Civil Court ruled that Lüth’s behaviour was a civil wrong, as it fell within the boundaries of section 826 of the German Civil Code (BGB) which sets out: A person who, in a manner contrary to public policy, intentionally inflicts damage on another person is liable to the other person to make compensation for the damage.
Lüth applied to the Constitutional Court. He claimed that the ruling limited his constitutional right to freedom of expression. The Constitutional Court ruled that Lüth’s constitutional right to freedom of expression was an objective value of the legal system. The term ‘public policy’ found in section 826 of the BGB should be interpreted in a manner consistent with this objective value. Considering the case’s circumstances, Lüth’s behaviour should not be considered as offensive to ‘public policy’. Lüth’s constitutional right to freedom of expression was directed towards the state alone. Lüth had no constitutional right to freedom of expression vis-à-vis the film’s producer. However, in the relationship between Lüth and the producer the objective value of freedom of expression did apply. This value influences the meaning of the private law’s provisions on ‘good morals’ and other ‘valve concepts’ of private law. We will now examine the different channels through which the objective aspects of the constitutional right affects private law (statute or the common law). I will discuss three main channels: (1) the interpretation of the private law; (2) referring to the gaps and the negative arrangement in private law; and (3) developing private law.
50 The use of the terms ‘subjective’ and ‘objective’ in this context was taken from German constitutional theory. The subject aspect regards the individual’s right vis-à-vis the state. The objective aspect is regarding the constitutional value which applies to all relationships in the law. 51 7 BVerfGE 198 (1958) (Ger).
Constitutional Rights and Private Law 389
B. First Channel of Application: Interpreting Private Law i. The Objective Aspects of Constitutional Rights and the Interpretation of Private Law The indirect application model assumes that constitutional rights apply only to the relationship between the individual and the state. However, each of the individual’s constitutional rights vis-à-vis the state expresses one of the legal system’s objective values. This objective value is not limited to the relationship between the individual and the state. It applies to all legal fields, and all legal relationships, including the relationship between individuals. It applies to private law.52 These objective value elements – derived from the individual’s constitutional rights vis-à-vis the state – have significant importance in private law. It is expressed in the interpretation of statutory private law. This interpretation in Israel is purposive interpretation.53 According to this interpretation, the statute text is given the interpretation which fulfills its purposes. This purpose, on a high level of abstraction, includes the fundamental values of the legal system.54 These make up ‘a normative umbrella spread out above all legislation’.55 The latter expresses the fact that the law is ‘a living creature within its environment’.56 The law’s environment includes ‘not only the legislative context in proximity, but also extended circles of accepted principles, fundamental purposes and basic standards’.57 These fundamental principles include the objective values at the foundation of the constitutional rights. These values penetrate into private law through purposive interpretation. Therefore, the rules of contract, tort or property found within private law should be interpreted in a manner which fulfills human dignity, liberty, property and the other objective values derived from constitutional rights. Like all fundamental principles which shape statutory purposes, they are in constant conflict.58 The objective aspect of one constitutional right (freedom of expression) conflicts with the objective aspect of a different constitutional right (privacy). Sometimes the objective aspect of the constitutional right conflicts with objective principles like public interest, such as security, public order and health. This conflict is not solved by the ‘victory’ of one objective aspect and the ‘loss’ of another or that of the public interest. The solution to this conflict within the legislative purpose is through balancing between the conflicting values according to their importance at the decision point. This is an interpretive balancing.59 Interpretive analogy can be used for the needs of this balancing from the element of proportionality stricto sensu from the rules of constitutional
CA 6601/96 AES System Inc v Saar 54(3) PD 850,860 [2000] (in Hebrew). See Barak, Purposive Interpretation (n 10) 370. 54 ibid 381. 55 See CA 165/82 Kibbutz Hatzor v Assessing Officer 39(2) PD 70, 75 [1985] (in Hebrew) (Barak J’s opinion). See also CA 2000/97 Lindorn v Carnit – Road Accident Victims Fund 55(1) PD 12, 29 [1999] (in Hebrew): ‘Alongside the specific purpose at the foundation of the defendant’s right is the general purpose. This is an objective purpose. It reflects the core values of the system . . . it is an expression of the understanding that legislation is not a one-off act of a passing legislator, but rather a link of legislative creation of a permanent legislator. It makes up a ‘normative umbrella’ spread above all of the legislation . . . This general purpose “trickles” into the purpose of all legislation. It is an expression of the structure of the system and its values’ (Court President Barak). 56 HCJ 58/68 Shalit v Minister of the Interior 23(2) PD 477, 513 [1970] (in Hebrew). 57 Kibbutz Hatzor (n 55). 58 See Barak, Purposive Interpretation (n 10) 153. 59 Regarding the interpretive balancing, see Barak, Proportionality (n 20) 72. 52 53
390 Aharon Barak proportionality.60 However, the Limitation Clause has no application. We are dealing here with a balancing for purposive interpretation and not for constitutional validity. ii. The Interpretation of Valve Concepts a. Interpretation in Light of the Objective Aspects of Constitutional Rights A special case of the interpretation of the private law exists when private law uses valve concepts (ventilbegriffe; conceffi volvola). Examples of these are good faith, public policy or negligence. These are shaped as principles (to distinguish from rules).61 They reflect society’s values and its basic principles. They provide the legal system with the appropriate flexibility. With their help, the legal system conforms to changing needs, yet still remains true to itself. Both legislation and the common law use these valve concepts to prevent stagnation within the law and to match up the law to changing conditions. The interpretation of the valve concepts which apply in private law is done on the background of the basic values of the legal system. These include the objective aspects of the constitutional rights to human dignity, freedom, property, privacy, freedom of movement and occupation. Indeed private law’s valve concepts take on a meaning which is influenced by the objective aspect of constitutional rights. This objective aspect has a ‘radiating effect’ on the valve concepts.62 The valve concept’s content is determined through the interpretive balancing between all the values which express it, and, within them, the objective aspect of the subjective constitutional rights. I will demonstrate this through two of private law’s important valve concepts: public policy and good faith. b. Public Policy One of the most important valve concepts is public policy. It operates in both public and private law alike. Through this concept ‘the legal system ensures the constant flow of fundamental values into the private law. This ensures harmony in the legal system. Through “public policy” the legal system ensures the proper behavior in interpersonal relationships’.63 The main components of public policy are the values which make up the objective aspect of constitutional rights. Public policy is the fulfilling of values such as human dignity, freedom, privacy, property and freedom of occupation. Needless to say, alongside these values are additional values which reflect the public interest. These values are in conflict. Often these basic principles conflict with one another . . . One’s freedom of occupation limits another’s freedom of contract. One’s freedom of expression limits another’s human dignity. This is particularly obvious within the framework of ‘public policy’ in contract law, as one of the basic principles of our legal system is the principles regarding freedom of contracts . . . and one of the fundamental rights of man that should be considered is that of the autonomy of the individual’s will. This is part of human dignity and liberty. Contractual relationship is an expression of this autonomy. Freedom of contracts is a constitutional right. When the con ibid 75. Regarding the differentiation between principles and rules in this context, see ibid 70. 62 Regarding the ‘doctrine of radiation’ see Alexy (n 12) 352. 63 Recanat (n 39) 363 (Court President Barak’s opinion). 60 61
Constitutional Rights and Private Law 391
tract’s content limits freedom of occupation, human dignity, freedom of movement, freedom of conscience and religion or freedom of expression – ‘public policy’ should not be shaped based on a simple use of these principles.64
The solution to this conflict is found through proper balancing which reflects the relative importance of conflicting values. ‘In determining the scope of “public policy” there must be an internal balance between conflicting interests and values’.65 Thus,66 only when the weight of the principles regarding freedom of contracts and the autonomy of the individual’s will is smaller than the weight of the opposing value – only then is there room to void a stipulation in a contract, or the entire contract, due to a conflict of public policy.
Procaccia J was right to state that ‘determining the scope of public policy requires a delicate and complicated balance between conflicting values and interests’.67 This balancing may be learnt, through interpretive analogy, from the rules of balancing set forth in the rules of constitutional proportionality.68 c. Good Faith Good faith, for the purpose of this chapter, is an objective standard. It expresses a standard of honesty, fairness and allegiance between adversaries.69 It is meant to determine that a person ensuring their matters, must do so in a manner which is fair and all the while considering the other party’s justifiable expectations. ‘Human being to human being, one cannot behave like a wolf, but one is not required to be an angel. Human being to human being, one must act like a human being’.70 The honesty and fairness between rivals are required by good faith and shaped by the values which make up the objective aspect of the constitutional rights. Good faith reflects the consideration of objective values regarding human dignity, privacy, property and liberty. Good faith is indeed a ‘pipeline’71 through which the objective values of the legal system flow through, and reflect the values of the constitutional right into private law. Additional values, related to the interpersonal behaviour of those with opposite interests, also flow into it. These values are in constant conflict. Good faith reflects the proper balance between them. This balance expresses the relative social importance of the conflicting values.72 In determining this balance it is possible to learn, through analogy, from the rules of balancing that are part of constitutional proportionality.73
Kestenbaum (n 38) 533 (Barak J’s opinion). CA 5258/98 A v B 58(6) PD 209, 222 [2004] (in Hebrew) (Court President Barak’s opinion). See also Saar (n 52). 66 Kestenbaum (n 38) 533 (Barak J’s opinion). 67 CA 11081/02 Dolev v Kadosh 62(2) PD 573, 603 [2007] (in Hebrew) (Procaccia J’s opinion). 68 See A Barak, Proportionality: Constitutional Rights and their Limitations (Srigim Li On, Nevo Publishing, 2010) 101 (in Hebrew). 69 FH 22/82 Beit Yules v Raviv 43(1) PD 441, 484 [1989] (in Hebrew); CA 1966/07 Ariel v Pension Fund of Egged Members Ltd (9 August 2010), Nevo Legal Database (by subscription) (in Hebrew). 70 LCA 6339/97 Roker v Salomon 55(1) PD 199, 279 [1999] (in Hebrew) (Court President Barak’s opinion). See also HCJ 59/80 Beer-Sheba Public Transport Services Ltd v National Labour Court 35(1) PD 828, 834 [1980] (in Hebrew); Beit Yules (n 69). 71 Roker (n 70) 279. 72 See Barak, Proportionality (n 68) 430. 73 ibid 101. 64 65
392 Aharon Barak iii. Interpretation of Private Law and its Validity What is the case if the interpretation of private law – interpretation which takes into consideration the objective aspect of the constitutional rights – leads to a result which does not completely express this aspect? This occurs when private law determines arrangements which do not match up with the interpretive balance between the objective aspects of constitutional rights or between the other constitutionally protected objective values. What is the effect of this private law? The answer is that the constitutionality of this private law will be tested. Indeed, this private law is an act of the state (whether legislation or common law). To the extent that this private law can limit the individual’s constitutional rights vis-à-vis the state, the limitation will lead to the unconstitutionality of the private law unless it fulfills the requirements of the Limitation Clause in general and the requirements of proportionality stricto sensu within it in particular. Thus, the radiation of the objective values at the foundation of constitutional rights onto private law does not ensure, in and of itself, that this private law is constitutional. It is possible that due to linguistic limitations, and the existence of additional and contradictory values, the means used by legislation and its interpretation, will lead the interpreter to conclude that private law limits individuals’ constitutional rights vis-à-vis the state, and that this limitation is not proportional. C. Second Channel of Application: A Negative Arrangement and Lacunae i. The Silence of Private Law The interpretation of private law assumes the existence of relevant legal text. What happens if this text does not exist? If the absorption of the constitutional rights as objective values in private law is done through private law’s provisions – what happens if there is no relevant private law? Is this really the failure point of the indirect application? In order to answer this question we must list the different meanings of silence in private law.74 This silence has different voices. For this purpose, the two relevant meanings that can be given to such a silence are negative arrangement and lacuna. ii. Negative Solution A negative solution is based on an interpretive conclusion, according to which the solution determined in the law contains an implicit negative solution regarding matters not regulated by law. At times the negative solution will reflect the objective aspects of constitutional rights. In this situation, the indirect application is resolved by the negative solution. But what is the case if a negative solution refuses the radiation stemming from the constitutional right’s objective values? What is the destiny of the indirect application in this situation? Does the latter mean that the indirect application of constitutional rights is not possible? My answer is that the non-expression of the objective aspects of constitutional rights may disproportionally limit the individual’s constitutional rights vis-à-vis the state which determined the negative solution. The result may be that the negative solution is See Barak, Purposive Interpretation (n 10) 67.
74
Constitutional Rights and Private Law 393
unconstitutional. Indeed, the destiny of a negative solution regarding constitutionality is the same as that of a positive solution regarding constitutionality. They both can disproportionally limit the individual’s constitutional right vis-à-vis the state. iii. Lacunae We assume that the private law does not determine a text which should be interpreted with the constitutional right’s objective aspect in the background. We also assume that the lack of legislative text stems from a lacuna. Is the significance of the latter that there is no room for indirect application? In order to answer this question we must first examine the rules of lacunae and the reaction to the lacunae. If the court does not have the authority to complete the lacunae in the legislation – as is the case in most common law legal systems75 – then the lacunae remains as is. The solution to this legal problem will exhaust itself within the boundaries of the common law. The common law makes up a ‘security net’ which is meant to solve, amongst others, most of the legislative lacunae. The only thing to examine is if this common law – similar to that of a negative solution – is constitutional. In most continental legal systems and in Israel, the court has the authority to complete a lacuna in the law.76 This complementary law is based on completion through analogy from a similar legislative norm from private law. Without a suitable analogy the judge looks to the general principles. Completion of lacunae is not merely an authority given to the judges. It is a duty imposed upon them. Once the lacuna in the legislation has been discovered, the judges must complete it according to the rules used in their legal system to do so. Once the completion of the lacuna has been completed, new law has been added to private law. This complementary law was created by the judicial authority. This is sub-constitutional law. It adds a complementary norm to the legislation. Like the rules of the common law, this rule may also limit the individual’s constitutional right vis-à-vis the state. The constitutionality of this limitation will be determined according to its proportionality. D. Third Application Channel: Creating New Private Law i. New Legislation and Developing the Common Law The interpretation of private law – including its valve concepts – may lead to the conclusion, that these solutions cannot ‘absorb’ the radiation of the objective aspect of the constitutional rights. In addition, this interpretation may lead to the conclusion, that the lack of solution in private law regarding the absorption of objective values is not a negative solution nor does it mean there is a lacuna. Has this situation led to the end of the road for the indirect application model? My answer is in the negative. If private law does not exist – it should be created. The legislator (in the framework of the legislation) and the judge (in the framework of the common law) cannot wash their hands of this. They should create new law (legislative or case law) which can express the objective values of the constitutional rights. ibid 69. ibid.
75 76
394 Aharon Barak ii. New Legislation Much of private law is legislated. If there is no legislation which can ‘absorb’ the objective aspects of constitutional rights, there is no escape from new legislation. The legislator will legislate new laws to solve the problem not yet solved in private law. Through its content, the new legislation will express the objective aspect of the constitutional rights. This will be done through the legislative recognition of new individuals’ sub-constitutional rights vis-à-vis other individuals. iii. The Development of the Common Law In the common law legal systems the judge’s power to create law is recognised. Contract and tort law in most common law legal systems are case law, where little legislation is involved. This development in the case law must take place here as well. When a judge encounters a legal problem regarding an interpersonal relationship whose basic values are derived from the objective aspect of the individual’s constitutional rights vis-à-vis the state, the judge should develop the common law in a manner that the basic constitutional values will be absorbed into private law. The development of the common law will translate the objective aspect of the constitutional rights into sub-constitutional law creating rights of individuals vis-à-vis other individuals. This approach was adopted by the Constitution of South Africa that provides:77 (a) in order to give effect to a right in the Bill, must apply, or if necessary develop, the common law to the extent that legislation does not give effect to that right; and (b) may develop rules of the common law to limit the right, provided that the limitation is in accordance with Section 36(1).
From this provision it appears that the development of the common law is the duty of the judicial authority. It is needless to repeat that this development of the common law is possible only when the legislation does not determine a contradictory solution. The common law cannot change legislation. The judicial activity in the development of the common law, as in Holmes’ much quoted phrase, is ‘interstitial’.78 The Israeli Supreme Court acted along these lines when it recognised compensation for a limitation of the individual’s will where surgery was carried out without the patient’s informed consent.79 Similarly the Supreme Court was asked to recognise procedural arrangements which would allow for the exposure of the identity of online commentators for slander. The request was denied by the majority.80 iv. Proportionality of New Legislation and the Development of the Common Law New legislation and the development of common law must be proportional. Indeed, development of private law by the legislature or the judiciary is a state action. This action may limit a constitutional right of an individual. This limitation will be constitu South African Constitution, 1996, s 8(3). See also the South African cases (n 7 above). See Southern Pacific Co v Jensen 244 US 205, 221 (1917). 79 CA 2781/93 Daaka v Carmel Hospital 53(4) PD 526 [1999] (in Hebrew). 80 LCA4447/07 Mor v Barak ESS Company for International Telecommunications Ltd (25 March 2010), Nevo Legal Database (by subscription) (in Hebrew). 77 78
Constitutional Rights and Private Law 395
tional only if it is proportional.81 Therefore it must secure that the balance between the conflicting objective aspects will coincide with the balance between the individual’s constitutional right vis-à-vis the state and the constitutional rights of others. This balance will be determined by proportionality stricto sensu. v. New Legislation and the Development of the Common Law – Discretion or Obligation? a. Negative and Positive Rights The indirect application model is based on existing private law, including the completion of lacunae, which ‘absorbs’ the objective aspect of the constitutional rights. Where such private law does not exist, the absorption of the objective aspect will be carried out by the new law anchored in legislation or common law, which develops private law. The question that arises in this matter is, is the development of this private law a discretionary activity by the legislator (in developing legislation) or the judge (in developing the common law) or a duty imposed upon them? If development of private law is not an obligation it raises the question of the efficiency of the indirect application model. The answer to this question leads us to an important issue of constitutional law, regarding the ‘positive’ aspect of constitutional rights.82 The question is if the constitutional rights are but negative rights, meaning, rights which impose duties on the state (legislative, executive or judiciary) not to limit the right holder, or whether they are also positive rights, meaning, those rights which impose a duty on the state to protect the right holder. When the approach is that the state duty is only a negative one, no duty to protect the individual’s constitutional right is imposed. This means the lack of duty to develop the private law in the spirit of the indirect application model. If, on the other hand, the duty of the state is also a positive duty, then the state must protect the individual’s constitutional rights vis-à-vis from infringement by another individual. The state will have fulfilled its duties by developing the law according to the objective aspect of the constitutional right. The state will do so, inter alia, if it recognises the subconstitutional right of the right holder vis-à-vis other individuals. It will make up part of the private law. The answer to the question of the ‘positive aspect’ of the constitutional rights changes from legal system to legal system. American law does not recognise, in principle, the positive aspect.83 The Bill of Rights is meant to prevent limiting constitutional rights by the state. They are not meant to impose upon it the duty to protect the individual. Other legal systems recognise the positive aspects of a number of rights, but not them all.84 See Barak, Proportionality (n 20) 118. See also South African Constitution, 1996, s 8(3)(b). Regarding this issue, see Barak, Proportionality (n 20) 422. See also D Grimm, ‘The Protective Function of the State’ in G Nolte (ed), European and US Constitutionalism (Cambridge, Cambridge University Press, 2005) 137, 138; A Mowbray, The Development of Positive Obligations Under the European Convention on Human Rights by the European Court of Human Rights (Oxford, Hart Publishing, 2004); S Freedman, Human Rights Transformed: Positive Rights and Positive Duties (Oxford, Oxford University Press, 2008); H Shue, Basic Rights: Subsistence, Affluence, and US Foreign Policy, 2nd edn (Princeton, Princeton University Press, 1996) 155. 83 See Jackson v City of Joliet 715 F 2d 1200, 1203 (1982). 84 Eg, the Convention for the Protection of Human Rights and Fundamental Freedoms (European Convention on Human Rights, as amended) (ECHR) recognizes the positive aspect of a number of rights. See Mowbray (n 82). 81 82
396 Aharon Barak German constitutional law recognises the general positive aspect of all constitutional rights (the schutzpflicht)85 and this is the case in the Constitution of South Africa.86 The situation in the Israeli legal system is not free from doubt. It seems that the right to life, body and dignity has both a positive and negative aspect.87 The negative aspect (the state’s duty not to limit a constitutional right) is expressed in section 2 of Basic Law: Human Dignity and Liberty:88 There shall be no limitation of the life, body or dignity of any person as such.
The positive aspect (the state’s duty to protect the constitutional right) is determined in section 4 Basic Law: Human Dignity and Liberty:89 All persons are entitled to protection of their life, body and dignity.
Through this positive aspect the state in Israel is under a duty to protect the individual’s life, body and dignity from their infringement by other individuals.90 The question which has not yet been resolved in Israel is if the positive aspect of the constitutional right applies only regarding the right to life, body and dignity included in section 4 of the Basic Law or does it extend to the entirety of the constitutional rights in this Basic Law (property, liberty, mobility and privacy) and in Basic Law: Freedom of Occupation (freedom of occupation). Recognizing the positive aspect of the constitutional rights (all or some of them) raises two questions. First, do we find vis-à-vis the state’s duty the individual’s right? Second, how can an individual realise the positive duty? b. Fulfilling the Duty in Positive Rights Regarding the first question, I believe that if a duty is imposed on the state to protect certain values, then opposite such duty stands the individual’s right.91 The state’s duty is the individual’s right. Indeed, Basic Law: Human Dignity and Liberty, explicitly states that ‘all persons are entitled’92 to protection of their life, body and dignity. Human dignity is not only a ‘negative’ right, but also a ‘positive’ right. There is a constitutional obligation on the legislature, the executive and the judiciary to protect positive rights of individuals. It is not at their discretion – it is an obligation. Therefore, the judiciary is obliged to develop the common law in a manner that will guarantee protection of his rights vis-à-vis the state.93
See Grimm (n 82) 143. See South African Constitution, 1996, s 7(2). 87 See HCJ 2557/05 Majority Camp v Israel Police 62(1) PD 200 [2006] para 10 of Barak J’s opinion (in Hebrew); HCJ 6427/02 Movement for Quality Government v Knesset [2006] 61(1) PD 619 [2006] para 31 of Barak J’s opinion (in Hebrew). 88 Basic Law: Human Dignity and Liberty, s 2. 89 ibid, s 4. 90 HCJ 366/03 Commitment to Peace and Social Justice Society v Minister of Finance [2005] 60(3) PD 464, 479 (Court President Barak’s opinion). 91 See Grimm (n 82) 153; Alexy (n 12) 301. 92 Basic Law: Human Dignity and Liberty, s 4. 93 See Carmichele (n 7) in which the South African Constitutional Court imposed a duty on the courts to develop tort law so that it will recognize the duty of the police to take actions in order to protect individuals from the aggression of other individuals. 85 86
Constitutional Rights and Private Law 397
How does the individual fulfill his positive right? My answer is that he will do so in one of the fashions available to him to realise his negative right. The German Constitutional Court has more than once ordered the legislature to use its legislative powers to fulfill the positive aspect of the constitutional rights.94 In common law legal systems the right holder can ask the court to use its power to create common law.95 Needless to say, if this common law is not to the legislator’s liking, it has the power to void or change this case law. The positive aspect of the constitutional rights – similarly to the negative aspect – does not recognise an absolute positive right. This is a relative positive right. It can be limited by a proportional legal provision (statute or common law). Therefore, if not every positive aspect of the constitutional right is protected by the legal system, it is possible that this situation is constitutional as it is also proportional.96 New legislation that protects the positive right must be proportional. This is also the case for new common law. Legislation which voids or changes the common law and fulfills the positive aspect is constitutional only if it is proportional. IV. THE INDIRECT APPLICATION IN ISRAEL
The Basic Law: Human Dignity and Liberty and the Basic Law: Freedom of Occupation determine a list of constitutional rights. The language of the Basic Laws does not determine who are the right holders. There is no explicit determination to the question if these rights are directed towards the state only or also towards other individuals. Both Basic Laws set forth that ‘all governmental authorities are bound to respect’97 the rights determined within them. It was not determined that this obligation is imposed only on the state. The question of the application of the constitutional rights in interpersonal relationships (horizontal), is seemingly open and given to interpretive determination by the court. The issue arose for the first time during the Jerusalem Community Burial Society v Kestenbaum case.98 This case dealt with the question of whether relatives of the deceased could be granted the right to add onto the tombstone the deceased’s name in Latin letters as well as her date of birth and death in numbers. In my opinion I noted that this right could be based on the rules of private law. In this context I mentioned that the condition regarding having solely Hebrew letters, which arose from the contract signed by the deceased’s husband, was contrary to public policy and was therefore void. This conclusion was based on the indirect application model, while noting that the question of direct application remained open.99 The line of thinking on which the case was examined was:100 [T]he basic principles of the system in general and a person’s human rights in particular are not limited to public law. The differentiation between public and private law is not so severe. The See Grimm (n 82) 153. See also Oliver and Fedtke (n 3) 153. See Rami Mor (n 80) Rubinstein J’s opinion. 96 See Barak, Proportionality (n 20) 429. 97 Basic Law: Human Dignity and Liberty, s 11; Basic Law: Freedom of Occupation, s 5. 98 Kestenbaum (n 38). 99 ibid 530. 100 ibid. 94 95
398 Aharon Barak legal system is not a confederation of legal fields. It makes up a unification of method and just ice. Therefore, basic principles are principles of the entire system, and not solely of public law.
The question I discussed in my opinion was ‘how do basic principles of public law flow into private law, and from there they are delivered into the individual’s behavior in his relationship with other individual’s’.101 In my answer I referred to the principles of interpretation.102 In addition, I mentioned that ‘private law includes a number of doctrines, which use tools through which the basic principles of the system in general and the person’s basic rights in particular flow into private law’.103 One of those doctrines is that of public policy. I noted that ‘this provision . . . channels into private law the basic principles of the law in general and basic human rights in particular’.104 I emphasised that: [B]y transferring the basic principles of the system in general and the person’s human right in particular from public law into private, they were changed. The state’s duty to fulfill human rights is different in content to the individual’s duty to fulfill human rights. It is clear the state’s duty is more severe and that the protection of basic right vis-à-vis it is more expansive. However, every individual must also respect another individual’s basic rights. Indeed, the individual’s duty to fulfill the other’s basic rights is derived from the same source that the state’s duty is derived from – these are the basic social understandings and the basic legal principles that the legal system is built upon.105
The question of the indirect application arose again in the AES System Inc v Saar case.106 In this case an employee signed a non-compete agreement. The question was if this agreement was valid. As in the Jerusalem Community Burial Society v Kestenbaum case,107 it was decided that the agreement was void as it contradicted public policy. The indirect application model of constitutional rights has been recognised in all fields of private law.108 Thus, for example, it applies in contract law,109 tort law110 and labour law.111 V. CONCLUSION
With the recognition of the state’s duty to act – as well as the scope of this duty – the cycle is complete. The subjective constitutional right which operated in the relationship between the individual and the state on the constitutional level becomes the individual’s ibid. ibid. 103 ibid. 104 ibid 531. 105 ibid; see also CA 3156/98 Ben Yishai v Veingarten 55(1) PD 939, 952 [1999] (in Hebrew). 106 Saar (n 52). 107 Kestenbaum (n 38). 108 See Hoffman (n 3). 109 HCJ 2911/05 Elhannati v Minister of Finance 62(4) PD 406 [2008] (in Hebrew). D Barak-Erez and I Gilad, ‘Human Rights in Contract Law and Tort Law: The Quiet Revolution’ (2009) 8 Kiryat Hamishpat 11 (in Hebrew). See also D Bhana and M Pieterse, ‘Towards a Reconciliation of Contract Law and Constitutional Values: Brisley and Afrox Revisited’ (2005) 122 South African Law Journal 865. 110 Barak-Erez and Gilad (n 109). 111 See A Barak, ‘Constitutional Rights and Private Law: Application in Labor Law’ in Essays in Honor of Elika Barak-Ussoskin (Tel-Aviv, Nevo Publishing, 2012). See also LA 90/08 Isakov Inbar v Commissioner for Women Labour (8 February 2011), Nevo Legal Database (by subscription) para 8 (in Hebrew); LA 33680-08-10 Dizengoff Club v Zoili (16 November 2011), Pador Legal Database (by subscription) paras 11, 25 of Court President Arad’s opinion (in Hebrew). 101 102
Constitutional Rights and Private Law 399
subjective sub-constitutional right vis-à-vis other individuals. This metamorphosis takes place through three stages: during the first stage objective constitutional values are drawn from subjective constitutional rights of the individual vis-à-vis the state. During the second stage these objective values penetrate into private law – through interpretation and the completion of lacunae. During the third stage, the objective aspect of the constitutional right penetrates into private law through new legislation or new common law. If this legislation and development are the legislator’s and the judge’s duty – as they are if the legal system recognises positive rights – then they have fulfilled their duty through the legislative or common law recognition of subjective rights at the legislative or common law level (at the sub-constitutional level) of the individual vis-à-vis the individual. This sub-constitutional law is constitutional only if it is proportional. It is important to note, that the indirect application model affects the interpersonal relationships. However, it does not turn these relationships into constitutional relationships. At the constitutional level it is the relationship between the individual and the state. The relationship between individuals, whose content is affected by the indirect application, is not at the constitutional level. It is at the sub-constitutional level. The constitutional level expresses the individual’s constitutional right vis-à-vis the state to its fullest extent. The constitutional possibility to limit this scope does not reduce it, but rather affects the constitutionality of the sub-constitutional law which limits it.112 This sub-constitutional norm – which reflects the private law as it was designed according to the effects of the indirect application – does not express the full scope of the constitutional right. Private law will express the constitutional right as it was limited by the other’s constitutional right and by the public interest. The individual’s right in the private law vis-à-vis another individual, will always be narrower in scope than the individual’s right vis-à-vis the state.113 The individual’s right vis-à-vis another individual, within private law, reflects the proportional limitations on the fulfillment of the constitutional right. It is an expression of the understanding that the individual’s constitutional right vis-à-vis the state is conveyed into the private law as an individual’s right vis-à-vis another individual. This right expresses the proportional limitations on the realization of the constitutional right vis-à-vis the state in private law. While the scope of the individual’s constitutional right vis-à-vis the state is wider than the scope of the sub-constitutional right, the realization of the constitutional right vis-à-vis the state also happens within the sub-constitutional law. Hence the similarity between the two. Nonetheless, this similarity does not mean they are identical. This is due to the difference between the principles and values at the balance.114 In many cases, the result reached by the indirect application model is not different from the one achieved by the direct application model.115 A study of comparative law emphasises that at times in the same legal system, two models of application are used, without the court having defined the difference between them.116 A legal system (such as German constitutional law) where all the stages occur – the transformation of a subjective constitutional right into an objective value, the penetration of objective values into See Barak, Proportionality (n 20) 32, 83. Kestenbaum (n 38) 531. 114 See HCJ 488/11 Medical Organization of Israel v Minister of Health (19 June 2011), Nevo Legal Database (by subscription) (in Hebrew); Ben Yishai (n 105) 952. 115 See Kum and Comella (n 12). 116 See Oliver and Fedtke (n 3) 520–57. 112 113
400 Aharon Barak private law through interpretation and the completion of lacunae, the duty to develop the private law by the legislator – will make up solutions according to both very similar methods of application. However, the difference between them should not be blurred. The results will not always be the same; the objective aspect of every constitutional right will not always be recognised; the purposive interpretation which grants an interpretive status to the objective purpose of every legal norm will not always apply; the legal system will not always agree to the penetration of objective values into its valve concepts; a duty will not always be imposed on the legislator and judiciary to develop the law in a fashion which fulfills the objective aspect of the constitutional rights. In any case where one of the stages is not activated in a complete fashion, the difference between the direct and indirect application models will be emphasised. Even if all stages are completely activated, there will still be a clear difference between the direct and indirect application models. This difference is expressed by the fact that in the direct application model the individual has a constitutional right vis-à-vis another individual. The indirect application model does not recognise a constitutional right between individuals. The indirect application model recognises the individual’s sub-constitutional right (legislative or common law) vis-à-vis another individual. This difference may have little practical significance, because in both models the realization of the right, and the remedies for its limitation, are on a sub-constitutional level. There may be little difference between the content of a sub-constitutional law that realises the right of an individual vis-à-vis the state and the content of a sub-constitutional law that realises the right of an individual vis-à-vis another individual. Nonetheless, there is still a difference. The reason for this is that the interests and values taken into account in forming a sub-constitutional law that deals with the right of an individual vis-à-vis the state differs greatly from the interests and values taken into account in forming a sub-constitutional law that deals with the right of an individual vis-à-vis another individual.
26 Human Rights in Private Law: Hybridization of the Balancing Tests MICHAL TAMIR
I. INTRODUCTION
T
HE APPLICATION OF human rights in private law poses a complex challenge to legal scholars.1 It combines private and public law in a way that calls for an in-depth grasp of the basic assumptions of each and every legal field as well as the need to adapt theories and narratives that propel law into an era in which the boundaries between the ‘private’ and the ‘public’ are blurred.2 It embodies the implementation of constitutional doctrines (such as the rights discourse and the balance theory) within the developed and comprehensive doctrine of private law and its basic assumptions. The core of the clash between these worlds dates back to the philosophical origins of political, social and legal discourse whereby the values of human dignity and equality – which are dominant in constitutional law – clash with the values of freedom and autonomy which characterise private law.3 In Israel, the application of human rights in private law is a given. It is the outcome of a set of statutory arrangements4 alongside a set of fundamental judicial laws originating in Court President Aharon Barak’s judgment in the matter of Kestenbaum.5 Moreover, Israeli law and the ways in which it copes with the issue constitute a comparative model for a great many countries which have recently been grappling with the development of the constitutional doctrine and the application of human rights in private law.6 1 MJ Horwitz, ‘The History of the Public/Private Distinction’ (1982) 130 University of Pennsylvania Law Review 1423, 1424. The attack on the public/private distinction was the result of a widespread perception that so-called private institutions were acquiring coercive power that had formerly been reserved to governments. 2 To the distinction between private and public law, see A Barak, Interpretation in Law: Constitutional Interpretation, vol 3 (Jerusalem, Nevo, 1994) 649–50 (in Hebrew); D Barak-Erez, ‘Public Law and Private Law – Overlaps and Mutual Influences’ (1999) 5 Law and Government 95, 95–99 (in Hebrew); A Bendor and M Tamir, ‘Constitutional Clichés – Between Public Expression and Private Dignity’ (2002) 32 Mishpatim 623, 632–39 (in Hebrew). 3 cf N Cohen, ‘Equality vs Freedom of Contract’ (1993) 1 HaMishpat 131, 131–32 (in Hebrew); H Keren, Contract Law from a Feminist Perspective (The Hebrew University of Jerusalem, Sacher Institute for Legislative Research and Comparative Law, 2005) 267–353 (in Hebrew). 4 Among the dominant laws are: Prohibition of Discrimination in Products, Services and in Entry into Places of Entertainment and Public Places Law, 5761-2000; Employment Law (Equal Opportunities), 57481988; Equality Right for People with Disability Law, 5758-1998. 5 CA 294/91 Jerusalem Community Burial Society v Kestenbaum 46(2) PD 464 [1992] (in Hebrew). 6 Eg H Hershkoff, ‘“Just Words”’: Common Law and the Enforcement of State Constitutional Social and Economic Rights’ (2011) 62 Stanford Law Review 1521, 1582.
402 Michal Tamir Notwithstanding the above, whether applying human rights in private law is justified or not is still a debatable issue among scholars.7 Furthermore, one can hardly say that the topic was thoroughly addressed by the Israeli Supreme Court. Nonetheless, it appears as though the discourse on this matter – worldwide as a whole and in Israel specifically – left the justification stage behind a long time ago8 and switched to focusing on the particular applications of the statutory arrangements and the application of the judicial model for constitutional rights in private law. Thus, the main challenge currently facing Israeli jurists is that of defining legal tools for the development and proper conceptualization of how, and to what extent, to apply human rights in private law. In particular, it appears as though a significant question that requires development is the relation between customary constitutional balancing9 and the balancing involved in the application of constitutional rights in private law. It is an indisputable fact that once constitutional rights become part of the fabric of private law, the balancing picture changes and the current constitutional doctrines of balancing do not offer all the tools required to achieve an appropriate application of constitutional rights. Thus for example, one convention that highlights the difference is that the Limitation Clause10 – which is part of the Basic Laws and embodies ‘vertical’ balance between individual rights and public interest11 – cannot be applied in an analysis of human rights in private law. It further appears that the simplistic implementation of the ‘horizontal balancing’ test developed in constitutional law to cope with the clash between two human rights of identical constitutional status12 is irrelevant because of the different properties of private law which do not allow for its application.13 This chapter is also based on the assumption that despite the clear difference between the action of public authorities and that of individuals and despite the difficulty involved in infringing on the autonomy of private will, the application of human rights must be recognised in the modern state, even in private law. The chapter will accordingly focus on attempting to depict the characteristics pertinent to the continued development of law – namely to the judicial implementation of the application of human rights in private law in concrete cases, primarily in contract law. The premise is that the differences between public and private law need to be reflected in the discussion surrounding the application of human rights in private law.14 As a fundamentally liberal country, Israel 7 H Dagan, ‘Values: The Limited Autonomy of Private Law’ (2008) 56 American Journal of Comparative Law 809, 811. 8 See, eg D Barak-Erez and I Gilead, ‘Human Rights in the Laws of Contracts and Torts’ (2009) 8 Kiryat Hamishpat 11, 36 (in Hebrew). 9 For some recent and helpful principles of constitutional balancing, see A Barak, ‘Proportionality and Principled Balancing’ (2010) 4 Law & Ethics of Human Rights 2; S Gardbaum, ‘A Democratic Defense of Constitutional Balancing’ (2010) 4 Law & Ethics of Human Rights 78. 10 See n 24 below and the accompanying text. 11 In his judgment in the Kestenbaum case, Court President Barak already made clear that the basic balancing in vertical relationships does not apply to horizontal relationships between individuals, because the substance of the duty to protect human rights and the extent of the protection are different (Kestenbaum (n 5) 531). Nonetheless, Professor Aharon Barak’s recent writings suggest the need for a broader use of the proportionality test, embodied in the Limitation Clause, even in private law. See Barak, ‘Proportionality’ (n 9) 116. 12 See, eg HCJ 2481/93 Dayan v Wilk 48(2) PD 456, 475–76 [1994] (in Hebrew). 13 There are those who claim that the distinction between vertical and horizontal balancing stems from its importance pursuant to the Basic Law since both types of balancing are included within the scope of proportionality. See Barak, ‘Proportionality’ (n 9) 305. 14 cf BR Snyder, ‘Private Motivation, State Action and the Allocation of Responsibility for Fourteenth Amendment Violations’ (1990) 75 Cornell Law Review 1053, 1061 (explaining that when a private party acts,
Human Rights in Private Law 403
shall not forego freedom of contract or the autonomy of individual will and the courts must be cautious about establishing balance when the legislator fails to do it. In part II, I will briefly review the development of the application of constitutional rights in private law in Israel. Part III will focus on the groundbreaking decision in the Kestenbaum case. In part IV, I will attempt to propose several relevant indices for balancing between constitutional rights in private law, which will likely be instrumental toward defining criteria to achieve ‘private balancing’ between constitutional rights in the realm of private law. Part V briefly sums up and demonstrates implementation of the suggested criteria on the Kestenbaum case. II. GENERAL BACKGROUND
In 1992, the State of Israel underwent a ‘Constitutional Revolution’15 through the enactment of two Basic Laws protecting human rights,16 namely: Basic Law: Human Dignity and Liberty and Basic Law: Freedom of Occupation. This revolution emphatically placed the question of the application of these constitutional human rights in private law on the public agenda.17 The Basic Laws – which achieved supra-legal constitutional status by virtue of Supreme Court judgments18 – raised a question which could also apply to other constitutions worldwide: does the constitution also apply in private law? Is the rationale underpinning human rights confined solely to protecting the individual against governmental violations, or are the interests protected by the Basic Laws also relevant to violations by private individuals?19 In Israel, this question may be particularly complex. Whereas the Basic Laws expressly state that these rights ‘shall be upheld in the spirit of the principles set forth in the Declaration of the Establishment of the State of Israel’,20 and that their purpose is to establish in a Basic Law ‘the values of the State of Israel as a Jewish and democratic state’21 – that is that they are intended to shape the character of Israeli society – the specific provisions in the Basic Laws expressly address the legislature and governmental authorities. The section pertaining to rights violations, which has earned the title of ‘the Limitation Clause’, addresses the legislature and states:
it does so on its own behalf. In contrast, government actors ultimately derive their authority and power from the citizenry, who expect them to actively promote the public interest). 15 See R Gavison, ‘Round Table: Israeli Constitutionalism’ (2005) 6 Yale Israel Journal 25, 27. 16 See AL Bendor, ‘Is It a Duck? – On the Israeli Written Constitution’ (2005) 6 Yale Israel Journal 53. 17 See D Friedmann and D Barak-Erez, ‘Introduction’ in D Friedmann and D Barak-Erez (eds), Human Rights in Private Law (Oxford, Hart Publishing, 2001) 1, 1 (explaining that human rights were originally conceived as rights and freedoms vis-à-vis the state and other public authorities). 18 See CA 6821/93 United Mizrahi Bank Ltd v Migdal Cooperative Village 49(4) PD 221 [1995] (in Hebrew) extracted in (1997) 31 Israel Law Review 764, 770. See also, A Barak, ‘Human Rights in Israel’ (2006) 39 Israel Law Review 12 (discussing the two major periods in Israeli history: the first period – from the formation of the State in 1948 up to the enactment of the Basic Laws pertaining to Human Dignity, Liberty and Freedom of Occupation in 1992; and the second period – from the enactment of these Basic Laws to the present). 19 cf A Barak, ‘Constitutional Human Rights and Private Law’ in D Friedmann and D Barak-Erez (eds), Human Rights in Private Law (Oxford, Hart Publishing, 2001) 13, 13 and the related footnotes (determining that the applicability of constitutional human rights within the sphere of private law is an issue of major importance, which has arisen in most legal systems). 20 Basic Law: Human Liberty and Dignity, s 1A; Basic Law: Freedom of Occupation, s 1. 21 Basic Law: Human Liberty and Dignity, s 1; Basic Law: Freedom of Occupation, s 2.
404 Michal Tamir There shall be no violation of rights under this Basic Law except by a law befitting the values of the State of Israel, enacted for a proper purpose, and to an extent no greater than is required or by regulation enacted by virtue of express authorization in such law.22
In other words, the Basic Laws chart an express course for the Court, which requires it to examine cases in which there is a clash between the right and public interest, but is silent regarding clashes between rights as such. Moreover, the application clause, also known as the ‘Respecting Clause’, provides that ‘all governmental authorities are bound to respect the rights under this Basic Law’,23 so that the Basic Laws bind governmental authorities, but not individuals, to respect rights. These problems are not purely semantic. Protecting one’s rights in private law means violating the rights of another. It follows that if the Court assumes the task of creating formulae for balancing rights, it will be violating rights without statutory authorization.24 On the other hand, if it fails to protect a civil right which has been violated by private law, the Court will be disregarding basic rights vested by our constitutional document upon ‘a person per se’.25 Concurrently, the legislation regulating the clashes between human rights in private law developed and generated its own balances. Thus, for example, the Defamation Law, 5725-1965, which preceded the Basic Laws, regulates the clash between reputation and freedom of expression; section 2 of the Equal Opportunity in Employment Law,26 regulates the issue of equality in private employment relations and prohibits an employer from discriminating against employees or job seekers on grounds set forth in the law; the Prohibition of Discrimination in Products Law,27 prohibits any person or private body selling a product or public service from discriminating against a person on the grounds set forth in that law. It may therefore be said that, except where housing is concerned, legislation in Israel currently regulates most areas in which rights clash in private law. Under these cir cumstances – where the matter is regulated by legislation – given that this legislation is constitutionally valid, the question of the application of rights in private law does not arise and the only issue to be considered is the interpretation and implementation of the legislation. However, the difficult questions have inevitably been left unresolved. Thus, for example, is a person entitled to refuse to let his or her apartment to an Arab, an Ethiopian or a homosexual? There are several models in various countries addressing the question of the application of constitutional rights in private law, when the legislature does not give its own answer in a specific enactment. The main models are: the Canadian ‘non-application’ model, according to which the Charter is applicable only in regard to the relations between the citizens and the federal authorities or the authorities of the various provinces;28 the American model of ‘application to the judiciary’, according to which although the constitution is not applicable in the private law, courts would not give a Basic Law: Human Liberty and Dignity, s 8. See also Basic Law: Freedom of Occupation, s 4. Basic Law: Human Liberty and Dignity, s 11. See also Basic Law: Freedom of Occupation, s 5. 24 cf Barak, ‘Constitutional Human Rights’ (n 19) 42 (stating that it is imperative to take a clear constitutional stand on the application of human rights in relationships between private parties, and that it is desirable that such a stand be taken by the constitutional legislative body rather than left to the judiciary). 25 cf ibid 28. 26 Employment Law (n 4). 27 Prohibition of Discrimination in Products, Services and in Entry into Places of Entertainment and Public Places Law(n 4). 28 See RWDSU v Dolphin Delivery Ltd [1986] 2 SCR 573 (Can). 22 23
Human Rights in Private Law 405
remedy which infringes upon human rights since it is a ‘state action’;29 and the ‘direct application’ model, explicitly applied by the South African Constitution,30 and prevailing in a number of European countries by case law, according to which the constitution is applicable in the private law.31 The Israeli Court has chosen the ‘indirect application’ model, developed in the German Constitutional Court.32 In this model, constitutional human rights permeate private law through private law doctrines and mainly ‘safety valves concepts’ (concetti valvola) such as ‘good faith’, ‘public order’ and ‘reasonableness’. These concepts, which assess human behaviour with extra-legal yardsticks, are affected by constitutional rights.33 Justice Aharon Barak adopted this model in the Kestenbaum case34 and further fine-tuned it to adhere to the ‘reinforced indirect model’. According to this model, where existing private law fails to grant a proper remedy to the right infringement, private law is revised to provide a remedy as needed.35 Because the model of indirect application attracted followers in Israel, constitutional human rights produced an impact on private law. Particularly prominent is the example of the employment relationship with regard to the contractual sphere.36 Historically, contracts restricting the business of a person following termination of her employment were examined on the basis of ‘public order’.37 This concept is not defined by law and its content is subject to courts’ interpretation. Generally it means ‘the main values, principles and interests that a given society wish to maintain in a given time’.38 When the basic right of freedom of occupation began to permeate the safety valve concept of ‘public order’, the latter changed its character and options to restrict employees following the termination of their work are now much more limited.39 III. THE KESTENBAUM CASE It appears eminently clear that the founding principles of the legal system in general and basic human rights in particular are not limited solely to public law. The distinction between public 29 See Shelley v Kraemer, 334 US 1, 12–19 (1948); S Gardbaum, ‘The “Horizontal Effect” of Constitutional Rights’ (2003) 102 Michigan Law Review 388, 412–14. 30 South African Constitution, 1996, s 8 (‘Application’). See also S Woolman, ‘Application’ in S Woolman, T Roux and M Bishop (eds), Constitutional Law of South Africa, 2nd edn (OS, 2006) chs 31, 62–74 (describing the current legal situation and the advisable one). 31 See Barak, ‘Constitutional Human’ (n 19) 14–17 (describing the adoption of the ‘direct application model’ in Switzerland and Germany). 32 The Lüth case, 7 BVerfGE 198 (1958) (Ger), described and discussed in DP Kommers, The Constitutional Jurisprudence of the Federal Republic of Germany, 2nd edn (Durham, Duke University Press, 1997) 361–69; C Starck, ‘Human Rights and Private Law in German Constitutional Development and in the Jurisdiction of the Federal Constitutional Court’ in D Friedmann and D Barak-Erez (eds), Human Rights in Private Law (Oxford, Hart Publishing, 2001) 97, 106–07. 33 Barak, ‘Proportionality’ (n 9) 163. 34 Kestenbaum (n 5). 35 See Barak, ‘Constitutional Human’ (n 19) 30. 36 See G Mundlak, ‘Human Rights and the Employment Relationship: A Look through the Prism of Juridification’ in D Friedmann and D Barak-Erez (eds), Human Rights in Private Law (Oxford, Hart Publishing, 2001) 297, 298. 37 See, eg CA 312/74 Cables and Electric Threads Company v Kristianpoler 29(1) PD 316, 319–21 [1974] (in Hebrew) (explaining that the restrictions must comply with two conditions: they are necessary to protect legit imate interests of the employer and they are compatible with the public interest). 38 See A Barak, The Judge in a Democracy (Haifa, Haifa University Press, 2004) 259 (in Hebrew). 39 See, eg CA 6601/96 AES System Inc v Saar 54(3) PD 850, 873–77 [2000] (in Hebrew).
406 Michal Tamir and private law is not that sharp. Legal method is not a confederation of legal fields – rather, it unites method and law. Indeed, the basic principles are those of the method as a whole and not of public law alone. Basic human rights do not target the government exclusively; they also affect mutual relations between individuals themselves.40
These words written by Justice Barak in the Kestenbaum case, and frequently quoted subsequently, mark the budding recognition of the impact of human rights in Israel on private law. In this case, the Court considered the clash between human rights and freedom of contract in a case where the Jewish Burial Society refused to allow a man to write an inscription on his wife’s tombstone in non-Hebrew characters, as the contract contained a provision incorporating the Burial Society’s regulations, which Kestenbaum acknow ledged with his signature. Justice Barak based his finding whereby the contractual provision was void on three alternative grounds: first, the fact that the provision was a depriving condition in a standard contract, which entails its invalidation according to the Standard Contracts Law.41 Second, the fact that the Burial Society was a hybrid body,42 which was also subject to obligations imposed by public law. As such, it had to draw a balance between human dignity and public interest in the Hebrew language, and under these circumstances, the right to human dignity outweighed public interest. Third, the percolation of the right to dignity through the concept of ‘public order’ in contract law. Within the context of public order, a balance had to be drawn between human dignity and the Burial Society’s freedom of contract – a balance where the right to human dignity ultimately outweighed the freedom of contract. Are Justice Barak’s grounds in the Kestenbaum case alternative? Let us modify the data somewhat. In the theoretical situation, there is not one – but several burial societies overseeing the burial of Jews in Jerusalem and each one of them holds different values. The first one holds the Hebrew language sacred, the second one values pluralism, the third one focuses on foreign languages, and so forth. Would the verdict have been different had Kestenbaum insisted on a burial society that holds the Hebrew language sacred? Let us bear in mind that this is a legally competent person who has the option to choose. Under these circumstances, could one have still claimed that the Burial Society was offending his dignity by insisting on the Hebrew language? Is the contractual provision still void for being contrary to public order? Although the Kestenbaum decision constitutes a breakthrough regarding the implementation of the indirect applicability model, it should be interpreted bearing in mind that the Burial Society is also a hybrid body. The verdict broke the ground for the implementation of human rights in private law, however – perhaps because of the fact that the outcome could also have been established based on balances belonging exclusively to public law – no clear criteria were defined for the application of human rights in private law. In the next part, I will attempt to define such criteria. Kestenbaum (n 5) 530. Standard Contracts Law, 5743-1982; see also V Lusthaus and T Spanic, Standard Contracts (Tel-Aviv, Nevo, 1994) (in Hebrew) 47–57 (defining deprivation), 49 (discussing the Kestenbaum case as an example in which the Court weighed several rights and interests and reached the decision that the condition was prejudicial). 42 See A Harel, Hybrid Bodies – Private Bodies in Administrative Law (Tel-Aviv, The Israel Bar-Publishing House, 2008) 56 (in Hebrew) (bringing up the case as an example of a private body administering governmental authorities). 40 41
Human Rights in Private Law 407
IV. AN INCLUSIVE ASSESSMENT OF THE CIRCUMSTANCES
This chapter proposes some principles designed to tackle the issues raised by the clash between human rights in private law which have not been resolved by legislation. These are not principles from which one can derive clear-cut rules as to how to solve all of the complex questions that arise; rather, this is an attempt to set guidelines for seeking a balance between two axioms. The first one involves the autonomy of the free will at the root of private law, which is contrary to the principle of allegiance dominating public law (according to which the authorities are liable for the public and thus have to act only within their authorization and for the public faith); the second axiom concerns the fact that human rights are inherent to all human beings, including vis-à-vis other individuals and not just the government. The postulation is that in order to make sure that rights are essentially, rather than mechanically, applied in private law and genuinely secured, each clash of rights must be handled by thoroughly assessing all the circumstances of the case. Within this context, I wish to propose a list of criteria whose shared characteristic is that they are non- dichotomous. Each trait has the potential for a whole set of possibilities and the point’s position in the set should affect the discussion. Nonetheless, each criterion may have a typical standard in public law, however the standard in private law, is likely to be different. Within this context, the following criteria must be weighed: A. The Substance of Each of the Clashing Rights The process of addressing clashing values should – in my opinion – take into account several aspects that could affect the balancing equation. First, one must take into account that the constitutional norm permeating private law modifies the intensity of its validity during the shift from vertical to horizontal relationships and this change is likely to have an effect on its implementation method. Professor Alexy distinguishes between constitutional rules and constitutional principles. A constitutional rule is a norm that requires a decisive implementation of an objective value. On the other hand, a constitutional principle is a norm at a higher level of abstraction which requires the conceptualization of a value to an optimal extent in relation to the existing normative system.43 This distinction has crucial implications on the balancing method because whereas in constitutional law, balancing tests are designed as rules (for example, the Limitation Clause), in private law, they are designed as principles. Nonetheless, it appears as though it is actually the high level of abstraction that can help to create a balance that is closer to the basic assumptions of private law. The indices that will be proposed hereafter for balancing between constitutional rights in private law, strive to actualise the constitutional norm by taking into consideration the horizontal norms between the parties as well as the implications of implementing the constitutional value on the area in private law in which it is actualised.
See R Alexy, ‘The Construction of Constitutional Rights’ (2010) 4 Law & Ethics of Human Rights 21.
43
408 Michal Tamir Secondly, it is necessary to take the constitutional hierarchy of human rights into account. The constitutional discourse in the State of Israel does not frequently address the issue of the hierarchy within the pyramid of constitutional norms, primarily due to the fact that Israeli constitutional law is – relatively speaking – in the initial phases of defining the constitutional norm through court decision. Nonetheless, it appears as though it is possible to draw a sketchy picture of a normative pyramid at the constitutional level. The upper portion of the pyramid clearly depicts the key civil and political rights such as freedom of movement,44 freedom of political expression45 and rights that are inherently and closely tied to human dignity, such as equality of suspect groups.46 The bottom of the pyramid depicts the socio-economic rights47 to the extent that they pertain to aspects beyond the minimum subsistence level with dignity.48 Constitutional hierarchy is also likely to affect private law in such a way that when a high-ranking right in the hierarchy clashes with a purely economic right, the latter will recede. For example, in a situation where a person only lets one room in an apartment in which she herself continues to dwell, it would seem that her right to privacy, combined with her economic rights to the property, overrides the right to the equality of others.49 Nonetheless, despite the fact that hierarchy in private law is affected by constitutional law, it is likely to be different. In private law, there is bound to be more emphasis on economic rights such as freedom of contract and property. For instance, freedom of political expression is of primordial relevance in the public arena in such a way that it will overcome almost any censorship.50 Conversely, when the journalist’s freedom of expression clashes with the newspaper owner’s freedom of property, it is somewhat limited nonetheless.51 Thirdly, besides the need to weigh the constitutional relevance of the value injected into private law, one needs to consider its sub-constitutional value. This particularly concerns specific constitutional arrangements which already embody constitutional values as well as the implications of these arrangements on the balancing equation in private law. Thus, for example, a review of various constitutional arrangements in the realm of private law will reveal that the right to equality – although not expressly anchored in the Basic Law but rather a derivative of human dignity – garnered stronger defence within the scope of legislation in many private law fields such as employment relations, public services, handicapped persons, etc. This fact is presumably significant in cases where private law is developed through common law as part of the aspiration toward constitutional harmony. In this matter, it shall be emphasised that the absence of equality arrangements in non-regulated legislative areas does not constitute a negative arrangement.52 44 See HCJ 5016/96 Horev v Minister of Transportation 51(4) PD 1 [1997] paras 59–60 of Justice Barak’s opinion (in Hebrew). 45 See HCJ 10203/03 Hamifkad Haleumi v Attorney General 62(4) PD 715 [2008] paras 20–27 of Justice Naor’s opinion (in Hebrew). 46 See HCJ 7052/03 Adalah v Minister of Interior 61(2) PD 202 [2006] para 71 of Justice Barak’s opinion (in Hebrew). 47 See HCJ 1715/97 Investment Managers Bureau v Minister of Finance 51(4) PD 367 [1997] paras 5–8 of Justice Dorner’s opinion (in Hebrew). 48 HCJ 366/03 Commitment to Peace and Social Justice Association v Minister of Finance 60(3) PD 464, 481 [2005] para 15 of Justice Barak’s opinion (in Hebrew). 49 Barak, Interpretation in Law (n 2), 691–92. 50 HCJ 399/85 Kahana v Broadcasting Authority Management Board 41(3) PD 255 [1987] (in Hebrew). 51 See NLC 53/3-223 Palestine Post Ltd v Yechiel 27 PDA 436 (17 October 1994) (in Hebrew). 52 cf Barak, ‘Proportionality’ (n 9) 126 (explaining that completion of the lacuna, at the sub-constitutional level, should be achieved according to the proportionality rules).
Human Rights in Private Law 409
Fourthly, there is an important distinction between negative rights (which grant the right not to be involved in the activity of the individual) and positive rights (which impose the obligation to take action and in certain cases, the obligation to provide resources). Certain rights can be classified as unmistakably negative (such as the right to freedom against imprisonment) and others as unmistakably positive (such as the right to a minimal supply of water). This issue is also highly relevant to the application of human rights in private law. When a ‘negative’ constitutional value permeates private law, it entails less constitutional significance than in the case of a ‘positive’ constitutional value due to the various obligations that are imposed in both cases. For example, stating that an individual’s freedom of employment should not be restricted for motives of non-competition is not the same – from a constitutional standpoint – as stating that a private developer constructing an apartment building in a disadvantaged neighbourhood needs to adhere to minimal standards of a dignified existence. B. Constitutional versus Administrative Equality Is a private individual entitled to refuse to rent his or her home to an Arab, an Ethiopian or a homosexual? Is a private individual entitled to discriminate between his or her various lessees by charging a different rent? It appears as though feelings regarding these types of discrimination differ. Therefore, making the distinction between ‘constitutional equality’ and ‘administrative equality’,53 is vital in regard to the ‘privatization’ of the right to equality, that is applying the right to equality in private law. Constitutional equality – or ‘the principle of non-discrimination in the substantive sense’ deals with ‘suspect groups’ – that is groups who are structurally and historically discriminated against – and it affords a substantial autonomous right not to be discrim inated against for belonging to a certain group.54 In the matter of Poraz, Justice Barak said that ‘equality must be upheld between persons who belong to religions, nations, denominations, races and parties who subscribe to different points of view espoused by entities and groups. Needless to say, this list is not exhaustive’.55 Indeed, alongside the classic grounds of race, nation, religion and sex – whose premise is that any distinction based on them is forbidden unless there is any clear rationale to deviate from this premise56 – over the years, more recent grounds have been added to the list, such as age57 and sexual orientation.58 Administrative equality addresses the administrative authority’s obligation to treat people with no pertinent differences between them consistently for the purpose of exercising authority, regardless of their group affiliation.59 This principle is also called the See M Tamir (Itzhaki), ‘Equality of Gays and Lesbians’ (2000) 45 The Lawyer 94, 109–11 (in Hebrew). See Y Benkler, ‘Non-Discrimination in Housing: Scope, Applicability and Remedies for Enforcement in Israeli Law’ (1991) 16 Tel-Aviv University Law Review 131, 160 (in Hebrew). 55 HCJ 953/87 Poraz v Mayor of Tel Aviv 42(2) PD 309, 332 [1998] (in Hebrew). 56 Y Zilbershats, ‘The Right of the Majority to Choose its Residence’ (2001) 6 Law and Government 87, 96 (in Hebrew). 57 See HCJFH 4191/97 Recanat v National Labour Court 54(5) PD 330, 369–71 [2000] (in Hebrew). 58 See APA 343/09 Jerusalem Open House for Pride and Tolerance v Municipality of Jerusalem (14 September 2010), Nevo Legal Database (by subscription) paras 53–59 of Justice Amit’s opinion (in Hebrew). 59 cf A Bendor, ‘Equality and Governmental Discretion – on Constitutional Equality and Administrative Equality’ in Y Marzel (ed), Shamgar Book, vol 1 (Tel-Aviv, The Israel Bar-Publishing House, 2003) 287, 302–03 (in Hebrew) (defining the right to ‘administrative equality’ as ‘the right to consistency’). 53 54
410 Michal Tamir ‘principle of non-discrimination in the procedural sense’ and it demands equality by law and therefore the obligation not to discriminate applies to authorities operating by law.60 In public law, this distinction has various implications. Although administrative authority cannot mar administrative or constitutional equality, the different types of equality generate different judicial criticism. As long as the claim pertains to violating the right to equality in its strong constitutional sense, the tendency is to assess the violation from a broad perspective of the authority’s activity, to evaluate the criteria applied and to intervene where necessary. On the other hand, as long as the authority defines criteria that do not presumably violate the value of equality in its constitutional sense, but rather distinguish between X and Y based on considerations within the authority’s realm of responsibility, the judicial criticism of the authority’s activity will focus on checking procedural or discretional deficiencies. Moreover, the application of constitutional equality is binding not only where constitutional law is concerned but, in a broader sense, in all areas of law.61 The distinction between constitutional equality and administrative equality should be weighed as part of the process of how and to what extent constitutional values are applied in horizontal relationships. My position is that whereby the dimension of constitutional equality firmly permeates private law, the dimension of administrative law permeates private law at a much weaker level. Thus, the distinction yields a typological axis of equality obligations that starts with imposing the administrative equality obligation and ends with imposing constitutional equality obligations in order to promote the empowerment of weakened groups in private law itself. As long as the required obligation is one of constitutional equality, the tendency will be to apply it since – to my understanding – within the scope of the indirect applicability model, the equality value permeating the law is constitutional equality. Of course this does not imply that constitutional equality fully applies to private law since injecting the value from constitutional law into private law is just the beginning of the task, which ends with establishing a balance between the values. Nonetheless, using the distinction helps during both stages. First, it helps to stress the relevant aspects of equality that need to be injected into private law; secondly, it helps to grant constitutional equality the relevance it deserves within the scope of the balance. C. Whether the Rights are Waivable The difference between protecting a right in public law and the ability to waive it in private law stems from the basic perception of human rights in democracy. Human rights are perceived as a defence against governmental power and as such, the government is not entitled to violate them other than in the light of criteria of constitutionality and proportionality. On the other hand, in the private sphere, human rights are granted to the individual as freedoms and as such, at least on a basic level, the individual is entitled to detract from them by applying free will. Accordingly, a waiver of rights in the public sphere depends on its constitutionality, that is the permission must be in a law, for a proper purpose and in a proportional manner. However, elsewhere, I expressed my See Benkler (n 54) 160. Tamir (n 53) 112, 118–22.
60 61
Human Rights in Private Law 411
fundamental position whereby a waiver of constitutionality in private law would be best assessed through the prism of public order.62 Another issue is the question of at which level the waiver should be discussed – at the level of the parties among themselves (a level that is considerably developed within the scope of contract law) or at the level between the parties and the public as a whole through safety-valve concepts such as public order and probability. To a large extent, the answer to these questions depends on the ideological perception behind the law. Autonomous perceptions of private law, as the will theory, dictate a policy whereby due to the individual’s complete ownership of the legal right, one needs to adopt a position whereby she is entitled to waive her constitutional freedoms upon her will. On the other hand, more instrumental perceptions of private law – like the public attitude toward contract law or the philosophical interest doctrine – will grant significance to general public values and interests that deviate from the question of the free will of the individual waiving her rights. It appears as though these questions are still in their infancy.63 The waiver issue raises tough questions. In which areas can an individual waive her constitutional rights in private law? To what extent does the applicability of human rights in private law generate commodification of constitutional rights and to what extent can this commodification be broadened? For example, it is obvious that an employment contract imminently restricts human freedom, but the waiver does not entail any difficulty. Nonetheless, the more one penetrates the nuclear dimensions of constitutional rights, the closer one gets to its cogent dimension64 regarding which there is a consensus that it is unwaivable, even in private law.65 In this context, inspiration can be drawn from the constitutional development of the perception of human dignity in Israeli law. Kvod ha-adam (human dignity) is a basic right with many facets whose interpretation encompasses a set of derived rights.66 Because of this complexity, various models were proposed in Israel to clarify the constitutional right to dignity and define its scope. Yehudit Karp suggested an interesting interpretation by likening human dignity to a stone cast into a lake and making ripples: One could perceive the value of ‘human dignity’ as surrounded by ripples of content, as if the legislator cast the ‘human dignity’ stone into the waters of the Lake of Basic Laws. Its impact on the water creates ever expanding ripples which bounce and feed off each other. Each ripple is the outcome of another ripple and they flow toward each other and distance themselves from their source until they fade away.67 See M Tamir and A Harel, ‘On Human Dignity and Privatization’ (2011) 41 Mishpatim 663, 677–82. See D Snyder, ‘The Law of Contract and the Concept of Change: Public and Private Attempts to Regulate Modification, Waiver, and Estoppel’ (1999) Wisconsin Law Review 607, 673 (explaining that the concept of waiver is problematic and hard to define). 64 cf Y Shany and O Ben-Naftali, International Law between War and Peace (Tel-Aviv, Ramot, 2006) 197– 201 (in Hebrew) (discussing cogent attributes of human rights in domestic and international law). 65 Eg among critical thinkers, there is a discussion being held over the question of whether it is appropriate to recognize prostitution contracts, which reflect a flagrant and grave violation of women’s human dignity, if only to enable the woman to enjoy the agreed-upon consideration. See Keren (n 3) 273–74 (referring to MJ Radin, ‘Market Inalienability’ (1984) 100 Harvard Law Review 1849; MJ Trebilcock, The Limits of Freedom of Contract (Cambridge, Harvard University Press, 1993)). 66 See O Kamir, ‘Honor and Dignity Cultures: the Case of Dignity and Human Dignity in Israeli Society and Law’ in D Kretzmer and E Klein (eds), The Concept of Human Dignity in Human Rights Discourse (Leiden, Martinus Nijhoff Publishers, 2002) 231, 236–345 (elaborating four different terms in the English language expressing four aspects of dignity). 67 Y Karp, ‘Some Questions on Human Dignity According to Basic Law: Human Dignity and Liberty’ (1995) 25 Mishpatim 129, 137 (in Hebrew). 62 63
412 Michal Tamir There appears to be a consensus whereby a person cannot be allowed to waive the ‘core’ ripple, even in exchange for a contractual or other benefit she may receive.68 However, the further one gets from the core, toward more far-reaching circles of dignity, the greater the potential to forego a right in exchange for contractual gain.69 I agree with the claim whereby the issue of the waiver of rights cannot hold water in its own right; rather, it must in any case be ascertained whether the behaviour of one of the parties to the contract infringes upon the right of the person waiving her right to the extent that public order is affected. Within this context, one must take into consideration the overall competing values, inter alia, the value pertaining to the autonomy of the individual will of both parties when entering into the contract.70 Thus, awareness of the individual’s ability to waive her right in private law could – in certain cases – lead to a weakened defence of public values within the scope of indirect applicability and – in other cases – could actually reinforce their defence. Thus for example, the fact that there are so many ways of contracting an alliance could omit claims of discrimination or violation of the contractee’s human dignity because of her ability to use free market tools in order to ensure her well-being in an egalitarian way. D. Nature of the Entities This parameter addresses another crucial development, namely that of the publicization of certain private entities and their definition as hybrid bodies. In the 1980s and 1990s in Israel – as a result of the policy of several governments – there was extensive privatization71 which led to a growth in the private sector at the expense of the public sector and to the conversion of a great many public products from the public supply model to the marketing supply model.72 Israeli law underwent two key processes simultaneously, which – to a certain extent – were a reaction to privatization and led to the application of public law to private law players. The initial process at the core of this chapter is the establishment of the indirect applicability model and the influx of values horizontally within private law.73 The second process is the courts’ growing willingness to reclassify – by means of functional tests – incorporated entities in private law and define them as hybrid bodies to which certain norms from the realm of constitutional and administra-
68 cf DN MacCormick, ‘Rights in Legislation’ in PMS Hacker and J Raz (eds), Law, Morality and Society: Essays in Honor of HLA Hart (Oxford, Clarendon Press, 1977) 189, 197 (dealing with unwaivable rights). See also, CA 2145/92 State of Israel v Gueta 46(5) PD 704, 725 [1992] (in Hebrew) (Justice Alon discusses basic rights which are part of the human spirit and hence cannot be waived). 69 Thus, in labour law there are cogent rights, intended to protect employers, and hence unwaivable. See O Yadlin, ‘“Good Faith” in Employment Relations: From Class to Individual’ (1999) 22 Tel-Aviv University Law Review 867, 869. 70 cf Woolman (n 30) 122–23 (claims that there is no such thing as a waiver – what appears to be a waiver is actually the interpretation of the right). 71 See A Shapira, ‘Central Privatization Processes in Israel’ (2010) 64 Parliament (in Hebrew): www.idi.org.il/ Parliament/2010/Pages/64_2010/C/c_64.aspx. 72 Y Dotan and B Medina, ‘Legal Aspects of the Privatization of the Supply of Goods and Services’ (2007) 37 Mishpatim 287, 292–95 (in Hebrew). 73 D Barak-Erez, ‘Human Rights in an Era of Privatization’ (2001) 8 Labour, Society and Law 209, 213 (in Hebrew); D Barak-Erez, ‘Civil Rights and Privatization in Israel’ (1999) 28 Israel Year Book of Human Rights 203, 207–12, 215–16.
Human Rights in Private Law 413
tive law apply.74 The Kestenbaum case is the most prominent in this area since it reflects both streams simultaneously. The significance of both processes is ultimately the application of norms from public law to incorporated entities that function in private law. Both paths show a willingness to blur the distinction between the private and the public. Nonetheless, a distinction must be made between them. The relevant difference for our purposes is that once a body is defined as a hybrid one, its relationship with the individual grows significantly closer to a vertical relationship between individual and authority. It is perceived as a body that essentially activates forces similar to those activated by the governmental authority, and thus the scope of the application of public norms is commensurate with the body’s degree of functional publicization. On the other hand, in cases of the application of human rights in private law, the horizontal relationship between the individuals themselves are maintained. In these cases, the applicability of human rights – according to the model of Israeli law – is only indirect and grants sub-constitutional entitlements according to the constitutional values being applied. Despite the fundamental difference between both models, one can sketch an axis of sorts, whose starting point is an entirely private body and end point, a statutory governmental authority. Along the way, there are the private entities to which certain obligations pertaining to public law may apply. The more we are willing to accept this typological continuum, the greater the incentive to make use of functional tests that help the court to assess the hybridity of a body for the purpose of evaluating the question of whether public law values should apply to it within the scope of indirect applicability. When the court evaluates an incorporated body in private law in order to determine whether it is a hybrid body, preference is given to the essence (nature) of the body over the formalistics involved (its form of incorporation). Within the scope of the essence, various characteristics are taken into account, such as the extent of the impact on public interest,75 the degree of proximity to government activity,76 etc. When implementing the indices, the court may definitely reach the conclusion that the body standing before it does not satisfy enough criteria to impose obligations pertaining to a hybrid body, but satisfies enough of them to justify the imposition of lesser obligations by virtue of the indirect applicability of human rights in private law. An illustration of such can be seen in the verdict of the Jerusalem Magistrate’s Court which addressed conflicting transaction laws.77 The Court imposed increased trust obligations on the bank in its capacity as a professional entity between the parties and based on the perception that despite its being an autonomous contractee in private law, the trust obligations applicable to it are similar to public law obligations.78
74 See CA 3414/93 On v Diamond Exchange Plants 49(3) PD 196, 206 [1995] (in Hebrew) Justice Zamir’s opinion. 75 See E Benvenisti, ‘The Applicability of Administrative Law to Private Bodies’ (1994) 2 Law and Government 11, 32 (in Hebrew). 76 See D Barak-Erez, Administrative Law, vol 1 (Tel-Aviv, Israel Bar-Publishing House, 2010) 15 (in Hebrew). 77 HP (Jm) 2247/03 Reshef v Levi (16 March 2004), Nevo Legal Database (by subscription) (in Hebrew). 78 See M Rubinstein and B Okon, ‘The Bank as a “Social Agency” III’ in Y Marzel (ed), Shamgar Book, vol 3 (Tel-Aviv, The Israel Bar-Publishing House, 2003) 819, 832 (in Hebrew).
414 Michal Tamir E. Type of Property Political and social changes such as the ‘constitutional revolution’79 or the privatization policy have also led to changes in basic philosophical and legal concepts such as the concept of property. In the spirit of the general philosophical clash between the public and the private, one can pinpoint two dominant rival perceptions of property in society and in Israeli law.80 The first one is the conservative libertarian proprietary perception that presupposes entitlement and long-term validity of ownership; the second one – which is best described by Hanoch Dagan81 – is the perception of socio-public property for which the justifications for private property are more communitarian.82 Based on the social perception, the concept of property must not only be understood in its libertarian, autonomous context, but also in its broader social context, just like the relationship between individuals among themselves and individuals and the community. Along with ownership and financial value, the property reflects a set of other values such as freedom, justice and more. Accordingly it is sometimes appropriate to deviate from formal entitlement granted by traditional proprietary rights and incorporate other considerations likely to expand or restrict the proprietary right. On the other hand, the conservative approach’s premise is the opposite. As a counter-reaction to the publicization of property laws, this perception aims to maintain long-term validity of individual ownership. It appears as though Israeli law has overlooked the dichotomy between ‘private property’ and ‘public property’ and is aware of the fact that there are assets involving aspects of private property coupled with aspects of public property.83 The literature has developed a typology of proprietary institutions, which makes it possible to create a spectrum between private property and public property. There are several important distinctions establishing this typology. The first distinction pertains to the nature of the property itself. In this context, one can distinguish between personal property (or founder property) and interchangeable property.84 The first end of the spectrum consists of non-convertible proprietary assets which are identified with and directly reflect the owner’s personality (such as a wedding ring). The other end of the spectrum consists of proprietary assets which serve the purpose of financial profit only and are convertible since they have no personality-related value (such as office buildings for rent). Different proprietary assets become meaningful in this sense based on their location between the two ends of the spectrum.85 The second distinction pertains to the relationship between the parties. This is where different relationships are assessed (eg between individual and local authority/community settlement, etc). The first end of the spectrum boasts an
79 See MD Birenhack, ‘Just Property: Three Concepts of Property’ (2005) 21 Bar-Ilan Legal Studies 439, 442–62 (in Hebrew) (dealing with how the ‘Constitutional Revolution’ broadened the concept of property and abolished the dichotomy between private and public property). 80 For a general survey, see A Lehavi, ‘The Public Domain of Private Property’ (2006) 9 Law and Government 619, 621–23 (in Hebrew). 81 H Dagan, Property at a Crossroads (Tel-Aviv, Ramot, 2005) (in Hebrew). 82 For a general survey of communitarian conceptions, see S Gardbaum, ‘Law, Politics, and the Claims of Community’ (1992) 90 Michigan Law Review 685. 83 On (n 74) 204 (Justice Levin providing examples such as a university campus or sports stadium). 84 Lehavi (n 80) 624. 85 See MJ Radin, Reinterpreting Property (Chicago, University of Chicago Press, 1993) 79–80.
Human Rights in Private Law 415
intimate relationship between the individual and his partner (reflected in partnership laws) while the other end features relationships between strangers.86 In this spirit, and upon the indirect (and even ‘reinforced’) applicability of objective principles seeping from public law to the private, it seems appropriate that the balance between the rights will be influenced by the type of property in question in light of the above parameters. Indeed, there is opposition to adapt public values in the realm of property laws due to the fact that the existing laws reflect the balances which the legislator deemed right.87 On the other hand, there are those who claim that in Israel these mechanisms do not always work – for example where the housing capacity of minorities is concerned.88 Either way, it appears as though even in earlier, less developed periods, common law had the knack of embodying ‘public good’ considerations within the scope of proprietary entitlements.89 Therefore in the present era – primarily within the scope of reinforced indirect applicability – the need to weigh the characteristics of property has grown. F. Type of Contract Different types of social and economic alliances are characterised by different types of contracts. First, there are general ‘contractual patterns’ whose importance is recognised by the law for the purpose of designing a certain contract. For example, the relational contract90 – the contractual pattern that sometimes dictates increased obligations between the parties to the contract – inter alia, because of their prolonged relationship, which generates mutual dependence;91 or the standard contract – the contractual pattern that characterises contracts in which one of the parties to the contract is the stronger, more sophisticated one whereas the other party is part of the alliance but has no true control over the contents of the contract.92 Secondly, there are types of alliances that are regulated by specific contract legislation. Prominent illustrations of such in Israeli law are the insurance contract,93 the sales contract,94 the rental contract,95 the present contract,96 the contract work contract97 Lehavi (n 80) 627. cf ibid 640–42 (asserting that one should not apply general doctrines in a private dispute, when there are legal and administrative mechanisms for the decision). 88 H Sandberg, The Land of the State of Israel – Zionism and Post-Zionism (Jerusalem, The Sacher Institute, 2007) 15–25, 25–35, 37–49, 149–51 (in Hebrew). 89 The common law courts have recognized the right to ‘rule’ a property regarding property affected with a public interest. Some types of property were deemed ‘public’, and thus could not be privately owned. See A Reichman, ‘Professional Status and the Freedom to Contract: Towards a Common Law Duty of NonDiscrimination’ (2001) 14 Canadian Journal of Law and Jurisprudence 79, 89–124 (bringing the examples of the market place or highway). 90 I Macneil, ‘The Many Futures of Contracts’ (1974) 47 Southern California Law Review 691; I Macneil, Contracts: Exchange Transactions and Relations, 2nd ed (1978); I Macneil, ‘Relational Contract Theory: Challenges and Queries’ (2000) 94 North Western University Law Review 87. 91 See Keren (n 3) 128–30. See also LCA 1185/97 Milgrom’s Estate v Mishan 52(4) PD 145, 160 [1998] (in Hebrew) (dealing with a standard contract between a person and a retirement home as a relational contract, characterized by long-term and complex relations). 92 A Bin-Nun, Law on Standard Contracts 5743-1982 (Jerusalem, The Sacher Institute, 1987). 93 Insurance Contract Law, 5741-1981. 94 The Sale Law (Apartments), 5733-1973. 95 Rental and Borrowing Law, 5731-1971. 96 The Gift Law, 5728-1968. 97 Contract for Services Law, 5734-1974. 86 87
416 Michal Tamir and more. For the most part, specific contractual regularization reflects the balancing of values. However, in many cases safety-valve concepts enable the court to carve content into the statutory balancing system.98 This content may also express public values within the scope of their indirect application in private law. The history of the application of human rights in the area of contract law was attentive to the relationship between the parties to the contract. Thus, for example, the rights to freedom of occupation, equality and association were recognised within the scope of contractual employment relations, which constitute a long-term and ongoing relationship; the latter boasts an almost permanent balance of power of the strong versus the weak and does not merely reflect the proprietary aspect of employment, but also the personality-related and creative aspect of the subject in the realm of employment.99 The perception ensuing from the Supreme Court ruling is that there is strong justification to apply human rights norms – and primarily equality norms – to ongoing relationships. Presumably the trend should be pursued in this area and the nature of the contractual relationship assessed. The key criterion in this area is that the more the contract type reflects power gaps between the parties, and the more the contract meets ‘public’ criteria, the greater is the tendency to apply public norms to a broader extent. Thus for example, for contracts in which one of the parties had no impact on their formulation, such as standard contracts100 or collective agreement, the tendency to intervene should be greater.101 Moreover, it appears as though another stage that should be aspired to is the development and elaboration of particular safety-valve concepts in specific contractual legislation in order to rely on them and not only on the conceptualization in general contractual legislation102. G. Nature of the Parties to the Contract This parameter examines whether the parties have equal power or whether there is in an inherent imbalance of power between them. In this area as well, there appears to be an axis of sorts at one end of which there are contracts drawn in a situation where there is equality in the power relationship between the parties; at the other end of the axis there are contracts that bring out significant power gaps between the parties. The question of defining the indices according to which the power gaps will be assessed is one that calls for an influx of public values. Hilla Keren places the identity of the person being discriminated against in the contractual alliance at the core of the assessment of the application of obligations to equality in contract law.103 According to her feminist outlook, which is critical of the law, the Eg the interpretation of the term prejudicial condition in a standard contract. See, eg Recanat (n 59) (dealing with the equality right of flying attendants in regard to their retirement age); HCJ 6845/00 Niv v National Labour Court 56(6) PD 663 [2002] (in Hebrew) (dealing with discrimination of women in regard to their retirement agreements). 100 See G Shalev, ‘Contract Law’ in I Zamir and S Colombo (eds), The Law of Israel: General Surveys (Jerusalem, The Sacher Institute, 1995) 197, 210 (‘Standard contracts constitute, ab initio, a deviation from the principle of freedom of contract, in that the formation of such contracts does not express freedom, and their formulation is a unilateral enterprise’). 101 See HCJ 104/87 Nevo v National Labour Court 44(4) PD 749, 768 [1990] (in Hebrew). 102 See, eg Rental and Borrowing Law, 5731-1971, s 22. 103 Keren (n 3) 297–325, 355–79. 98 99
Human Rights in Private Law 417
routine development of contract law is the outcome of an assumption whereby the ultimate contract is between two businessmen. Nonetheless, reality has shown that contractual alliances vary greatly and that it is always necessary to assess the relationship between the parties to the contract. Accordingly, contract law must focus – where appropriate – on the identity of the weak party and embody, through the doctrine of contract law, the power gaps between the parties. This is all the more so in cases when one of the parties to the contract does not have any other practical options; she does not have the option to choose and is the weak party whose contractual freedom is very limited or completely inexistent. Under these circumstances, the judicial intervention in the formal contractual freedom is designed to promote essential contractual freedom.104 Along with more extensive use of existing contractual tools and the need to take into account the gaps between the parties within the context of other civil laws such as tort law,105 it would also be appropriate for the assessment of the applicability of human rights in private law to take into account the need to minimise gaps between strong and weak parties.106 V. CONCLUSION
In this chapter, I attempted to suggest criteria for the balance called for in private law between human rights within the scope of the indirect applicability model. These criteria are not exclusive, and there may be other relevant variables. Moreover, it is presumably not right to develop excessively rigid tests since one of the advantages of common law stems from its ability to develop ad hoc tests and balancing according to the circumstances of concrete cases brought before it. Nonetheless, the criteria comprise a basic categorization of the key questions that the court faces when it has to establish a balance stemming from the application of constitutional principles in private law. The criteria set forth a series of basic typologies that grant the court tools for delimiting the questions and understanding the basic levels which make it possible to weight the constitutional values. The inference of applying these criteria to the Kestenbaum case is, in my opinion, that Justice Barak’s grounds must be cumulative and not alternative. If one were to select a different scenario, in which the burial society did not control most of the burials in Jerusalem and the widower could choose among different burial societies – it could undoubtedly be argued that a person who chose to enter into a contract with a company that championed Hebrew inscriptions waived his right. In fact, Justice Barak was the one who stressed the ability of individuals to waive rights: Indeed, we enable individuals – in well-defined areas – to waive their basic rights to a certain extent (but not fully or essentially). This waiver naturally affects the nature of the appropriate N Cohen (n 3) 132. See, eg CA 10064/02 Migdal Insurance Company Ltd v Abu Hana 60(3) PD 13 [2005] paras 33–40 of Justice Rivlin’s opinion (in Hebrew) (consideration within the scope of compensation for ‘lost years’ according to statistical earning aspects in cases in which the plaintiff’s earning potential is low because of his minority status). 106 For distributive justice in private law, see C Vinitzki and M Tamir, ‘Application of Distributive Justice in Private Law: Models’ (2012) 14 HaMishpat 469 (in Hebrew). 104 105
418 Michal Tamir balance between the different values. This balance differs from that accepted in public law since the human right infringement in this context is not founded on the victim’s will, but on governmental power.107
In the Kestenbaum judgment, Lionel Kestenbaum and the honour of his deceased wife were confronted with the Society’s contractual freedom. However, the economic right was enhanced by the Society’s aim to promote and sanctify the Hebrew language – an ideological aim at the Society’s core, which is also part of its dignity and freedom of expression. Accordingly, one cannot claim to be dealing with a clash between a civil right and an economic right, but rather a head-on collision between civil–political rights. If it were not a matter of a monopolistic company, would Kestenbaum’s waiver of his right to choose the language of the inscription on the tombstone be contrary to public order? In my opinion, the answer is no. Human dignity serves both as a reason and a rationale for all human rights, including the autonomy of individual will.108 In this case, it is not a matter of waiving one’s core dignity in the sense of ‘human dignity’, the waiver of which a democratic society cannot condone. The dignity that the judgment addresses pertains to much more far-reaching circles which a legally competent person – who has the option to choose – is entitled to relinquish in order to achieve his other interests and rights. It is against this background that one should interpret the criticism of Frances Raday of Justice Barak’s reasoning in the Kestenbaum case.109 Raday agreed with the outcome of the judgment, but thought that Justice Barak had chosen a formal model of ‘privatizing human rights’, which did not take into account the difference in respective strength and may ultimately undermine the objective of protecting the rights of the weak. Instead, she proposed a substantive model for privatizing human rights, which seeks to preserve the objective of human rights to protect socio-economic discrepancy. Nonetheless, there is a great deal of wisdom in the judgment rendered by Justice Barak in the Kestenbaum case. As aforementioned, using the argument of the right’s application to private law as grounds in their own right and part of the rationale for the decision (ratio decidendi) made it possible to apply the human rights doctrine in private law to Israeli law, as opposed to an obiter dictum. Thus, the tremendous significance of the judgment despite the fact that it is doubtful as to whether the three arguments set forth by Justice Barak were alternative rather than cumulative.
Kestenbaum (n 5) 535. HH Cohen, ‘On the Meaning of Human Dignity’ (1983) 13 Israel Year Book of Human Rights 226, 231. 109 F Raday, ‘Privatization of Human Rights and the Abuse of Power’ (1994) 23 Mishpatim 21, 32–34 (in Hebrew). 107 108
27 Private Actors and Constitutional Rights STEPHEN GARDBAUM*
T
HE ISSUE OF whether and how constitutional rights bind private individuals as well as government actors has been a formative and paradigmatic one in the growth and development of comparative constitutional law in recent years. Perhaps rivalled only by the topic of proportionality, this issue has been the subject of a fruitful dialectic between the theory and practice of comparative constitutional law that has helped to drive the discipline forward over the past two decades. Scholars have first observed the range of internal constitutional practices in various systems, then attempted to systematise the topic by providing conceptual and normative maps of the available options, and judges in turn have employed these maps to forge increasingly convergent answers to the general issue. Like the principle of proportionality, what is generally referred to as indirect horizontal effect, or the indirect application of constitutional rights, has come to be the dominant model within both the academy and the courtroom of comparative constitutional law. It is now, perhaps, part of the post-war paradigm.1 Remarkably, one person has single-handedly combined both roles in making a unique and seminal contribution to this topic. Even more remarkably, this same person has also made the same dual contribution to the topic of proportionality. I am referring, of course, to Aharon Barak whose chapter2 – a distillation of his thoughts and actions on the issue – I have the honour to comment on, alongside that of Michal Tamir.3 In his comprehensive and wide-ranging chapter, ‘Constitutional Rights and Private Law’, Justice Barak begins by setting out four models for answering the basic question of whether an individual’s constitutional rights apply only against the state (vertically) or also against fellow individuals (horizontally). In addition to describing and distinguishing the four models, he presents concise normative critiques of the first three before focusing for the remainder of the paper on the fourth model – indirect application – which, he argues, originated in the seminal Lüth opinion in Germany and has since become the dominant one comparatively. The second half of the paper provides a detailed explanation of how this model works in action both generally and in Israel, where it was adopted by the Supreme Court in the early 1990s as the result of landmark opinions written by Justice Barak. * This chapter was written for the Israeli Constitutional Law in the Making – Comparative and Global Perspectives Conference, 22–24 May 2011. 1 The term is Loraine Weinrib’s. L Weinrib ‘The Post-war Paradigm and American Exceptionalism’ in S Choudhry (ed), The Migration of Constitutional Ideas (Cambridge, Cambridge University Press, 2006). 2 Ch 25 in this volume. 3 Ch 26 in this volume.
420 Stephen Gardbaum In my comments, I shall be focusing on the first part of Justice Barak’s paper and, in particular, on his four models for answering the basic question. This is both because that part of the paper best enables me to fulfil my assigned function of helping to put the issue in the broader context of constitutional theory and global processes, and because some of my own past writing on the subject has attempted a similar task.4 Indeed, one of the things I shall be doing is identifying one small island of difference between us amidst a large sea of agreement. The first of Justice Barak’s four models is the ‘direct application’ of constitutional rights against private individuals, or direct horizontal effect as it is commonly labelled in the literature. In response to the familiar normative arguments for this comparatively uncommon position in practice, Justice Barak argues, first, that it is unnecessary to protect weaker individuals in society against stronger ones through constitutional rights as sub-constitutional law is sufficient for the task and, secondly, that the potential evasion and erosion of constitutional obligations via privatization of government services is adequately and more properly countered by rights-conscious judicial review of privatization statutes themselves, as well as by suitable changes in sub-constitutional private law. Second is ‘the non-application model’, in which constitutional rights not only apply directly vis-à-vis the state alone (as they do in most of the jurisdictions under consideration) but also are limited to public law and the public sphere without any penetration into private law and, hence, into inter-personal relations. The well-known Canadian case of Dolphin Delivery5 is cited as an example of this position. Justice Barak’s critique of this model is that ‘it creates too deep a rift between constitutional law and private law’6 and fails to view the legal system as a coherent whole reflecting and expressing certain fundamental, comprehensive, or overarching (ie ‘constitutional’) values and principles. The third model, stated to have been developed in American constitutional law, is ‘the application through the judiciary model’. Here, although constitutional rights are directed towards the state alone, the judge is deemed to be a state organ and so, like the others, must act consistently with constitutional rights. In particular, this means first that judge-made common law must be developed in line with constitutional rights and, secondly, that the judge ‘must act within the confines of the specific [private law] dispute in a fashion which conforms to the parties’ constitutional rights vis-à-vis the state’.7 According to Justice Barak, this first duty, exemplified by New York Times v Sullivan8 is sound; but the second, exemplified by Shelley v Kraemer9 is not. This is because it transforms a constitutional right to equality vis-à-vis the state into a constitutional right to equality vis-à-vis another individual, which judges are only empowered to do under the direct applicability model. Justice Barak argues that this second aspect of the application by the judiciary model, however, is not accepted within comparative law and has not 4 See S Gardbaum, ‘The “Horizontal Effect” of Constitutional Rights’ (2003) 102 Michigan Law Review 387; S Gardbaum, ‘Where the (State) Action Is’ (2006) 4 International Journal of Constitutional Law 760; S Gardbaum, ‘The Myth and the Reality of American Constitutional Exceptionalism’ (2008) 107 Michigan Law Review 391; S Gardbaum, ‘The Structure and Scope of Constitutional Rights’ in T Ginsburg and R Dixon (eds), Research Handbook in Comparative Constitutional Law (Northampton, Elgar, 2011). 5 RWDSU v Dolphin Delivery Ltd [1986] 2 SCR 573 (Can). 6 Barak (n 2) at 383. 7 ibid, at 384. 8 New York Times v Sullivan 376 US 254 (1964). 9 Shelley v Kraemer 334 US 1 (1948).
Private Actors and Constitutional Rights 421
been sustained in America. Rather, it has effectively been overruled by subsequent development of the US state action doctrine, so that the horizontal application of constitutional rights by judges is conditional on the individual against whom a rights claim is made exercising a state function. Finally, the fourth model is that of the ‘indirect application’ of constitutional rights, which holds that the individual’s constitutional rights apply directly against the state and indirectly against fellow individuals. Originating in the German Federal Constitutional Court’s 1958 decision in the Lüth case, the model is premised on the assumption that each of the individual’s subjective constitutional rights against the state also has an objective value that ‘radiates’ into all areas of the law, including private law. As a result, ‘Every provision of private law must be compatible with this system of [objective] values, and every such provision must be interpreted in its spirit’.10 Accordingly, although constitutional rights apply directly to the government only, they apply indirectly to private individuals via their influence on private law. By contrast, Justice Barak states that the US state action doctrine of the third model ‘operates outside of private law’.11 When he first developed his fourfold model in the early- and mid-1990s, Justice Barak was a pioneer in the field and writing on a largely blank slate. Those of us contemplating the field now do so standing on his shoulders. The main difference between Justice Barak’s conceptual map of the terrain and my own is a relatively small one, and may even ultimately boil down to labels or level of abstraction/generality – although I am not quite convinced of this yet. In essence, I believe there are three models or answers to the basic question of the scope of constitutional rights within a legal system and not four. In particular, I think there is no separate, independent or distinct third model – the model of judicial application. Rather, this ‘model’ is one way among several in which a legal system may institutionalise the position of indirect application. That is, it is one technique for achieving indirect horizontal effect, not an alternative to it. Similarly, the German model of conceptualizing constitutional rights as objective values that operate within private law is also another specific route to, or version of, indirect applicability and is neither a necessary one nor constitutive of this position itself – although, as I shall argue, Germany also routinely employs aspects of the judicial application technique. In other words, in my view, Justice Barak’s model of indirect application is actually a more general one than he conceives of it. What defines this more general position and distinguishes it from both the direct and non-applicability models is the proposition that all law (including all private law), but not all conduct, is subject to the constitution and its set of rights.12 In this way, although constitutional rights apply directly only against the government (contra the direct applicability model), they also apply indirectly against private individuals by governing the law that regulates their legal relations with one another (contra the non-applicability model). 10 Lüth, translated in DP Kommers, The Constitutional Jurisprudence of the Federal Republic of Germany (Durham, Duke University Press, 1997) 363. 11 Barak (n 2) at 387. 12 The fact that all law but not all conduct is subject to constitutional rights under this general position does not mean that only law and no conduct is also so subject. Thus, for example, the official conduct of state officials – including judges – in administering, executing and enforcing the law is also typically governed by the constitution under the basic vertical approach, which is included in, but does not exhaust, the model of indirect application. The main type of conduct that is excluded is that of private individuals.
422 Stephen Gardbaum Accordingly, I disagree that Germany and the United States adhere to different models or approaches to the basic question. Thus, in my view, Lüth and New York Times are in essence structurally identical instances of the same general position – indirect application. In both, the respective lower courts were held to have violated the constitutional rights to free speech of the claimants by applying the relevant private law without adequately taking these rights into account. In Lüth because the lower court failed to take the objective value of freedom of expression adequately into account in interpreting and applying ‘good morals’ under section 826 of the German Civil Code (BGB). In New York Times because the lower court failed to take the newspaper’s constitutional right to free speech adequately into account in interpreting and applying the common law of defamation. It is true that in Lüth, the Federal Constitutional Court (FCC) did not invalidate section 826 but required it to be interpreted in light of the objective value of free expression, whereas the US Court invalidated the state common law of defamation. But that was only because the very general and capacious language of section 826 could be so interpreted. If it could not have been, the FCC would have invalidated it under the first Lüth prong: ‘every provision of private law must be compatible with this system of values’.13 Consequently, the only real difference between Lüth and New York Times is the existence in Germany of a constitutional duty on the lower courts to interpret private law consistently with the constitutional value where possible; although if not possible, the FCC will invalidate that law as unconstitutional. Even here, however, there is a constitutional canon in the United States, if not a duty, that where a provision of law is ambiguous, courts should apply the meaning that is consistent with the constitution rather than exercise the power of judicial review, unless plainly contrary to legislative intent.14 And this difference may, in turn, simply reflect the institutional difference between decentralised and centralised judicial review – as in the latter, lower courts only have interpretive and not invalidation powers. In short, in Germany no less than the United States, constitutional rights are applied through the judiciary. Courts are bound by, and may violate, constitutional rights15 (indeed, the majority of successful constitutional complaints are against lower courts) and all private law – BGB and common law – is subject to them. This, in turn, means that in both countries although constitutional rights apply directly only against the government, they apply indirectly against individuals in the context of private litigation. As further support for my view that there is a single general position of indirect applicability deriving from the proposition that all law, including all private law, is subject to constitutional rights, let us briefly consider the case of Shelley v Kraemer that Justice Barak considers exemplifying the mistaken second prong of the separate and distinct application by the judiciary model. Recall that Justice Barak approves of New York Times as exemplifying the valid, but distinct, first prong of this model. Although Justice Barak argues that the principle he identifies as underlying the decision in Shelley – that 13 Lüth (n 10). See Gardbaum, ‘The “Horizontal Effect”’ (n 4) 405–06; G Taylor, ‘The Horizontal Effect of Human Rights Provisions, the German Model and its Applicability to Common Law Jurisdictions’ (2002) 13 King’s College Law Journal 187, 196–97. 14 Edward J DeBartolo Corp v Fla Gulf Coast Bldg & Constr Trades Council 485 US 568, 575 (1988). 15 As expressly stated in the Grundgesetz für die Bundesrepublik Deutschland [Grundgesetz][GG][Basic Law], 23 May 1949, BGBl I, Art 1(3) (Ger): ‘The following basic rights shall bind the legislature, the executive, and the judiciary as directly enforceable law’.
Private Actors and Constitutional Rights 423
within a private law dispute, the judge must act consistently with the parties’ constitutional rights vis-à-vis the state – is mistaken, he suggests elsewhere in his paper that the same result could, and perhaps should, be reached under his fourth model of indirect applicability. This involves balancing the objective value of equality against the objective value of the freedom to contract, finding the former weightier than the latter in the particular context, and so deeming the racially restrictive covenant void as against public policy.16 Thus, rather than holding that by enforcing a racially restrictive covenant a court violates the purchaser’s (external) constitutional right to equality vis-à-vis the state, the fourth model would hold that the covenant violates the (internal) contractual norm of public policy based on the objective aspect of the constitutional right to equality. Now although this is a difference, it does not seem like a significant enough difference to me to say that two separate and independent models of the application of constitutional rights to private individuals are at work here. In both cases, constitutional rights apply indirectly to individuals, rather than either directly or not at all. Especially if one notes that for jurisdictional reasons, the US Supreme Court could not have taken the approach that Justice Barak suggests would be the more proper one and the one that the Israeli Supreme Court would take. This is, of course, because for federalism reasons the US Supreme Court does not have the power to interpret or develop state private law but only the power to declare state action (including state private laws) constitutional or unconstitutional. But here, I think, is the important point: the FCC would likely analyze the issue raised by Shelley in a way that is more similar to the US than the Israeli Supreme Court. And if I am correct, this casts further doubt on the distinctiveness of the third and fourth models. Imagine a purchaser in Shelley’s position in Germany. As in both Lüth and Shelley itself, the German Shelley makes his constitutional rights claim in the lower courts but they reject it out of hand. The purchaser then files a constitutional complaint with the FCC against the lower courts, as in Lüth. The FCC is quite likely to find that the lower court violated the purchaser’s constitutional right to equality by not taking its objective aspect sufficiently into account in resolving the private law dispute. Again, for jurisdictional reasons – though for slightly different ones than in the US case – the FCC cannot itself interpret and apply the private law of contract but only to declare the action of the state court constitutional or unconstitutional, as in Shelley itself. The precise routes may differ between the US and Germany – the Supremacy Clause’s command that all state law is subject to the Constitution and the Fourteenth Amendment’s that no state actor shall deprive any person of equal protection versus the conception of constitutional rights as forming an objective order of values throughout the legal system – but the analysis and result is essentially the same. In both, state courts have violated an individual’s constitutional right to equality in how they resolved a private law dispute. My point in all of this is that the specific differences in how the US, German and Israeli courts handle the issue of the effect of constitutional rights in private law disputes – differences stemming to a significant extent from jurisdictional powers and limitations17 – Barak (n 2) section IIIDi. For other and more general aspects of the importance of jurisdiction on the issue, see M Kumm and VF Comella, ‘What is So Special about Constitutional Rights in Private Litigation? A Comparative Analysis of the Function of the State Action Requirements and Indirect Horizontal Effect’ in A Sajo and R Uitz (eds), The Constitution in Private Relations: Expanding Constitutionalism (Utrecht, Eleven, 2005). 16 17
424 Stephen Gardbaum do not seem sufficient to me to say they are giving different answers to the general question of the scope of constitutional rights within a legal system. Rather, they each adhere to the essential and general position of indirect application, or indirect horizontal effect, in which all private law is subject to constitutional rights – thereby indirectly subjecting private individuals to them. Once again, this is what distinguishes this inter mediate position from both the direct and non-applicability models. Accordingly, my own conceptual map of the terrain identifies three clear and distinct general positions. First, what Justice Barak refers to as the non-applicability model and I refer to as strong vertical effect: constitutional rights apply to public law and only those private law disputes in which the state is a party (for example, as employer or landlord). As Justice Barak correctly and insightfully notes, this is more restrictive than the very common basic vertical position in which constitutional rights apply directly to the state only because, as we have seen, this latter position does not rule out significant indirect impact on private individuals through its applicability to private law. The point of this first position is precisely to prevent constitutional rights from leaking into the private sphere at all. Secondly, indirect horizontal effect through the impact of constitutional rights on private law. I sub-divide this position into two, not as between judicial application and objective values (for the reasons above), but between a stronger and weaker version. The strong version of indirect horizontal effect is where constitutional rights apply directly to all private law (and so indirectly to individuals) because all private law must be consistent with it and all private law can be challenged as violating constitutional rights. By contrast, weak indirect horizontal effect operates where constitutional rights apply indirectly to private law; that is where they do not directly govern that law or apply directly to the private law dispute at hand. The typical mechanism for this indirect application to private law is the general power or duty of courts to interpret/develop private law, often the common law, in line with constitutional ‘values’ in the context of an ordinary, non-constitutional cause of action, as distinct from a specific rights claim. For the reasons given, in my book both Germany and the United States are strong indirect horizontal effect jurisdictions: Germany because, under Lüth, the interpretive duty is not exhaustive so that a provision of private law which cannot be interpreted consist ently with a right’s objective value is unconstitutional; the United States because all law is directly subject to the Constitution under its Supremacy Clause. Canada is an example of weak indirect horizontal effect because – unlike private law statutes – the common law at issue in private litigation is not directly subject to the Canadian Charter of Rights and Freedoms (despite the Charter’s Supremacy Clause, Charter rights are stated to bind only the legislative and executive branches of government, and not the judicial), but courts ought to take Charter rights into account in developing the common law.18 I am not sure into which of my categories Israel falls; this will likely depend on whether private laws that cannot be interpreted consistently with objective values are unconstitutional, as in Germany. Finally, at the other end of the spectrum is what Justice Barak calls the direct applicability model and I refer to as direct horizontal effect. On this last model, Justice Barak’s critique of the fairly standard arguments for direct applicability of constitutional rights against private individuals that I referred to above is expressed in universal terms. But his arguments might also be thought of as condi See Dolphin Delivery (n 5).
18
Private Actors and Constitutional Rights 425
tional. That is, where certain conditions apply, sub-constitutional law will be sufficient to address the problem of protecting weaker individuals in society against stronger, or of rendering private law suitable for privatization. Where these conditions do not apply, however, then there may be no sufficient alternative solution and only direct applicability can solve a genuine problem that sub-constitutional law is unable to do. This in fact seems to be one of the reasons for the rapid spread of the direct applicability model in many Latin American countries since the transition to democracy in the 1990s.19 Faced with either or both weak legislative systems, or ones captured by representatives of the powerful, courts have helped to fill the resulting vacuum and bring about greater protection of the weak through constitutional causes of action by one private individual against another, such as the writs of tutela (Colombia) and amparo (Argentina). Indeed, on a global level, this rapid recent development of the direct applicability model in Latin America is a significant trend and cuts against too easy a dismissal of this model in terms of comparative constitutional experience. In ‘Human Rights in Private Law: Hybridization of the Balancing Tests’, Michal Tamir takes as her starting point the Israeli Supreme Court’s choice of the indirect application model and proposes a framework for its judicial implementation in concrete cases, primarily in contract law. As she views it, the model results in frequent clashes of rights between ‘the autonomy of the free will at the root of private law’ and ‘the fact that human rights are inherent to all human beings, including vis-à-vis other individuals and not just the government’.20 Such clashes that have not been resolved by (valid) legislation, she argues, should be assessed in the light of all their particular circumstances, and the main part of the chapter proposes several principles or criteria to serve as guidelines for courts in performing the required case-by-case assessments. These principles include the substance of the clashing rights (eg the normative hierarchy of the human right at issue, whether a negative or a positive right is at stake), whether ‘constitutional equality’ (the obligation to refrain from discriminating against members of suspect groups) or only ‘administrative equality’ (the requirement of consistency in the treatment of like cases) is involved, whether the rights are waivable, the nature of the entities involved, and the type of property or contract under consideration. These principles provide a basic categorization of the key questions the court should ask when balancing private law values and human rights under the indirect applicability model. Tamir applies them to justify the outcome of the actual Kestenbaum case,21 in which the Jewish Burial Society held a monopoly, offering the widower little choice, and so was treated as a hybrid body subject to certain public law obligations, and to show how a hypothetical version in which it competes on language policy with other burial societies ought to affect the balance, due to a clearer waiver of rights. She also suggests that a private landlord is more likely to be able to discriminate among lessees in rent charged (administrative equality) than on the basis of race or sexual orientation (constitutional equality). The chapter clearly and thoroughly identifies the major legal and factual variables that might be used to distinguish cases and their outcomes under the model of indirect application as it has been accepted in Israel. One possible concern with such a context 19 See W Rivera-Perez, ‘International Human Rights and the Doctrine of Horizontal Effect in Latin America’ (SJD dissertation, UCLA School of Law, 2010) (on file with the author). 20 Tamir (n 3) at the beginning of part IV. 21 CA 294/91 Jerusalem Community Burial Society v Kestenbaum 46(2) PD 464 [1992] (in Hebrew).
426 Stephen Gardbaum sensitive, all-things-considered approach is that it may potentially exact too high a cost from individuals in terms of the standard rule of law values of legal certainty and predictability to make this a suitable solution for private law disputes. It is perhaps one thing to subject parties to the uncertainties of a multi-pronged balancing exercise in the context of constitutional adjudication and another in the context of everyday private law disputes unregulated by legislation. It is true that ordinary negligence law, for example, often imposes the ex ante indeterminacy of the reasonableness standard on individuals, but whether doing X is reasonable in the circumstances is still a somewhat more constrained exercise than weighing all the factors in such a multi-pronged test. A second reaction is that, at times, there is some confusion in her account between the direct and indirect application models. So, on the one hand, Professor Tamir’s capacious conception of human rights quoted above appears to envisage their direct application to private individuals. On the other, however, she seems to deem it important (and struggles) to distinguish the ‘firm permeation’ of private law by the principle of constitutional equality that she proposes from its ‘full application’ to private law,22 as if this latter would be incompatible with indirect application rather than perfectly consistent with it. These quibbles aside, this chapter is a fine and helpful systematization of the issues faced by Israeli courts in the process of applying Justice Barak’s favoured fourth model.
Tamir (n 3) at 410.
22
28 The National Security Constitution and the Israeli Condition DAPHNE BARAK-EREZ*
I. INTRODUCTION
I
SRAEL OFFERS A unique case study for the assessment of the constitutional regulation of national security. The special value attached to the Israeli example derives from several factors: first, Israel has experienced continuous existential threats ever since its establishment. Therefore, in Israel, the constitutional regulation of national security is not only a matter for the law books. It is the result of a constant challenge to both the existence of the country and the rule of law.1 Second, national security threats have served as one of the reasons working against the aspiration to finalise Israeli constitution-making. Third, Israel presents a working laboratory to one of the constitutional models for dealing with national threats – a model based on the power to declare an emergency regime. Against the background of the controversy around this model, Israeli law serves as an example of both its advantages and weaknesses. Fourth, the Israeli Supreme Court also practices judicial review in matters related to the military and national security, and thus makes the regulation of security matters in Israel a living legal reality. The following analysis is aimed at reviewing and evaluating the Israeli case study, taking into consideration its unique traits as well as its potential contribution to understanding the relative advantages and disadvantages of models for regulating emergency conditions in other systems. Following this Introduction (which serves also as part I), part II describes the basic features of the regulation of security matters in Israeli law and part III elaborates on the issues regulated on the constitutional level by provisions of the Basic Laws. Part IV takes the analysis another step forward by describing the de facto application of the constitutional regime in this area. Part V offers another perspective on the Israeli case study by looking into other models of regulating the issue of emergency conditions in other legal systems. Two other important aspects discussed at this stage are the impact of judicial review on the de facto regulation of security issues (part VI) and more specifically the impact of norms of international law as applied by the courts * The author would like to thank Liav Orgad for his comments.
1 See also P Lahav, ‘A Barrel Without Hoops: The Impact of Counterterrorism on Israel’s Legal Culture’ (1988) 10 Cardozo Law Review 529; D Barak-Erez, ‘Israel’s Anti-terrorism Law: Past, Present and Future’ in VV Ramraj, M Hor, K Roach and G Williams (eds), Global Anti-Terrorism Law, 2nd edn (Cambridge, Cambridge University Press, 2012) 597.
430 Daphne Barak-Erez (part VII). Towards the end, part VIII offers a relatively new perspective on the Israeli case study, by looking into the impact of new laws which set special securityoriented norms and limit their durability to a limited period of time (using ‘sunset’ provisions). This development is also evaluated in the conclusion vis-à-vis the more traditional building blocks which constitute the regulation of emergency conditions in Israel (part IX). II. FOUNDATIONS
Israel did not fully accept a formal constitution as promised by its Declaration of Independence2 in part because its establishment was accompanied by constant existential threats. Within these circumstances, the idea of a constitution was risky in more than one way. On the one hand, having a constitution necessarily meant the limitation of government powers during a time in which it was vital to give the government maximum strength in order to secure the prospects of the state to protect itself against its enemies. On the other hand, a constitution that would have merely reflected the necessities of the time could not set a vision for the country. This could have been an unbalanced and potentially dangerous constitution. Eventually, the compromise of refraining from enacting a formal constitution (and instead legislating Basic Laws that would eventually be consolidated into such a document)3 proved to be the pragmatic solution in several contexts including, namely, national security.4 Lacking a formal constitution, the normative framework of national security law in Israel has found its grounding in legislation, judicial precedents, and later, also in Basic Laws. (1) Legislation – statutory law in this area has traditionally included both legislation originating from the days of the British Mandate in Palestine5 and original Israeli legislation. The Law and Administration Ordinance6 provided the first constitutional principles in this area, although formally speaking it only had the force of ordinary legislation. This Ordinance was enacted in the first days of the state and was aimed at reflecting the main principles regarding the roles and powers of the various branches of government. The most important provision of the Ordinance in this context was section 9, which acknowledged Parliament’s power to declare an ‘emergency situation’. It further stated that when the government declares an ‘emergency situation’ it has the power to promulgate ‘emergency regulations’ that carry the power to abolish or amend any existing law
Declaration of the Establishment of the State of Israel, 1 LSI 7 (1948). This compromise is known as the ‘Harari Decision’: ‘The first Knesset charges the Constitution, Law and Justice Committee with the task of preparing a constitution for the country. The constitution will be built chapter by chapter, so that each one will in itself be a basic law. The chapters will be submitted to the Knesset as the Committee concludes its task and, together, all these chapters will become the constitution of the country’. DK 5 (1950) 1743 (in Hebrew). 4 Another area of controversy which posed difficulties was the regulation of law and religion. For more background on the early stages of the Israeli constitutional process, see D Barak-Erez, ‘From an Unwritten to a Written Constitution: The Israeli Challenge in American Perspective’ (1995) 26 Columbia Human Rights Law Review 309. 5 The main statutory framework from this time is the Defence (Emergency) Regulations, 1945, Palestine Gazette 1442, 1055 (Isr). 6 The Law and Administration Ordinance, 5708-1948. 2 3
The National Security Constitution 431
for a period not exceeding three months. Indeed, section 9 succinctly included the basic features of the constitutional regulation of national security in Israel: the power of the Parliament to wave its sovereignty and sole power to legislate by declaring a state of emergency; and the power of the government to overstep the boundaries set between the executive branch and the legislative branch for a limited period of time. It reflected a willingness to grant the executive branch additional power, subject to Parliament’s final word – its power to revoke the declaration of the emergency situation. In addition to the power to promulgate emergency regulations, Israel enacted several laws that included provisions stating that they would be in force only when a declaration of a state of emergency is in force.7 (2) Judicial precedents – the Israeli Supreme Court, sitting as the High Court of Justice, has also opened its gates to petitions in matters of national security from its very first days. It is possible to indicate petitions in matters such as judicial review of administrative detentions even from the time of the War of Independence.8 However, until the 1990s, the scope of judicial review was limited to the invalidation of executive decisions.9 Legislation was immune from the power of judicial review, following the British tradition of sovereignty (which has been gradually deserted later on in both Israel and Britain).10 Thus, executive decisions in the area of national security were subjected, first and foremost, to the ultra vires principle, meaning that the government, including the military, cannot act without statutory authorization.11 (3) Basic Laws – among the Basic Laws, the most important one for the foundations of national security law is Basic Law: The Government,12 which defines the powers of the executive branch. This Basic Law currently includes the updated regulation of the issue of an emergency situation (replacing the Law and Administration Ordinance in this regard, as detailed below). Another important aspect of the constitutional regulation of national security is found in Basic Law: The Military, which clarifies the subjection of the military to the government. Section 2(a) of this Basic Law states in this regard: The Army is subject to the authority of the Government.
Other provisions of the Basic Law detail the chain of command and control of the military by the government. Section 3(b) of the Basic Law states that [t]he Chief of the General Staff is subject to the authority of the Government and subordinate to the Minister of Defence.
7 See, eg Emergency Land Requisition (Regulation) Law, 5710-1949; Commodities and Services (Control) Law, 5718-1958. A notable example of a newer law that applies only in times of emergency is that of the Emergency Powers Law (Detention), 5739-1979, authorizing the use of administrative detentions. These laws were designed to serve as ‘dormant’ legislation. In practice, however, they have not been dormant in any practical sense – they are in force de facto since their legislation (due to the continuous nature of the declaration of emergency situation, as described later). 8 HCJ 7/48 Al-Karabutli v Minister of Defence 2(1) PD 5 [1949] (in Hebrew). 9 See Barak-Erez, ‘From an Unwritten to a Written Constitution’ (n 4). 10 In Britain, laws are currently subject to judicial review based on the Human Rights Act, 1998. In Israel, the change has developed gradually during the 1990s, as explained below in part VI. 11 In this sense, Israeli law presents an example of the ‘legislative’ model of terrorism law. For this terminology, see, D Barak-Erez, ‘Terrorism Law between the Executive and Legislative Models’ (2009) 57 American Journal of Comparative Law 877. 12 On the various versions of this Basic Law, see nn 13–16 and the accompanying text.
432 Daphne Barak-Erez III. THE EMERGENCY CONSTITUTION AND THE BASIC LAWS
As indicated, the constitutional basis for Israeli national security law was incorporated into Basic Law: The Government. The first version of this Basic Law, enacted in 1968,13 did not address the issue, but it was eventually regulated by Basic Law: The Government from 199214 and Basic Law: The Government from 2001,15 the current valid version of this Basic Law.16 Section 1 of Basic Law: The Government states: The Government is the executive authority of the State.17
Thus, it does not authorise the government to act beyond its statutory powers.18 Indeed, the government has residual power to act in areas not regulated by existing legislation,19 but this power has been interpreted as inapplicable to actions which infringe human rights or to new initiatives aimed at the regulation of new areas of law (where power lies only with the legislature).20 Therefore, the government may take steps considered necessary for safeguarding national security, when these are not legislated, only within its power to promulgate emergency regulations (formerly based on section 9 of the Law and Administration Ordinance, and currently on Basic Law: The Government). Sections 38 and 39 of Basic Law: The Government adhere to the principles of the regulation of emergency situations as they had been originally drafted in 1948.21 According to section 38, the power to declare an emergency situation is given to the Knesset, the Israeli Parliament. Section 38(a) states: Should the Knesset ascertain that the State is in a state of emergency, it may, of its own initiative or, pursuant to a Government proposal, declare that a state of emergency exists.
When such a declaration is in force, the government is empowered to promulgate emergency regulations that may change current laws, for a period not exceeding three months. Section 39(a) states that [d]uring a state of emergency the Government may make emergency regulations for the defence of the State, public security and the maintenance of supplies and essential services, Basic Law: The Government, 5728-1968, SH No 540, p 226. Basic Law: The Government, 5752-1992, SH No 1396, p 214. 15 Basic Law: The Government, 5761-2001, SH No 1780, p 158. 16 The two earlier versions of Basic Law: The Government are identical on this point. The following analysis reviews ss 38–39 of Basic Law: The Government from 2001 (which rehearse the wording of ss 49–50 of the Basic Law from 1992). The differences between the two versions are connected to the fact that Basic Law: The Government from 1992 tried to introduce direct elections to the position of the Prime Minister, a reform which was abolished by its predecessor from 2001. They do not have a bearing on the regulation of emergency regimes, the focus of this analysis. 17 All the citations from the Basic Laws are taken from the English translation published on the Knesset website: knesset.gov.il/description/eng/eng_mimshal_yesod1.htm. 18 This view represents the adherence of Israeli constitutional law to the so-called ‘legislative’ model of anti-terrorism law. See n 11 above. This emphasis is warranted against the background of the use of executive power in the area of national security in the US, although the breadth of this power is subject to controversy as illuminated by cases such as Hamdi v Rumsfeld 542 US 507 (2004) and Hamdan v Rumsfeld 548 US 557 (2006). 19 S 32 of Basic Law: The Government states: ‘The Government is authorized to perform in the name of the State and subject to any law, all actions which are not legally incumbent on another authority’. 20 See HCJ 11163/03 Supreme Monitoring Committee for Arab Affairs in Israel v Prime Minister 61(1) PD 1 [2006] (in Hebrew). 21 For additional comparison between the two, see B Bracha, ‘Checks and Balances in Protracted State of Emergency – The Case of Israel’ (2003) 33 Israel Year Book of Human Rights 123. 13 14
The National Security Constitution 433
and section 39(c) states that
[e]mergency regulations may alter any law temporarily suspend its effect or introduce conditions, and may also impose or increase taxes or other compulsory payments unless there be another provision by law.
Section 39(f) adds the factor of the time limitation on emergency regulations, stating that [t]he force of emergency regulations shall expire three months after the day of their enactment unless their force is extended by law, or they are revoked by the Knesset by law, or pursuant to a decision of a majority of the members of Knesset.
The current version of Basic Law: The Government includes several additional limitations on the powers of government during a state of emergency.22 First, the declaration of a state of emergency may not exceed one year, although it may be renewed. Section 38(b) states in this regard: The declaration will remain in force for the period prescribed therein, but may not exceed one year; the Knesset may make a renewed declaration of a state of emergency as stated.
The practical implication of this limitation is that the Knesset is obligated to revisit its decision in this matter at least once a year. Second, the Basic Law subjects the power to promulgate emergency regulations to several substantive limitations regarding their contents. Section 39(d) states that [e]mergency regulations may not prevent recourse to legal action, or prescribe retroactive punishment or allow infringement upon human dignity.
Section 39(e) adds another limitation by stating that [e]mergency regulations shall not be enacted, nor shall arrangements, measures and powers be implemented in their wake, except to the extent warranted by the state of emergency.23
Since 1992, Basic Law: The Government provides for another important matter, not previously regulated – the declaration of war (and not only of emergency). In the current version of the Basic Law, it is governed by section 40. Section 40(a) states in this regard that, in principle, [t]he state may only begin a war pursuant to a Government decision.
At the same time, section 40(b) clarifies that [n]othing in the provisions of this section will prevent the adoption of military actions necessary for the defence of the state and public security.
In addition, section 40(c) states that the Knesset (more specifically, its Foreign Affairs and Security Committee) must be notified of decisions to begin war or of military 22 These additional limitations had been introduced for the first time in Basic Law: The Government, 57521992 (n 14). 23 As explained below, this limitation was already introduced by judicial precedents even prior to its express stipulation in the Basic Law. See A Shinar, ‘Constitutions in Crisis: A Comparative Approach to Judicial Reasoning and Separation of Powers’ (2008) 20 Florida International Law Journal 115; C Friedberg and RY Hazan, ‘Israel’s Prolonged War against Terror: From Executive Domination to Executive–Legislative Dialogue’ (2009) 15 Journal Legis Studies 257.
434 Daphne Barak-Erez actions. With regard to war, the Prime Minister should also give a notice to the Knesset plenum. The constitutional significance and force of these provisions of Basic Law: The Government is connected to the fact that they are not subject to change through emergency regulations. According to section 41 of the Basic Law, [n]otwithstanding the provisions of any law, emergency regulations cannot change this Basic Law, temporarily suspend it, or make it subject to conditions.
It is difficult to circumvent this limitation because according to section 44(a) the Basic Law itself may be repealed only by a special majority.24 Thus, in contrast to regular legislation that may be changed by a simple majority vote in the Knesset, the required majority in this context is at least 61 Knesset Members (out of the 120 in the full plenary). This provision secures for Basic Law: The Government certain durability. At the same time, in terms of political reality, the majority needed does not constitute a high barrier to cross, since every stable coalition government needs at least this majority in the Knesset in order to be able to govern effectively.25 Another important limitation on the use of emergency regulations derives from Basic Law: Human Dignity and Liberty, which includes the main constitutional provisions on the protection of human rights in Israel. According to section 12 of this Basic Law, it cannot be suspended or amended by emergency regulations. In addition, emergency regulations may limit the application of the rights guaranteed by the Basic Law, only subject to the condition that the denial or restriction shall be for a proper purpose and for a period and extent no greater than is required.
Whether such emergency regulations meet this standard is a question subject to judicial review.
IV. THE EMERGENCY CONSTITUTION IN PRACTICE
The Israeli regime of emergency powers is formally based on the assumption that emergency law is the exception to the norm and that the government is subject to close parliamentary scrutiny in this area (since the Knesset controls the power to declare and prolong an emergency situation). Practice, however, presents a different de facto balance between the executive and the legislature. In fact, the emergency situation originally declared in Israel in May 1948, pursuant to its establishment, has never been abolished for several reasons. First and foremost, threats on Israel’s security never subsided after its independence, culminating in both wars and terrorist attacks. Hence, one might understand why the Knesset found it diffi24 S 44(a) of Basic Law: The Government (n 15) states: ‘This Basic Law can only be changed by a majority of the Knesset members; the majority under this sub-section will be required for decisions of the Knesset plenum in the first, second and third readings; for purposes of this sub-section, “change” is either explicit or by implication’. 25 The requirement of special majority to amend Basic Law: The Government is thus very different from the strong preservation of the status quo originating from the harsh procedure of constitutional amendment in the US. See US Constitution, Art V.
The National Security Constitution 435
cult to take the initiative and stop renewing this declaration. Second, from a political perspective, the coalition-based system of government in Israel does not allow for a very independent parliamentary supervision of the executive. Knesset Members who belong to the parties that comprise the government tend to support it, especially in core matters such as national security. This way, Israeli emergency law demonstrates once again the danger of analyzing legal regimes only by reference to the ‘law in the books’. In these books, the emergency regime is the exception. In the realm of ‘law in action’, it is the standard.26 In 1999, the Association of Civil Rights in Israel (ACRI) petitioned the High Court of Justice arguing that the declaration of an emergency situation had become unreasonable with the passage of time. The prospect of abolishing the declaration necessitated a serious process of reviewing current legislation in order to assess whether some of the laws which include provisions conditioning their applicability in the existence of a declaration of emergency situation should be amended (by applying them during periods with no declaration of an emergency situation in force). Accordingly, the Ministry of Justice initiated such a process and the court allowed it to do so, holding the petition open until this process is finalised and the court is updated regarding its results.27 Eventually, however, with no obvious sign that the threats on Israel had declined, the court decided to dismiss the petition, expressing its wish that the Ministry of Justice continue to promote the effort it had initiated in reviewing statutory schemes which currently apply only during emergency situations.28 V. THE EMERGENCY CONSTITUTION FROM A COMPARATIVE PERSPECTIVE
The regulation of national security in Israeli constitutional law should be assessed also by reference to the theoretical and comparative study of exceptions made for emergencies. The idea that emergency circumstances merit the suspension of the ordinary con stitutional regime is not new. Rather, it is derived from the institution of dictatorship in the Roman Empire. In Rome, a dictator was appointed as a temporary leader with special powers for a limited period of time, in order to confront threats to the empire. As Clinton Rossiter writes: The splendid political genius of the Roman people grasped and solved the difficult problem of emergency powers in a manner quite unparalleled in all history, indeed so uniquely and boldly that a study of modern crisis government could find no more propitious a starting point than a brief survey of the celebrated Roman dictatorship.29
At the same time, history teaches us that the power to declare an emergency situation may be abused. The most notable example of such gross misuse has been the rise of the Nazi regime, which took advantage of and circumvented the Weimar Constitution, by
For this distinction, see R Pound, ‘Law in Books and Law in Action’ (1910) 44 American Law Review 12. See HCJ 3091/99 Association for Civil Rights in Israel v Knesset (1 August 2006) (in Hebrew) (hereinafter: Declaration of Emergency case). The Supreme Court decided to postpone the hearing of this petition without dismissing it. 28 See HCJ 3091/99 Association for Civil Rights in Israel v Knesset (8 May 2012) (in Hebrew). 29 C Rossiter, Constitutional Dictatorship: Crisis Government in the Modern Democracies (Princeton, Princeton University Press, 1948) 15. 26 27
436 Daphne Barak-Erez suspending several of its protections in the name of national security.30 Concerns are inspired not only by historical experience; they are based also on the theoretical understanding that exceptions to the norm tend to gradually eliminate it.31 In principle, the challenge posed by extraordinary threats to national security (or other types of crisis) to the constitutional order raises several questions.32 The first question is whether to insist on maintaining the regular legal order (thus giving preference to human rights and democracy under all circumstances) or to allow for accommodations that curtail the regular level of human rights protections or democratic processes. Israeli law, like many other systems, has chosen the latter option, as explained below.33 The second question concerns the choice between different versions of accommodations – introducing express constitutional provisions which allow for the partial suspension of some constitutional arrangements; adopting lenient constitutional interpretations that allow for broadening executive power without express regulation of emergency regimes;34 or assuming that government will choose to break the rules when this becomes absolutely necessary.35 Here, the Israeli system opted for the first alternative. In general, the argument for the express regulation of emergencies in the constitution derives from the recognition that it avoids the following contradicting evils – the inability to effectively protect the state from its enemies (if only the regular constitutional regime applies) and the destruction of the rule of law as a guiding principle (if emergency extra-legal measures are taken). This preference also finds grounding in the approach of international law. Several human rights conventions allow for ‘derogation’ from the regular level of protection of human rights in states of emergency.36 The main shortcoming of this model concerns the possibility of overusing it. This concern is doomed to be in the background; the answer to it should lie in the domain of political culture (that has to push against this possibility to the maximum extent possible). 30 Article 48 of the Weimar Constitution (the Constitution of the German Reich, 1919 (Die Verfassung des Deutschen Reichs)) stated: ‘In case public safety is seriously threatened or disturbed, the Reich President may take the measures necessary to reestablish law and order, if necessary using armed force. In the pursuit of this aim he may suspend the civil rights described in articles 114, 115, 117, 118, 123, 124 and 154, partially or entirely’. English translation available at: www.zum.de/psm/weimar/weimar_vve.php. 31 The concern regarding the effect of the exception on the norm has been inspired by the writings of Carl Schmitt. See C Schmitt, Political Theology (Chicago, University of Chicago Press, 1985). 32 The description of these choices in the following paragraph is similar to the one offered by EA Posner and A Vermeule, ‘Accommodating Emergencies’ (2003) 56 Stanford Law Review 605, 606–07. 33 In theory, each of these alternatives has different merits and vices. Express provisions regulating emergency situations create stability since they clarify the ‘rules of the game’. At the same time, they are susceptible to overuse, due to the simplicity and prima facie legitimacy associated with the act of suspension of the regular constitutional order. 34 This alternative, based on the interpretation of vague constitutional provisions, necessitates the government to act under a cloud of constitutional uncertainty. It may thus deter the government from overusing special powers, and at the same time may make it hard to act effectively, even when executive action is needed. 35 The last alternative mentioned is, in fact, an extra-legal one. It wishes to narrow the recourse to emergency measures to the minimum degree possible. However, it may in fact eventually lead to scepticism regarding the rule of law altogether. This proposal has been promoted by O Gross, ‘Chaos and Rules: Should Responses to Violent Crises Always be Constitutional?’ (2003) 112 Yale Law Journal 1011. 36 See, eg Art 4 of the International Covenant on Civil and Political Rights (adopted 16 December 1966, entered into force 23 March 1976) 999 UNTS 171 states the following: ‘(1) In time of public emergency which threatens the life of the nation and the existence of which is officially proclaimed, the States Parties to the present Covenant may take measures derogating from their obligations under the present Covenant to the extent strictly required by the exigencies of the situation, provided that such measures are not inconsistent with their other obligations under international law and do not involve discrimination solely on the ground of race, color, sex, language, religion or social origin’. See E Hafner-Burton, LR Helfer and CJ Farris, ‘Emergency and Escape: Explaining Derogations from Human Rights Treaties’ (2011) 65 International Organization 673.
The National Security Constitution 437
As a matter of practice, several modern constitutions include provisions regulating the possibility to declare emergency regimes during which ordinary constitutional protections do not apply to the same extent. It is worthwhile to compare these examples to the regulation of emergencies in Israel. The brief analysis below focuses on three important case studies – those of South Africa, Germany, and Spain – which had suffered during the twentieth century from dictatorial regimes and abuses of emergency powers. Section 37 of the South African Constitution of 1996,37 which serves as an example for a modern and sophisticated constitutional document, recognises the possibility of declaring a state of emergency38 by an Act of Parliament when the life of the nation is threatened by war, invasion, general insurrection, disorder, natural disaster or other public emergency,
and when ‘the declaration is necessary to restore peace and order’. In such circumstances legislation may derogate from the ordinary level of protection of constitutional rights. However, the constitution puts additional limitations on such legislation, which may do so only when ‘the derogation is strictly required by the emergency’ and consistent with the obligations of the state under international law. In addition, the South African Constitution recognises a table of non-derogable rights. It awards absolute protection to human dignity and life, and authorises only limited derogations from other rights. The German Basic Law39 and the Spanish Constitution40 shed light on the possibility of distinguishing between several types of emergencies (a distinction unknown to Israeli law). The German Basic Law distinguishes between ‘internal emergency’ (regulated by Articles 91 and 87a(4)), ‘state of tension’ (regulated by Articles 12a(5)–(6) and 80a) and ‘state of defence’ (regulated by Articles 115a–115l). The most severe situation is that of ‘state of defence’, limited to circumstances in which ‘federal territory is under attack by armed force or imminently threatened with such an attack’. A determination that such a situation exists ‘shall be made by the Bundestag with the consent of the Bundesrat’ based ‘on application of the Federal Government’ and it requires ‘a two-thirds majority of the votes cast, which shall include at least a majority of the Members of the Bundestag’. During such time legislative powers may be exercised with modifications, including the passing of federal legislation in areas which are normally regulated by the Länder (Art 115c) and the use of a special enhanced legislative process (Art 115d), sometimes even only by the Joint Committee, rather than by the Bundesrat and Bundestag (Art 115e). These extraordinary laws ‘shall suspend the operation of incompatible law so long as they are in effect’. However, the last word is given to the democratic process. According to Article 115l, the Bundestag may, with the consent of the Bundesrat, at any point repeal laws enacted by the Joint Committee (Art 115l(1)) and may also at any time terminate the declaration of the state of emergency (Art 115l(2)).
South African Constitution, 1996, s 37. For the possibilities of regulating emergencies in South Africa in the past, see S Ellmann, ‘A Constitution for all Seasons: Providing against Emergencies in a Post-Apartheid Constitution’ (1989) 21 Columbia Human Rights Law Review 163. 39 Grundgesetz (GG). 40 Constitución Española, BOE n 311, 29 December 1978. 37 38
438 Daphne Barak-Erez The Spanish Constitution also distinguishes between three types of emergencies: alarm, emergency, and siege (Article 116(1)). Siege, the highest level of emergency situation, can only be declared by the absolute majority of the House of Representatives at the exclusive proposal by the Government.
The House of Representatives shall determine the territorial scope, duration, and conditions of the declaration. Article 116(5) states that [t]he House of Representatives may not be dissolved while any of the states contained in the present article are in effect,
and Article 116(6) clarifies that [t]he declaration of the states of alarm, emergency, and siege shall not modify the principle of the responsibility of the Government or its agents as recognised in the Constitution and in the laws.
In addition, Article 55 accepts the possibility of suspending basic rights when a state of emergency or siege is declared. The US Constitution, which dates from the eighteenth century, includes a much less sophisticated regulation of emergency. In fact, formally speaking, it does not recognise special rules for times of emergency, and only authorises the possibility of suspending the Writ of Habeas Corpus. According to Article I, section 9, clause 2: The privilege of the writ of habeas corpus shall not be suspended, unless when in cases of rebellion or invasion the public safety may require it.41
This provision falls short of constituting a full scheme of emergency powers, but it forms a pragmatic alternative to the need to do so, by barring the access to court of people whose liberty is being infringed by the executive.42
41 US Constitution, Art I, §9, cl 2. In practice, the use of suspension power in the US has been a source of public debate. The most well-known (and controversial) example of such suspension comes from the decision of President Lincoln during the Civil War. Richard Posner wrote in this regard: ‘One response to Lincoln’s actions might be to say that if he was acting justifiably, we should amend the Constitution to authorize presid ents to suspend habeas corpus in emergencies . . . The alternative, which has been chosen by default, is to say that we are not going to give the president that legal authority but we are going to expect him to suspend habeas corpus if doing so is necessary (as Lincoln believed) to save the nation’. RA Posner, Not a Suicide Pact – The Constitution in a Time of National Emergency (Oxford, Oxford University Press, 2006) 153–54. This analysis adheres to the model of using extra-legal measures. See n 35 above. 42 For an interpretation of the Suspension Clause as a source of emergency power in the substantive sense of the word, see AL Tyler, ‘Suspension as an Emergency Power’ (2009) 118 Yale Law Journal 600. Against this background and following the events of 9/11, Bruce Ackerman has suggested going beyond the traditional regulation of the matter and introducing in the US as well the possibility of declaring an emergency regime, during which the President (as the head of the executive branch) will be authorized to take measures necessary for the defence of the country and its people. See B Ackerman, ‘The Emergency Constitution’ (2004) 113 Yale Law Journal 1029; B Ackerman, Before the Next Attack: Preserving Civil Liberties in an Age of Terrorism (New Haven, Yale University Press, 2006). This proposal builds upon the regulation of emergencies in other constitutions (and accommodates this idea to the US context by the proposal to enact a framework statute in this regard). In the US, however, this proposal was generally speaking received with concern. Critiques offered various reasons – ranging from concerns regarding the ability to protect rights, see, eg LH Tribe and PO Gudridge, ‘The Anti-Emergency Constitution’ (2004) 113 Yale Law Journal 1801, to analytical arguments against the effectiveness of regulating the matter only by a statute, see, eg A Vermeule, ‘Self-Defeating Proposals: Ackerman on Emergency Powers’ (2006) 75 Fordham Law Review 631.
The National Security Constitution 439
Evaluating the Israeli condition against the above comparative background leads to several poignant points. First, the express regulation of emergency regimes is currently considered the lesser evil by various modern constitutional documents. Second, the primary problem with the Israeli regulation of emergency situations does not derive from the law in the books, but rather from the continuity of the declaration of emergency in practice. At the same time, putting formal limitations on the duration of emergencies is not a sensible solution, given the palpable threats the country faces. However, the possibility of amending the relevant provisions in a manner that will allow for distinctions between various degrees or levels of emergencies (as in Germany and Spain) should be considered.43 Such a reform will decrease the scope of exceptional executive powers in times of ‘ordinary’ emergencies. Indeed, as the analysis so far reveals, the constitutional text by itself may not suffice for safeguarding a balanced use of emergency powers. Even a constitutional regime which acknowledges distinctions between degrees of emergency is not a full guarantee against a political choice to use the most extreme measure. However, introducing these nuances into the constitutional system may diminish the prospects for the overuse of emergency powers. VI. JUDICIAL REVIEW AND NATIONAL SECURITY
The regulation of national security in constitutional texts proves to be important only if it is enforced by the courts. As already noted, in Israel, executive decisions in the area of national security have been subject to judicial review since the early days of the Israeli Supreme Court.44 This is also the case today, and even more so, as the Israeli Supreme Court has significantly narrowed preliminary barriers for judicial review, including the justiciability doctrine.45 The relatively significant role exercised by the judiciary in the Israeli context is often juxtaposed with the relatively weak parliamentary supervision of decisions in the area of national security (due to various factors, including the parliamentary coalition system which guarantees an almost automatic support for government decisions, as well as the background perception that Israel is under constant threat).46 It is important to note, however, that although the role of judicial review is important it is also limited. The decisions of the Supreme Court in this area do not challenge the basic structure of the regulation of the emergency regime (nor the declaration of the emergency situation itself),47 but rather criticize specific decisions in the area of national 43 Distinctions between different emergencies are also found in Canada, where the matter is regulated not by the Canadian Charter of Rights and Freedoms, but rather by a statute – the Emergencies Act 1985 (Can). See R Martin, ‘Notes on Emergency Powers in Canada’ (2005) 54 University of New Brunswick Law Journal 161; KL Scheppele, ‘North American Emergencies: The Use of Emergency Powers in Canada and the United States’ (2006) 4 International Journal of Constitutional Law 213, 230–31. 44 See Al-Karabutli (n 8). For the growth of judicial review of decisions in the area of national security, see B Bracha, ‘Judicial Review of Security Powers in Israel: A New Policy of the Courts’ (1991) 28 Stanford Journal of International Law 39. 45 D Barak-Erez, ‘Broadening the Scope of Judicial Review in Israel: Between Activism and Restraint’ (2009) 3 Indian Journal of Constitutional Law 118. 46 See Shinar (n 23); Friedberg and Hazan (n 23). 47 See nn 27–28 above and accompanying text.
440 Daphne Barak-Erez security within this framework.48 In other words, judicial review has proved more effective when it addresses specific decisions to use emergency regulations, and not the basis of the emergency regime as such. The landmark precedents exercising judicial review over decisions in the area of national security have touched on decisions or actions that incongruously prioritised national security over basic rights (and thus were deemed ultra vires, unreasonable or disproportionate). These cases include the ruling to invalidate the military censor’s decision not to disclose the name of a retiring head of Mossad (the Israeli intelligence authority)49 and the ruling which prohibited the use of physical interrogation measures by the General Security Services.50 Traditionally, the Israeli Supreme Court has intervened in the area of national secur ity mostly by reviewing executive decisions, due to the constitutional convention prevalent in the formative years of the Israeli legal system – of legislative sovereignty. Thus, judicial review of legislation in the area of national security was beyond the reach of the court. However, the immunity of legislation from judicial review was removed with the enactment of the Basic Laws on human rights in 199251 – Basic Law: Freedom of Occupation52 and Basic Law: Human Dignity and Liberty. These Basic Laws were interpreted by the Court as also extending judicial review to legislative infringements of human rights.53 One notable example of litigation, which challenged the constitutionality of legislation in the area of national security, has been the controversy over the law which prohibits residents of the Occupied Territories from applying for residency in Israel (as well as from starting a process of naturalization) based on their family relations with Israeli citizens.54 VII. NATIONAL SECURITY BETWEEN CONSTITUTIONAL LAW AND INTERNATIONAL LAW
An analysis focused on the constitutional aspects of national security obscures the fact that, in the Israeli context, many of the important constraints in this area derive not only from constitutional law but also from international law. This has been the case due to 48 In general, the court is willing to intervene when the government tries to wield its power to promulgate emergency regulations although it has other means of action, including the possibility of initiating new legislation. See HCJ 2994/90 Poraz v Government of Israel 44(3) PD 317 [1990] (in Hebrew); HCJ 6971/98 Paritzky v Government of Israel 53(1) PD 763 [1999] (in Hebrew). 49 HCJ 680/88 Schnitzer v Chief Military Censor 42(4) PD 617 [1989] (in Hebrew). 50 HCJ 5100/94 Public Committee against Torture in Israel v State of Israel 53(4) PD 817 [1999] (in Hebrew). 51 See text to nn 9–10. 52 The original version of Basic Law: Freedom of Occupation, 5752-1992, SH No 1387, p 114 was replaced by a newer version – Basic Law: Freedom of Occupation, 5754-1994, SH No 1454, p 90. 53 See HCJ 6821/93 United Mizrahi Bank v Migdal Cooperative Village 49(4)PD 221 [1995] (in Hebrew). However, judicial review of infringements of Basic Law: Human Dignity and Liberty does not extend to laws which preceded this Basic Law, according to the express exception stated in s 10: ‘This Basic Law shall not affect the validity of any law [din] in force prior to the commencement of the Basic Law’. 54 The Court was bitterly divided in this matter, and petitions against the law were rejected twice by close majority–minority decisions. The justices in the minority stated that despite the importance of the legislative goal, the measure taken – enacting a sweeping rule against so-called ‘family unification’ with Palestinian resid ents of the Territories and thus infringing constitutional rights of their Israeli relatives – did not pass the proportionality requirement. See HCJ 7052/03 Adalah v Minister of Interior 61(2) PD 202 [2006] (in Hebrew); HCJ 466/07 Galon v Attorney General (11 January 2012) (in Hebrew). See also D Barak-Erez, ‘Citizenship and Immigration Law in the Vise of Security, Nationality, and Human Rights’ (2008) 6 International Journal of Constitutional Law 184.
The National Security Constitution 441
several reasons: first, the power of international law has generally expanded worldwide, as well as in the Israeli context where the Supreme Court gradually takes more guidance from the norms of international law;55 second, and more concretely, many of the challenges in the area of national security in Israel have concerned the Occupied Territories – beyond Israel proper. Accordingly, the international law of occupation has proved extremely relevant.56 In fact, some of the Israeli Supreme Court’s most interesting and important decisions were mainly inspired by international law more than by Israeli domestic constitutional law. Examples include the decisions on the security barrier57 and the decision on targeted killings.58 The impact of international law has consistently intensified due to the unprecedented willingness of the Israeli Supreme Court to engage with petitions concerning military actions that are taking place in real time, when they have an effect on human rights of civilians.59 The orders given by the Court may have been restrained, but in some cases even the willingness to hear a petition has a restraining effect on military actions.60
VIII. EMERGENCY LEGISLATION WITHOUT EMERGENCY – TOWARD A NEW PARADIGM?
The backbone of national security law, as described so far, has been the power to declare an emergency situation. Therefore, the focus was on the scope of the power to declare an emergency situation and possible reforms in the way it is regulated. In contrast to this traditional analysis, it is suggested here to shed light on the constitutional significance of a relatively new legislative policy of enacting new laws which regulate top priority national security matters without conditioning their force in the existence of a formal declaration of an emergency situation. More specifically, it is possible to point at two categories of new laws of this sort – ordinary laws and so-called temporary laws which include ‘sunset’ provisions. A representative example of this new legislative trend is the Incarceration of Unlawful Combatants Law, 5762-2002, which unlike the law on administrative detentions61 is not limited to emergency situations. A major legislative bill – the Draft Bill Struggle Against Terrorism, 201162 – emphasises the willingness to broaden this option. The Bill, designed 55 D Barak-Erez, ‘The International Law of Human Rights and Constitutional Law: A Case Study of an Expanding Dialogue’ (2004) 2 International Journal of Constitutional Law 611. 56 At the same time, there have been some doubts regarding the complete application of Israel’s Basic Laws on human rights in these areas: HCJ 1661/05 Hof Aza Regional Council v Israeli Knesset 59(2) PD 481 [2005] (in Hebrew). 57 HCJ 2056/04 Beit Sourik Village Council v Government of Israel 58(5) PD 807 [2004] (in Hebrew); HCJ 7957/04 Mara’abe v Prime Minister of Israel 60(2) PD 477 [2005] (in Hebrew). See also D Barak-Erez, ‘The Security Barrier: Between International Law, Constitutional Law and Domestic Judicial Review’ (2006) 4 International Journal of Constitutional Law 540. 58 HCJ 769/02 Public Committee against Torture in Israel v Government of Israel 62(1) PD 507 [2006] (in Hebrew). 59 See, eg HCJ 2936/02 Physicians for Human Rights v Commander of the IDF forces in the West Bank 56(3) PD 3 [2002] (in Hebrew); HCJ 2117/02 Physicians for Human Rights v Commander of the IDF forces in the West Bank 56(3) PD 26 [2002] (in Hebrew). 60 For a broader analysis, see Barak-Erez, ‘Broadening the Scope of Judicial Review’ (n 45). 61 See n 7 above. 62 Draft Bill Struggle Against Terrorism, 2011, HH (Government) No 611, p 1408.
442 Daphne Barak-Erez to form the basis for the new anti-terrorism law in Israel, aims at enacting a law in this area for all seasons, notwithstanding the existence of an emergency situation. In fact, the Bill even aims at amending several existing laws which currently condition their applicability in the existence of an emergency situation. Interestingly, in its decision on the continuity of the declaration of emergency, the Supreme Court has pointed to the new Bill as one of the steps taken by the Ministry of Justice in order to get prepared for the non-renewal of the declaration of emergency situation.63 The decision refrains from addressing the possibility that this achievement may become a ‘Pyrrhic victory’ if the practical result is that the declaration of emergency will end but the so-called normal legislation will become emergency-oriented in its substance. A similar development is the legislation of ‘temporary’ laws – laws with sunset provisions which limit their force to a period of a few years with no connection to the existence of an emergency situation. Representative examples in this regard include the law which limits ‘family unification’ with relatives of Israeli citizens who live in the Occupied Territories64 and the special law on investigations of suspects in national security offences, regarding the time period before they can meet with their attorneys and the procedures lengthening their arrests.65 Indeed, legislating special laws in times that are perceived as ‘special’ due to national security threats (but with no formal constitutional declaration of an emergency situation) is not special to Israel. Following the events of 9/11 this has been done by several Western democracies, such as the US, Canada and the UK. Partially at least, these laws were enacted for a limited period of time, in a way that contributed to their public legitimacy.66 The US Patriot Act of 2001 originally included a sunset clause (section 224) which limited some of the law’s provisions to four years. Later on the issue of prolonging these provisions was the subject of heated debates, and eventually most of them were indeed prolonged. In a similar manner, the Canadian Anti-terrorism Act of 2001 limited the force of two controversial powers – regarding preventive arrests and investigative hearings – to five years. Eventually, these powers were not prolonged. In the UK, the Anti-terrorism, Crime and Security Act 2001, limited its provisions on unlimited detentions of non-nationals declared as ‘suspected international terrorists’ to a period not exceeding five years. After this controversial power was declared as incompatible with the European Convention on Human Rights,67 the UK enacted the Prevention of Terrorism Act of 2005, which introduced the power to promulgate ‘control orders’ to national and non-nationals alike. This power was limited to a duration of one year with the possibility of prolonging it (section 3), and in fact, it was indeed prolonged. Formally speaking, ordinary legislation which includes sunset provisions is not part of the constitutional regulation of emergency situations. However, pragmatically, to a Declaration of Emergency case (n 27) para 16. Citizenship and Entry into Israel (Temporary Provision) Law, 5763-2003. 65 Criminal Procedure (Detainee Suspected of Security Offence) (Temporary Provision), 5766-2006. 66 JE Finn, ‘Sunset Clauses and Democratic Deliberation: Assessing the Significant Deliberation: Assessing the Significance of Sunset Provision in Antiterrorism Legislation’ (2010) 48 Columbia Journal of Transnational Law 442. 67 A v Secretary of State for the Home Department [2004] UKHL 56, [2005] 2 AC 68. 63 64
The National Security Constitution 443
large degree, this form of legislation makes the declaration of an emergency situation redundant. First, such laws provide the government with broad powers that are ready for use at times of threat to national security. Second, when these laws are challenged in court and accused of infringing the proper balance between the protection of national security and the protection of human rights, the court is expected to take into consideration the fact that these are ‘only’ temporary laws and therefore to limit the scope of judicial review. When the court does so, the result is that these laws in fact influence the constitutional framework of emergency.68 In the first decision, which reviewed the constitutionality of the limitations on ‘family unification’, the temporary nature of the law played an important role. Cheshin J referred to earlier precedents on temporary laws and stated that the less we declare temporary laws void, the better . . . Security reasons are reasons that change from time to time, and determining that a law is a temporary law means a reduction in the harm caused by it merely to the areas where security reasons so demand.69
Of course, the temporary nature of the law constitutes only one factor to be considered by the court, and it is still possible that the court will decide to invalidate even a law of a temporary nature.70 The shift toward the legislation of laws with sunset provisions in the area of national security poses an additional question: to what extent are these laws of a really temporary nature? Indeed, it seems reasonable to enact special powers in the area of anti-terrorism for limited periods. This way, it is possible to answer temporal challenges without changing the underlying principles of the legal system. This understanding has shaped the tolerant view of the Israeli Supreme Court toward temporary laws of this kind. However, this tolerance has to be reassessed against the background of prolonging laws with sunset provisions time after time. This practice may create a new balance between security needs and human rights under the disguise of temporary legislation, forming a reality of ‘temporary permanence’ or ‘permanent temporariness’. In many ways, this new practice resembles the basic problem with the constant prolonging of the declaration of emergency situation under Basic Law: The Government. It should therefore come as no surprise that the Israeli Supreme Court is gradually developing a growing (although limited) willingness to review also laws defined as ‘temporary’ when they are prolonged more than once.71
68 This judicial restraint in reviewing temporary laws has developed as a general policy and is not limited to the national security context. See, eg HCJ 726/94 Klal Insurance Co Ltd v Minister of Finance 48(5) PD 441, 486 [1994] (in Hebrew); HCJ 24/01 Ressler v Knesset 56(2) PD 699 [2002] (in Hebrew). 69 Adalah v Minister of Interior (n 54) p 450 (para 118). 70 Indeed in the Adalah v Minister of Interior decision itself, the minority justices thought that the invalidation of the law is merited despite its temporary nature. In another case – CHR 8823/07 Anonymous v State of Israel (11 February 2010) (in Hebrew) – the Court invalidated a statutory provision which recognized the possibility of prolonging the arrest of a suspect in terrorist related activities without his or her presence in court. 71 See Ressler v Knesset (n 68) (a petition against the exemption of students of religious institutions from military service was accepted after the temporary law in this matter was prolonged); Galon (n 54) (the second petition against the law which prohibited family unification with Palestinians from the Occupied Territories was dismissed but some of the justices expressed a growing dissatisfaction with the practice of the prolonging a law which was originally legislated as a temporary measure. Special notice should be taken of the opinion of Levi J, who was part of the majority in the earlier decision in this matter, and ultimately joined the minority justices).
444 Daphne Barak-Erez
IX. CONCLUSION
The analysis offered so far shows that although the Israeli case study is special it still resembles problems and issues that are debated also in other systems. Recognizing special powers that may be invoked in times of emergency is a practice accepted by several systems and in this sense the Israeli regime as enacted by Basic Law: The Government is a reasonable compromise, especially when one takes notice of the special security threats Israel has to cope with. The unique nature of the Israeli condition derives not from the ‘law in the books’, but rather from the constant prolonging of the declaration of the emergency situation. At the same time, putting a formal limitation on the length of an emergency situation is not a practical solution, taking into consideration the threats facing Israel. The comparative study sheds light on a possibility that has not yet been pursued by Israel – of differentiating between various levels of security conditions.72 Beyond the formal contours of the constitutional norms, this chapter has noted the growing tendency to enact temporary laws with ‘sunset’ provisions which derive their professed justification from increased security threats presented as temporary. When these laws are prolonged, they seem to informally change the constitutional balance between security threats and human rights. In addition, the analysis offered clarifies the position that although the constitutional regulation of Israel’s national security law is formally grounded in Basic Law: The Government, in fact it is much broader and includes also the de facto permanent nature of the declaration of an emergency situation, the influence of international law and the access to judicial review of decisions in this area.
72 As noted above, this differentiation is recognized by the German Basic Law and by the Spanish Constitution. Similarly, such differentiation is also recognized in Canada by the Emergencies Act, 1985. See Martin (n 43); Scheppele (n 43). It is worth noting that Israeli law accepted this differentiation for limited purposes – in the context of the Civil Defence Law, 5711-1951 (as amended), which regulates the powers to defend residents of civilian areas during a period of battles.
29 The Role of the Legislature in Determining Legitimate Responses to Security Threats: The Case of Israel BARAK MEDINA
I. INTRODUCTION
A
CTIVISM AND MINIMALISM are typically attributed to the scope of judicial involvement in determining the legitimacy of decisions taken by the political branches. But such evaluation may also address the role the legislature plays in delineating the measures the Executive Branch may (or should) take. From a normative, deontological perspective, at issue is what is the scope of the principle known as the rule of law, which requires governments to act on the basis of general, predefined norms, determined by an assembly of representatives. In addition, enforcing such requirement imposes the representatives to deliberate on the relevant aims and means, and may thus have a substantial positive effect on policies taken. The purpose of the current chapter is to contribute to the study of this issue by presenting and critically evaluating the Israeli experience regarding the role of the legislature (the Knesset – Israeli Parliament) in one specific context – responding to security threats, and in particular to terror attacks. At issue is the question of what political branch should decide the types of military and administrative measures that are legit imate and justified in certain circumstances? The descriptive part of this chapter presents a story of evolution in the role of the Israeli Parliament in this respect. Until approximately the second half of the 1990s, the Knesset’s role was mostly marginal. The choice of responses to security threats was considered in the exclusive domain of the Executive Branch. However, the last decade has witnessed a growing involvement of the legislature in delineating what measures can legitimately be employed and in what circumstances. This change reflects only in small part the aim of restraining unjustified use of force. The more dominant aim is precisely the opposite – legitimizing the use of certain measures that were otherwise considered prohibited. The chapter also aims to enquire as to what role the Israeli legislature should play. Given the fact that in practice the Knesset legislates mostly to legitimise employing certain measures rather than to restrain the government, it may seem as if the requirement of ex ante legislative authorization is superfluous. Nevertheless, I suggest that enforcing
446 Barak Medina this requirement is justified. I address three main arguments: first, that the requirement of legislation is beneficial, at least in terms of its deliberative values; second, that legislation is an essential tool for scrutinizing governmental activities; and third, that the concerns of the symbolic adverse effects of authorizing human rights infringements in legislation are unfounded. The chapter proceeds as follows: part II provides an overview of the scope of legislation in Israel in determining the legitimacy of various responses to security threats. Part III considers several explanations of this reality. Part IV then moves on to consider the normative aspects of the discussion, by identifying three main models of regulating the government’s response. Part V compares the models and argues for an enhanced involvement of the legislature, according to the third model. Part VI concludes. II. THE KNESSET’S ROLE IN LEGITIMIZING PRE-EMPTIVE MEASURES
In Israel, as in other liberal democracies, determining how to respond to security threats and what pre-emptive measures to take is traditionally considered to be in the domain of the Executive Branch. As indicated, the Israeli legislature adopted a rather minimalist approach in regulating the government’s actions in the fight against terror. This part describes this practice, and points to its evolution in recent years. A. The Minimalist Approach The prevailing approach in Israel used to be one of legislative deference to the Executive Branch in determining the legitimacy of the use of pre-emptive measures in response to security threats. Whereas punitive measures can be implemented exclusively by the courts, based on explicit legislation, pre-emptive measures are often taken without such authorization. Subject to minor exceptions, the Knesset refrained from determining in legislation the preventive powers of the Executive Branch (and to a great extent, did not apply other supervising powers as well). It neither granted the Executive Branch explicit authorization to take certain measures nor did the Knesset restrain the government by setting limits to its powers. A notable expression of this minimalism is the government’s decision to subject tens of thousands of Arab Israeli citizens to martial law in the years 1948–66.1 This policy came into force during Israel’s War of Independence, which was, in part, a civil war between the Jews and Arabs in Palestine. This decision can be considered legitimate in times of war. However, it remained in force long after the hostilities were over, for close to two decades, and, as indicated, the Knesset did not take any (formal) part in either imposing martial law or revoking it. In 1967, just one year after martial law was revoked, Israel reimplemented it, this time over the hundreds of thousands of Palestinians who lived in the West Bank and the Gaza Strip, areas which were occupied in the Six Days War.2 The Israeli martial law in the 1 Y Bauml, ‘The Military Government on the Israeli Arabs and its Cancellation’ (2002) 43 The New East 133 (in Hebrew). 2 M Shamgar, ‘Legal Concepts and Problems of the Israeli Military Government – The Initial Stage’ in M Shamgar (ed), Military Government in the Territories Administrated by Israel, 1967–80: The Legal Aspect (Jerusalem, Hebrew University, 1982) 13.
Legislative Responses to Security Threats 447
Gaza Strip ended in 2005, when Israel withdrew from the area, whereas in the West Bank this status is still in force, now for more than four decades. Here too, the Knesset refrained from authorizing the government to implement the military order and take specific measures, and it did not impose restrictions on the IDF’s (Israel Defence Forces) powers (although it did approve in legislation the government’s decisions to withdraw from the Gaza Strip in 2005, and to transfer certain sovereign powers to the Palestinian Authority in specific parts of the West Bank, in 1993). In the face of terrorist attacks, the Israeli security forces took harsh military measures against persons living in the Occupied Territories.3 Again, these measures were not explicitly authorised by the Knesset and were not subject to limitations set forth in legislation. This long practice of deference by the Knesset was possible, in terms of formal authorization, based on three main sources of law (in addition to section 40 of the Basic Law: The Government, discussed below). The first is legislation enacted during the British Mandate over Palestine, primarily the Defence Regulations (Emergency), 1945. This legislation consists of extensive governmental powers, which are highly uncommon in democratic regimes. It was implicitly incorporated into Israeli law by section 11 of the Law and Administration Ordinance, 5708-1948.4 Other provisions that empower the government with extensive powers to infringe basic liberties, also originated in the era of the British Mandate, and were similarly incorporated into the Israeli law. The second source that authorises the government to take actions in response to security threats is the norms of customary international humanitarian law (IHL). The international law of belligerent occupation authorises the occupying state to take all the measures in [its] power to restore, and ensure, as far as possible, public order and safety, while respecting, unless absolutely prevented, the laws in force in the country.5
The laws in force in the Gaza Strip and the West Bank prior to the Israeli occupation, as well as the general authority to take all measures required to ensure public safety, provide the government with extensive powers, making legislation by the Knesset unnecessary. The Israeli Supreme Court also ruled that the terror attacks of Palestinian organizations established an ‘international armed conflict’.6 The government is thus empowered to take actions such as targeted killings of suspected terrorists and mass curtailment of freedom of movement, including: imposing curfews and closures on the residents in certain areas, applying a siege on the Gaza Strip, and the construction of the separation barrier. This too enables the Knesset to refrain from taking part in the difficult moral and legal dilemmas about the conditions in which it is justified to use such measures.
3 D Kretzmer, The Occupation of Justice: The Supreme Court of Israel and the Occupied Territories (Albany, SUNY Press, 2002) 115–85. 4 HCJ 7/48 Al-Karabutli v Minister of Defence 2(1) PD 5, 9–10 [1949] (in Hebrew). 5 Hague Convention (IV) Respecting the Laws and Customs of War on Land and its Annex: Regulations Concerning the Laws and Customs of War on Land (adopted 18 October 1907, entered into force 26 January 1910) TS 539, Art 43. 6 HCJ 769/02 Public Committee against Torture in Israel v Government of Israel 62(1) PD 507 [2006] para 18 of Court President Barak’s opinion (in Hebrew). English translation available at: elyon1.court.gov.il/files_ eng/02/690/007/e16/02007690.e16.htm.
448 Barak Medina The third source is ‘emergency powers’. When a declaration on a state of emergency is in force, the government is authorised to employ several specific powers.7 In such times the government is also authorised by Basic Law: The Government, section 39(c) to enact emergency regulations, which may alter any law temporarily, suspend its effect or introduce conditions, and may also impose or increase taxes or other compulsory payments.
The Knesset declared a state of emergency just a few days after the State was founded, on 19 May 1948, and has avoided cancelling this declaration ever since. The Supreme Court refused to review the legitimacy of this practice.8 In recent decades the government has avoided using this power, mainly due to limits imposed by the judiciary regarding the power to enact emergency regulations.9 Thus, most powers of the Executive Branch that are employed in response to security threats are not based on the Knesset’s legislation. Only a handful of powers are regulated in legislation, most notably administrative detentions, set by the Emergency Powers Law (Detentions), 5739-1979, and the authority to declare an organization as a terrorist one, according to the Prevention of Terrorism Ordinance of 1948 (both acts are in force only during the time of a formal declaration of ‘a state of emergency’). B. The Knesset’s Greater Involvement in the Recent Decade In recent years, the Knesset has enacted several laws which are aimed, primarily, at expanding the security agencies’ powers and to removing doubts about the legitimacy of employing these powers absent such an explicit authorization in legislation. This legislation includes two main types of norms. One group consists of legislation that provides powers to interrogate suspected terrorists. The Knesset’s involvement in this area is the result of the tendency to limit the powers of the police, and accordingly to provide greater protection of basic liberties for criminal, non-terror related suspects. In setting such norms the Knesset explicitly exempted the security agencies from the general limitations, and authorised them to take more extensive measures in interrogating suspected terrorists. One example (which dates to the earlier period) is the Wiretap Law, 5739-1979. It prohibits investigation authorities from engaging in wiretapping, unless authorised to do so, for the purpose of preventing serious crimes, by a court order. However, the law exempts security agencies from this requirement and authorises them to wiretap ‘whenever needed for security reasons’, subject only to the approval of the Minister of Defence (section 4). Other examples deal with the rights of detainees. The Criminal Procedure Law (Enforcement Powers – Detentions), 5756-1996, compels investigation authorities to inform a relative of a person detained and a lawyer of his or her choice about the detention. Here too the Knesset empowered the government to infringe this basic right 7 There are thus certain similarities between this declaration and the decision of the US Congress known as the War Powers Resolution, Pub L No 93-148, 87 Stat 555 (1973) (codified as amended at 50 USC, ch 33, s 1541–48 (2006)). This Resolution authorizes the President to make brief interventions unilaterally – giving him 60 days to use military force without legislative approval. 8 HCJ 3091/99 Association for Civil Rights in Israel v Knesset (8 May 2012), Nevo Legal Database (by subscription) paras 11–13 (in Hebrew). 9 HCJ 6971/98 Paritzky v Government of Israel 53(1) PD 763 [1999] (in Hebrew).
Legislative Responses to Security Threats 449
in the case of persons suspected of terrorist activities (‘security offences’ in the jargon used in this and related legislation). The same is true in respect of the right to counsel, under the Criminal Procedure Law (Enforcement Powers – Detentions), 5756-1996 (sections 34 and 35). The Knesset also enacted the Criminal Procedure Law (A Detainee Accused in a Security Offence) (Temporary Provision), 5766-2006, which empowers the government to hold suspected terrorists in detention for 96 hours without a judicial warrant (as compared to only 24 hours in the case of persons who are suspected of nonsecurity related criminal offences); and authorises the government to ask, and the court to approve, that the court hearings on extending the period of detention will be held without the detainee’s presence, whenever stopping the interrogation (in order to bring the detainee to court) is likely to prevent pre-empting a security offence. Finally, the Criminal Procedures Law (Criminal Investigations), 5762-2002, which imposes a duty of written or video documentation of interrogations of the accused, provides a general exemption from this requirement in the case of investigating a person accused of terror related offences (section 17). Interestingly, in most of the above Laws the government is required to provide the Knesset with a periodical report of its use of the above exceptional powers, such as the number of wiretaps without a judicial warrant that were held during a specified period, the number of persons held in detention for more than 24 hours without a judicial warrant, and so on. This requirement strengthens the Knesset’s involvement in supervising the legitimacy of employing such extraordinary measures. A second type of norms enacted in recent years authorises the government to take specific measures in response to security threats. Among these is the Detentions Law, which empowers the Minister of Defence to order the holding of a person in detention for a period of up to six months (but administrative detention orders may be extended repeatedly), whenever there is a reasonable basis to suppose that the security of the State or of the public necessitates [this] person to be held in detention.10
The Israeli Supreme Court ruled that this power can only be used as a forward-looking preventive measure, against a person who poses an individual threat, and may not be used as punishment for past acts.11 In response, the Knesset enacted a new piece of legislation (without revoking the earlier one) – the Internment of Unlawful Combatants Law, 2002. It authorises the government to intern persons who are classified as ‘unlawful combatants’, to bypass the requirement of proving the detainee’s individual dangerousness. The law aims at replacing the requirement of ‘individual dangerousness’, with a ‘collective dangerousness’ approach, according to which a person can be interned whenever the terrorist organization that he is affiliated with poses a threat.12 In addition, internment under the Unlawful Combatants Law is not limited in time. Emergency Powers Law (Detentions), 5739-1979, s 2. See, eg HCJ 5784/03 Salama v IDF Commander of Judea and Samaria 57(6) PD 721 para 6 [2003] (in Hebrew). English translation available at: elyon1.court.gov.il/Files_ENG/03/840/057/A05/03057840.A05.HTM. See also CrimFH 7048/97 John Does v Ministry of Defence 54(1) PD 721 [2000] (in Hebrew), in which the Court invalidated a decision to hold a person in detention merely as a ‘bargaining chip’. English Translation available at: elyon1.court.gov.il/files_eng/97/480/070/a09/97070480.a09.pdf. 12 CrimA 6659/06 A v State of Israel 62(4) PD 329 [2008] (in Hebrew). English Translation available at: elyon1.court.gov.il/Files_ENG/06/590/066/n04/06066590.n04.htm. The Court ruled, however, that this power can be employed only when the detainee’s ‘individual dangerousness’ is sufficiently high. 10 11
450 Barak Medina Other examples of such legislation are: the Prohibition on Financing Terror Law, 2005, and the General Security Service Law, 2002. The former law prohibits financially transacting with entities declared by the government as terror organizations. The latter law authorises the General Security Service (GSS), Israel’s main security agency responsible for the fight against terror, to employ several measures to collect information, including the power to enter private premises, to obtain telecommunication data, and more. At the same time, it also sets limits on the GSS powers, both by setting express limitations,13 and by not granting its staff certain powers, most notably the power to employ coercive investigative methods. Finally, the government has recently published a draft of the proposed Fight Against Terrorism Law, 2010, which aims primarily to codify the existing legislation that was discussed above, but also to add several additional anti-terrorist measures. The above legislation reflects a considerable increase in the Knesset’s involvement in determining the appropriate responses to security threats in comparison to the past. However, the Knesset did not deviate from its traditional minimalist approach regarding ‘military’ activities, such as targeted killings, imposing siege, curtailing freedom of movement, and so forth. In these areas the Knesset did not respond to judicial decisions that substantially limited the government’s powers.14 Thus, the Knesset’s current approach cannot accurately be described as an activist one in absolute terms, but only relatively to its previous approach. In addition, the above review demonstrates that the aim of restraining the Executive Branch is the distinctively less dominant one. The legislation imposes some limits on the governmental powers, by setting the conditions for employing the relevant powers. A potential additional restraint may result from the requirement for the government to provide the Knesset with a periodical report of the use of some of its exceptional powers. However, it seems that the more dominant purpose of the recent tendency of greater involvement of the legislature is precisely in the opposite direction, as substantial parts of the relevant enactments legitimise the use of various measures in response to security threats. III. REASONS FOR THE KNESSET’S PAST MINIMALISM AND CURRENT (PARTIAL) ACTIVISM
The decision whether to enact norms that relate to governmental powers may be based on three main types of considerations – functional, deliberative and symbolic. The central element is the functional aspect. Legislation may serve to either legitimise an otherwise prohibited governmental measure or to prohibit an otherwise legitimate one. This consideration is contingent on two background conditions: the first is the nature of the governing constitutional norms, mainly the scope of judicial review of governmental activities, and the enforcement of the requirement of explicit legislative authorization as a prerequisite to employ certain powers. A second relevant condition is the legislature’s political motivation to restrain the government. A related functional role of legislation is to direct behaviour of soldiers and other officials. A permissive legislation may serve to shield soldiers from the risk of criminal liability, and a prohibitive one may be aimed at deterring them from taking certain measures. The General Security Service Law, 5762-2002, ss 8–12. See n 21 below.
13 14
Legislative Responses to Security Threats 451
The two other considerations are more nuanced. The deliberative aspect refers to the process of legislation, rather than its outcome. This process serves as a forum for public deliberation on the disputed issue. Thus, the decision whether to legislate is also a result of the prevailing political culture regarding the question whether inducing such an open deliberation is desirable or not. The third consideration is the symbolic consequences of legislation. As discussed below, the explicit authorization in legislation to take measures that infringe basic human rights may have adverse symbolic (domestic and international) political consequences. In general, legislatures do not show an interest, independent of that of the government, in restricting the powers of the Executive Branch in the context discussed here. Times of armed conflicts and threats of terror activities are characterised by a high level of uncertainty and ambiguity. The stakes are high and imposing or enforcing restraints by the Parliament on the use of certain types of military force is typically perceived as a politically unrewarding activity. It is not surprising then that legislatures in liberal democracies employ a rather minimalist approach to regulating the fight against terrorism, as discussed above.15 The same is true for the government’s incentive to initiate legislation. Legislation is costly, in terms of both the requirement of holding extensive deliberations and the symbolic effects of ex ante explicit authorization to infringe basic liberties. As long as the scope of judicial review is ‘bearable’ in terms of the measures it prevents the government from taking, these costs are usually prohibitive.16 Consider, along these lines, the Knesset’s practice in the two periods under discussion. The minimalism is mostly the result of the Court’s lack of activism in most parts of the first period. The Court avoided requiring ex ante legislation as a prerequisite for employing military measures, and it also recognised, as discussed above, British Mandate legislation and IHL as sufficient sources of power. The government had no interest in initiating legislation, and the Knesset did not show any preference, independent of that of the government, in restricting or otherwise delineating the powers of the Executive Branch. Rationalizing the recent legislation is more difficult, as here conflicting considerations are involved. As indicated, one purpose of this legislation is to restraint the Executive Branch. This part of the legislation can be explained by an ideological shift in the Knesset towards a wider protection of basic human rights. This shift is a part of what is known as the ‘Constitutional Revolution’ of 1992/95, which refers to the enactment of the Basic Law: Human Dignity and Liberty, and the judicial recognition of this Basic Law as Israel’s Bill of Rights. The greater attachment to the requirement to respect human rights and other constraints may also be explained on pragmatic (functional) considerations, mainly as a consequence of the enhanced (political) status of international law and international tribunals. One may also identify an evolving preference in the Israeli political culture towards more substantial public deliberations, which may have diffused into setting security policies as well.
15 See, eg JL Mashaw, ‘Due Process of Governance: Terror, the Rule of Law, and the Limits of Institutional Design’ (2009) 22 Governance 353; B Ackerman and OA Hathaway, ‘Limited War and the Constitution: Iraq and the Crisis of Presidential Legality’ (2011) 109 Michigan Law Review 447, 450–51. 16 cf S Issacharoff and RH Pildes, ‘Emergency Contexts without Emergency Powers: The United States Constitutional Approach to Rights During Wartime’ (2004) 2 International Journal of Constitutional Law 296.
452 Barak Medina However, as indicated, the more dominant purpose of the recent tendency of greater involvement of the legislature is precisely in the opposite direction, as substantial parts of the relevant enactments are aimed at legitimizing the use of certain measures. The functional element of the legislation, to authorise otherwise impermissible (or at least doubtful) measures, is probably the main purpose of this recent increase in legislation. One reason is the enactment of acts that provide greater protection of basic liberties of criminal, non-terror related suspects, such as a maximum period of holding a person in detention, the right to counsel, and more. These acts required the Knesset to explicitly address the powers of security agencies in dealing with suspected terrorists, since the majority’s position is that the fight against terrorism justifies infringing the criminal procedure rights of suspected terrorists. Thus, the aim is to neither expand the government’s powers nor restrain it, but merely to hold them unchanged, but given the change in the norms that generally apply in criminal proceedings, the exceptional measures taken against suspected terrorists had to be legislated for. A second reason is the moderate increase in the Supreme Court’s enforcement of the constitutional law doctrine which requires an explicit legislated authorization to take measures that infringe basic liberties. According to section 8 of the Basic Law: Human Dignity and Liberty, it is not permissible to infringe human rights unless the infringement is prescribed by law. This norm expands the criminal law principle of legality (nullum crimen sine lege). In its important 1998 Rubinstein v Minister of Defence decision, the Court expanded this requirement and held that employing powers which involve substantive moral dilemmas or raise public and political dispute must be authorised in legislation.17 The Court held that the Knesset should set the ‘primary’ arrangements, leaving the Executive Branch to deal only with ‘secondary’ norms. The Court held that the government’s powers are limited according to this doctrine.18 The Rubinstein v Minister of Defence case dealt with the policy of exempting Yeshiva (institution for religious studies) students from the draft, and in 1999 the Court implemented this doctrine to hold that it is impermissible to use coercive interrogation methods, not even in the case of a ‘ticking bomb’, at least as long as no explicit authorization to do so is promulgated in legislation.19 However, the enforcement of this doctrine is rather limited, as the international laws of armed conflict are considered a sufficient basis for authorization. The third motivation of the Knesset’s recent greater involvement is the Court’s growing activism in reviewing the legitimacy of various military activities.20 In several cases the Court declared invalid even anti-terrorist measures that were explicitly authorised in legislation.21 But these are exceptional, as in most instances the military activities that HCJ 3267/97 Rubinstein v Minister of Defence 52(5) PD 481, 502–24 [1998] (in Hebrew). ibid 524–30. cf the Canadian Supreme Court decision in Slaight Communications Inc v Davidson [1989] 1 SCR 1038, 1077 (Can) (‘it is impossible to interpret legislation conferring discretion as conferring a power to infringe the Charter, unless, of course, that power is expressly conferred or necessarily implied’). 19 HCJ 5100/94 Public Committee against Torture in Israel v State of Israel 53(4) PD 817, 832–36 [1999] (in Hebrew). See also HCJ 1437/02 Association for Civil Rights in Israel v Minister of Public Security 58(2) PD 746 [2004] (in Hebrew) (invalidating an order which denies detainees the right to counsel, for the lack of sufficient authorization in legislation for such an order). 20 See, eg Y Dotan, ‘Legalising the Unlegaliseable: Terrorism, Secret Services and Judicial Review in Israel 1970–2001’ in M Hertogh and S Halliday (eds), Judicial Review and Bureaucratic Impact: International and Interdisciplinary Perspectives (Cambridge, Cambridge University Press, 2004) 190. 21 CHR 8823/07 John Doe (Anon) v State of Israel (11 February 2010), Nevo Legal Database (by subscription) (in Hebrew) (invalidating s 5 of the Criminal Procedure Law (A Detainee Accused in a Security Offence) (Temporary Provision), 5766-2006, which authorized the government to ask, and the Court to approve holding 17 18
Legislative Responses to Security Threats 453
were found impermissible were not based on explicit legislation, and the Court was not required to refer to constitutional norms to declare them invalid. These include, for instance, the decision to ban the use of force (or the use of ‘humiliating’ techniques) in interrogations of suspected terrorists;22 the prohibition to target persons who do not take direct part in military activities;23 the decision that the army may not use enemy civilians as ‘human shields’;24 and the prohibition to order that certain roads will be closed to traffic of vehicles of all Palestinian residents of the Occupied Territories, following terrorist attacks against passengers on these roads.25 The Court preferred not to base its decisions on formal rules, such as the lack of explicit authorization in legislation, but rather referred mainly to broad standards such as the ‘proportionality’ doctrine. The result is that these decisions are often read as reflecting not only what is legal but also what is just and morally permissible, making the political cost of an override often a prohibitive one. In addition, the fact that the Court decided according to its interpretation of IHL means that legislation legitimizing the prohibited measures is perceived as overriding not only the Court’s ruling but also the dictates of IHL. For these reasons, in the majority of the cases, most notable of which is the prohibition against the use of coercive measures in interrogation, the government preferred ‘to bite the bullet’, and chose not to ask for legislative approval of such activities. IV. REGULATING RESPONSES TO SECURITY THREATS: MODELS OF LEGISLATIVE INVOLVEMENT
The extensive literature of recent years on the broad issue of the ‘Emergency Constitution’ discusses the applicability of the constitution in times of emergency. In particular, it addresses the question whether the prohibition against unjustifiable curtailment of basic liberties applies in times of emergency just as it does in times of normalcy. Some scholars argue that (at least substantial parts of) human rights law should be suspended or qualified in times of crises, and thus legitimise infringements that would have been proscribed otherwise.26 Others object to this position, arguing that one may not ‘skip the Constitution’27 even in times of emergency, as the solutions must be found within the constitutional framework. The debate is whether anti-terrorist measures should be taken ‘outside’ the constitution, and in an important sense ‘outside’ law in general (as in court hearings without the detainee’s presence); HCJ 8276/05 Adalah v Minister of Defence 62(1) PD 1 [2006] (in Hebrew) (invalidating a provision denying persons harmed in ‘conflict zones’ in the Occupied Territories from the right to receive compensation). English translation available at: elyon1.court.gov.il/files_eng/05/ 760/082/a13/05082760.a13.pdf. 22 Public Committee against Torture (n 19). 23 Public Committee against Torture (n 6). 24 HCJ 3799/02 Adalah v GOC Central Command, IDF 60(3) PD 67, 80 [2005] (in Hebrew). English translation available at: elyon1.court.gov.il/files_eng/02/990/037/A32/02037990.a32.htm. 25 HCJ 2150/07 Abu Safiyea v Minister of Defence (29 December 2009), Nevo Legal Database (by subscription) (in Hebrew). English translation available at: elyon1.court.gov.il/files_eng/07/500/021/m19/07021500.m19. htm. 26 See, eg RA Posner, Not a Suicide Pact: The Constitution in a Time of National Emergency (New York, Oxford University Press, 2006) 152–58; EA Posner and A Vermeule, Terror in the Balance: Security, Liberty, and the Courts (New York, Oxford University Press, 2007) 15. 27 D Cole, ‘How to Skip the Constitution’, New York Review of Books, 16 November 2006, www.nybooks. com/articles/archives/2006/nov/16/how-to-skip-the-constitution.
454 Barak Medina the slogan ‘necessity knows no law’),28 or within it. The discussion in this chapter is narrower. My underlying premise is that human rights laws bind at all times. The dilemma refers to the applicability of only one constitutional norm – the requirement of legislative authorization. It is assumed that the governmental decisions must stand the scrutiny of legality, and the question is basically whether the legislature, and not only the judiciary, should be required to take part in deciding on the constitutionality of the response to security threats. What role should the Israeli legislature play in determining the ways in which the state may respond to security threats? I start the normative discussion by delineating three main models of regulating the government’s response. According to one model, the Executive Branch holds ‘inherent’ powers (including those based on international laws of war), to employ all measures it deems necessary in response to security threats. The government may well be limited in employing its discretion to take a certain measure – mainly, it may do so only for ‘appropriate purpose’, and may inflict harm subject to the ‘proportionality’ requirements – but the measures it may use are not limited to those explicitly authorised by the legislature. A second model provides the Executive Branch with such ‘inherent’ powers only as long as the legislature explicitly declares that a security threat exists (eg ‘a state of emergency’). A form of this model is employed in the International Covenant on Civil and Political Rights,29 which provides the possibility of derogating from some of the rights enshrined in the Covenant during declared states of emergency. According to this model, while it is not necessary for the legislature to provide the government with the powers that may be used in response to the threat, a general authorization to hold such un- enumerated powers is required, as well as a formal declaration about a state of emergency. Arguably, this latter requirement makes explicit the exceptional nature of the measures the government may employ, and enforces the legislature to deliberate on the justification of authorizing the government to use special measures. Finally, according to a third model, the government’s powers are limited to those explicitly authorised in legislation. This model is based, primarily, on the concern that in times of crises the legislature cannot be expected, for political reasons, to appropriately scrutinise the activities of the Executive Branch. The only effective way to involve the legislature in the deliberation is by requiring a legislated authorization to take such actions.30 V. THE CASE FOR GREATER INVOLVEMENT OF THE LEGISLATURE
The comparison between the different models should be based on general, universal considerations, but it clearly has a substantial local and contextual dimension. The fol28 W Rehnquist, All the Laws But One: Civil Liberties in Wartime (New York, Alfred A Knopf, 1998); A Harel and A Sharon, ‘Necessity Knows No Law: On Extreme Cases and Uncodifiable Necessities’ (2011) 61 University of Toronto Law Journal 845. 29 International Covenant on Economic, Social and Cultural Rights (adopted 16 December 1966, entered into force 3 January 1976) 993 UNTS 3. 30 For a related typology, see D Barak-Erez, ‘Terrorism Law between the Executive and Legislative Models’ (2009) 57 American Journal of Comparative Law 877; J Ferejohn and P Pasquino, ‘The Law of the Exception: A Typology of Emergency Powers’ (2004) 2 International Journal of Constitutional Law 210, 216–18. The second model presented here is close to what Ferejohn and Pasquino refer to as ‘the legislative model’ while the third model is similar to what Barak-Erez calls ‘the legislative model’.
Legislative Responses to Security Threats 455
lowing discussion refers mainly to the Israeli case.31 I start with a short doctrinal analysis of Israeli constitutional law on this issue, and then move to policy analysis, in an attempt to provide support for the third model. A. Doctrinal Analysis The requirement that the government’s activities in response to security threats, both within Israel and in the Occupied Territories, will be explicitly authorised in legislation is based on section 8 of the Basic Law: Human Dignity and Liberty. It can also be based on the Rubinstein doctrine, which requires that all major policy decisions will be determined by the Knesset, in legislation.32 An important doctrinal question is raised by section 40(b) of the Basic Law: The Government, which provides that [n]othing in the provisions of this section will prevent the adoption of military actions necessary for the defence of the state and public security
and section 18 of the Law and Administration Ordinance, 5708-1948, which authorises the government to establish armed forces, which shall have authority to do all lawful and necessary acts for the defence of the State.
To my view, these provisions do not resolve the issue under consideration. Section 40(b) mainly addresses the armed forces power to take the measures necessary immediately, even prior to a formal decision of the government (according to section 40(a)) ‘to begin a war’.33 Section 40(b) provides that ‘Nothing in . . . this section will prevent’ taking actions, without explicitly providing the armed forces with an unlimited power to take all actions necessary to the defence of the state. Similarly, the ‘lawful’ requirement in section 18 can be interpreted as incorporating the requirement of specific authorizing legislation. The powers mentioned in these provisions refer only to the strict sense of ‘military actions’. This term should be narrowly interpreted, to include only measures directed against ‘combatants’ (including civilians ‘for such time as they take a direct part in hostilities’),34 immediately necessary for the protection of lives. Thus, while targeted 31 For an analysis which accentuates the contextual dimension see, eg Barak-Erez ibid 886–91. For localized comparisons between the models see, eg MJ Glennon, ‘The United States: Democracy, Hegemony, and Accountability’ in C Ku and HK Jacobson (eds), Democratic Accountability and the Use of Force in International Law (Cambridge, Cambridge University Press, 2003) 323; G Nolte, ‘Germany: Ensuring Political Legitimacy for the Use of Military Forces By Requiring Constitutional Accountability’ ibid 231; Y Boyer et al, ‘France: Security Council Legitimacy and Executive Primacy’ ibid 280; ND White, ‘The United Kingdom: Increasing Commitment Requires Greater Parliamentary Involvement’ ibid 300; B Medina, ‘Legislating Responses to Security Threats: The Requirement of Legislative Authority to Take Anti-Terrorist Measures’ in F Liao (ed), Constitutional Interpretation (forthcoming). 32 See n 17 above and accompanying text. For a discussion see, eg B Medina, ‘The No-Delegation Doctrine – A Reply to Yoav Dotan and Gideon Sapir’ (2012) 42 Mishpatim 449 (in Hebrew). 33 See M Kremnitzer and A Bendor, The Basic Law: The Army (Jerusalem, Hebrew University Press, 2000) 45 (in Hebrew). 34 Art 51(3) of the First Additional Protocol to the Geneva Convention (International Committee of the Red Cross (ICRC), Protocol Additional to the Geneva Conventions of 12 August 1949, and relating to the Protection of Victims of International Armed Conflicts (Protocol I), 8 June 1977, 1125 UNTS 3). See Barak-Erez (n 30) 891 (suggesting that actions of ‘preventive killings of terrorists . . . are considered part of the conduct of a military confrontation and, therefore, are not based on legislation’).
456 Barak Medina killings of suspected terrorists may be considered as a military action, if conducted to thwart an imminent threat posed by a person who takes a direct part in hostilities, it is implausible to interpret the above provisions as authorizing the armed forces to take measures which are distinctly outside the realm of what is typically conceived of as ‘military actions’. Activities such as demolitions of houses in which suicide bombers lived, mass curtailment of freedom of movement, including imposing curfews and closures on the residents in certain areas, the construction of the separation barrier, administrative detentions, methods of interrogations, employing ‘profiling’ practices for tracking potential terrorists, and similar activities, should be classified as infringements of basic liberties. These are subject to human rights law, including the requirement of explicit legislative authorization as a prerequisite for their legitimacy, and the other requirements set forth in section 8 of the Basic Law: Human Dignity and Liberty. B. Policy Considerations in Support of the Third Model Supporters of the first model, according to which the Executive Branch holds ‘inherent’ powers to employ all measures it deems necessary in response to security threats, base their opposition to the third (and to some extent to the second as well) primarily on the view that the benefits of insisting on legislation are marginal. As demonstrated by the Israeli experience, the legislature may be expected to mostly legitimise employing certain measures rather than to restrain the government, making the requirement of ex ante legislative authorization superfluous. It is also suggested that requiring legislation is dangerous: arguably, it is impossible to predict what types of measures will be required in response to possible security threats, and enforcing the requirement of authorised legislation might thus prevent the government from effectively addressing the threats. In addition, as mentioned above, the explicit authorization in legislation to hold certain powers may have adverse symbolic consequences. I find these arguments unpersuasive. I suggest that insisting on legislation is beneficial, at least in terms of the deliberative value of the process. Moreover, the main concern in the current context should not be from excessive restrictions of the executive but rather from the lack of efficient review, and legislation is an essential tool for scrutinizing governmental activities. It is in this respect that the third model is superior to the second one. Insisting on legislative authorization is also justified on the basis of the importance of ‘rulifying’ the fight on terror. In addition, I argue that concerns of the symbolic adverse effects of authorizing human rights infringements in legislation are unfounded. Finally, I briefly address the concern that enforcing the requirement of authorised legislation might prevent the government from effectively addressing the threats. Deliberation. The democratic principle of citizens’ equal participation lends support to legislative ordering.35 It also provides the executive with an explicit expression of popular support, which is often needed in order to take extreme measures.36 But the most important benefit of requiring legislation is probably the deliberation which the process of legislation induces.37 The process of legislation requires decisionmakers to present See, eg J Waldron, Law and Disagreement (New York, Oxford University Press, 1999) 232–34, 244–49. Ferejohn and Pasquino (n 30) 220. 37 F Michelman, ‘Relative Constraint and Public Reason: What is “The Work We Expect of Law”?’ (2002) 67 Brook Law Review 963, 971–84; Barak-Erez (n 30), 894–95. 35 36
Legislative Responses to Security Threats 457
their empirical evaluations of the relevant risks and the expected efficacy of the measures under consideration.38 It also requires explicitly stated moral positions, including, for instance, what weight should be given to the rights of ‘enemy civilians’, how to address uncertainties, and so forth.39 Essentially, the deliberative process may well shape pre ferences, rather than merely restating attitudes.40 Parliamentary deliberation over the constitutionality of the use of force serves to rebut moral disagreements by conducting a debate which is based on ‘public reason’.41 Under the Rawlsian constraint of public reason ‘participants in social decisions must stand ready to explain the consonance of their positions with some conception of a complete, legitimating constitutional agreement’.42 An appeal to public reason requires the decisionmakers to present valid justifications for inflicting (unintended) harm on enemy civilians, to justify a position which assigns different values to people’s lives according to their nationality, and so forth. The requirement to hold open deliberations may also contribute to mitigate the concern of the institutional bias of law enforcement and security agencies, whose primary goal is to fight crime and terror.43 One may not rule out the possibility that the political discussion will result in providing the government with more powers, as providing greater security is often more popular than protecting basic liberties of suspected terrorists. However, the doctrine under consideration is not based primarily on a consequentialist argument about the probable outcomes under each of the alternative models. Its central basis is the recognition of the intrinsic value of holding parliamentary, and thus also directly or at least indirectly, public deliberations. Scrutinizing governmental activities. In times of crises there is no effective way to appropriately scrutinise activities of the Executive Branch. The lack of adequate information and the high risk which is involved in such decisions make the efficacy of the Parliament’s supervisory powers very limited.44 Parliaments such at the Knesset are empowered to direct the government’s activities only through legislation. The second model presented above is not suited to address the concern of insufficient check on the government. It is doubtful whether the Parliament is suited to hold a thorough and independent review of the government call to declare a state of emergency. The Israeli experience demonstrates such a failure. It seems that the theory of Carl Schmitt, that it is impossible to develop effective constitutional constraints on the use of ibid 891. C Martin, ‘Taking War Seriously: A Model for Constitutional Constraints on the Use of Force in Compliance with International Law’ (2011) 76 Brook Law Review 611; O Greene, ‘Democratic Governance and the Internationalisation of Security Policy: The Relevance of Parliaments’ in H Born and H Hänggi (eds), The Double Democratic Deficit: Parliamentary Accountability and the Use of Force Under International Auspices (Aldershot, Ashgate, 2004) 28. 40 PF Diehl and T Ginsburg, ‘Irrational War and Constitutional Design: A Reply to Professors Nzilebe and Yoo’ (2006) 27 Michigan Journal of International Law 1239, 1249–50; A Gutmann and D Thompson, Why Deliberative Democracy? (Princeton, Princeton University Press, 2004) 1–20. 41 CT Tsai, ‘Presidential War Power in the Deliberative Moment – An Empirical Study of Congressional Constitutional Deliberation and Balance of War Power’ (SSRN, 16 July 2010): ssrn.com/abstract=1641226 (demonstrating that a higher level of congressional deliberation over a use of force influences Congress to impose a higher level of control over presidential war power). 42 J Rawls, Political Liberalism (New York, Columbia University Press, 1993) 226. See also, eg GK Hadfield and S Macedo, ‘Rational Reasonableness: Toward a Positive Theory of Public Reason’ (SSRN, 1 March 2011): ssrn.com/abstract=1785996. 43 TP Crocker, ‘Torture, with Apologies’ (2008) 86 Texas Law Review 569, 585–93; Hamdi v Rumsfeld 542 US 507, 545 (2004) (Souter J concurring). 44 H Hänggi, ‘The Use of Force Under International Auspices: Parliamentary Accountability and “Democratic Deficits”’ in Double Democratic Deficit (n 39) 3. 38 39
458 Barak Medina armed force for in moments of crisis such constitutional provisions will be ignored,45 is particularly powerful in reference to the second model of providing the government with unlimited powers whenever a state of emergency is declared.46 It seems that the only effective way to induce the legislature to take part in reviewing the executive is by the continued operation of a thick substantive notion of the rule of law at all times, and thus by applying the third model. The prediction that the legislature may take an important role in restraining an unjustified use of force is based mainly on the political costs that are involved in formally legitimizing taking certain measures. The Israeli experience in this respect is illuminating. As indicated above, while the GSS used force in interrogation for more than a decade, a policy which was approved by a public committee appointed to investigate the matter,47 the Knesset did not intervene to stop this practice. In 1999, when the Supreme Court held that absent explicit authorization such a practice is prohibited,48 the political costs of such legislation were proved to be prohibitive. The mere enforcement of the requirement to legislate was sufficient, despite the majority’s support of this measure, to change the practice. ‘Rulifying’ the fight on terror. Explicit authorization ensures that the government acts on the basis of general, overt, predictable norms, in accordance with the requirements of formal justice.49 Refraining from setting guidelines ex ante is unfair to actors who are exposed to the risk of bearing civil and criminal liability if it turns out that their judgement regarding the permissibility of the infringement is different from that of the ex post reviewer. In fact, prior authorization may be necessary to encourage risk-averse agents, who are reluctant to ‘dirty their hands’, to nevertheless promote the overall good when such action involves a justified infringement of a moral constraint.50 Thus, even those whose major concern is inducing the executive to take the necessary measures in response to security threats should in fact support the requirement of prior authorization in legislation.51 Such legislation may also narrow the scope of judicial review of military actions.52 Symbolic aspects. Some scholars base their objection to legislation by pointing at the expressive role of authorizing rights infringements. It rests on the notion that legal provisions express attitudes, shape public perceptions, and may thus inflict ‘expressive’ harm. For instance, a statute authorizing torture, even only in very rare circumstances, 45 Carl Schmitt, Political Theology, Four Chapters on the Concept of Sovereignty (G Schwab tr, University of Chicago Press, 2005). For a critical analysis of Schmitt’s theory, see, eg D Dyzenhaus, ‘Schmitt v Dicey: Are States of Emergency Inside or Outside the Legal Order?’ (2006) 27 Cardozo Law Review 2005, 2030–37; O Gross, ‘The Normless and the Exceptionless Exception: Carl Schmitt’s Theory of Emergency Powers and the “Norm-Exception” Dichotomy’ (2000) 21 Cardozo Law Review 1825. 46 Ferejohn and Pasquino (n 30) 216. 47 Commission of Inquiry into the Methods of Investigation of the General Security Service Regarding Hostile Terrorist Activity, Report (1987). For an English translation of excerpts, see (1989) 23 Israel Law Review 146. For a critical analysis, see M Kremnitzer, ‘The Landau Commission Report – Was the Security Service Subordinated to the Law, or the Law to the “Needs” of the Security Service?’ (1989) 23 Israel Law Review 216. 48 Public Committee against Torture in Israel v State of Israel (n 19). 49 See, eg HLA Hart, The Concept of Law (Oxford, Clarendon Press, 1961). 50 NA Sales, ‘Self Restraint and National Security’ (2012) 6 Journal of National Security Law and Policy 227; M Walzer, ‘Political Action: The Problem of Dirty Hands’ (1973) 2 Philosophy and Public Affairs 160. 51 See, eg J Nzelibe and J Yoo, ‘Rational War and Constitutional Design’ (2006) 115 Yale Law Journal 2512, 2530–32; Ferejohn and Pasquino (n 30) 216–18; DN Pearlstein, ‘Form and Function in the National Security Constitution’ (2009) 41 Connecticut Law Review 1549. 52 Issacharoff and Pildes, (n 16); Barak-Erez (n 30) 893.
Legislative Responses to Security Threats 459
can be considered disrespectful of human dignity. According to this objection, it is better that agents would act extra-legally when necessary, without explicit prior authorization.53 This position is implicit in the German Constitutional Court’s judgment to invalidate a statute authorizing officials to shoot down an aircraft that was being wielded as a deadly weapon.54 Arguably, prior authorization may turn the actor’s decision-making process into a rather technical assessment of whether the conditions set forth by the legislature are met, without giving sufficient attention to the nature of the action as an infringement of rights.55 I find these arguments unpersuasive.56 For one thing, if no principles govern the behaviour of agents in extreme cases, how would agents decide whether the circumstances they face are truly extreme? Once it is accepted that basic liberties may justifiably be infringed under certain circumstances, there is no escape from delineating those circumstances. In addition, this argument is relevant in the current context only with respect to the marginal effect of authorizing certain measures through legislation, since, at least in Israel, the judiciary sets rules of behaviour that justify taking numerous anti-terrorist measures, and the adverse symbolic effect already exists. It is doubtful whether the legislations’ marginal expressive effect is substantial, given forward-looking ‘judicial legislation’. Another argument that can be raised against ex ante authorization is that such author ization is likely to bring about unjustified activities.57 It may also induce the establishment of institutions that would make infringements an even more readily available option.58 It was also suggested that legislating taking extreme measures may enable officials to maintain what is known as ‘role distance’,59 in which an official disassociates themself from the action’s moral implications. While these slippery slope arguments cannot be ignored, they do not necessarily preclude predetermined guidelines for the permissibility of infringements. The absence of ex ante authorization may result in people refraining from taking justified measures, as suggested above. Moreover, such authorization may be accompanied by measures that deter unjustified activities. Legislation authorizing the government to take extreme measures in response to security threats may well have adverse symbolic effects. However, making these symbolic effects a decisive reason against legislation directly contradicts the very basic reasons for insisting on legislation in the first place, including the deliberative value of the process of legislation and the political costs associated with such legislation. If the 53 See, eg Posner (n 26) 152–58; O Gross, ‘Are Torture Warrants Warranted? Pragmatic Absolutism and Official Disobedience’ (2004) 88 Minnesota Law Review 1481, 1526–34; Harel and Sharon (n 28). 54 115 BVerfGE 118, 128 (2006) (Ger). 55 cf ES Anderson and RH Pildes, ‘Expressive Theories of Law: A General Restatement’ (2000) 148 University of Pennsylvania Law Review 1503, 1512; SH Kadish, ‘Torture, the State and the Individual’ (1989) 23 Israel Law Review 345, 353. 56 For a critical discussion of this objection, see E Zamir and B Medina, Law, Economics, and Morality (New York, Oxford University Press, 2011) 117–22. 57 H Shue, ‘Torture’ (1978) 7 Philosophy and Public Affairs 124, 141; D Luban, ‘Liberalism, Torture, and the Ticking Bomb’ (2005) 91 Virginia Law Review 1425, 1446; HH Koh, ‘Can the President be Torturer in Chief?’ (2006) 81 Indiana Law Journal 1145, 1165; JH Langbein, ‘The Legal History of Torture’, in S Levinson (ed), Torture: A Collection (New York, Oxford University Press, 2004) 93, 101; JB Elshtain, ‘Reflection on the Problem of “Dirty Hands”’ ibid (Torture: A Collection) 77. 58 See, eg H Shue, ‘Torture in Dreamland: Disposing of the Ticking Bomb’ (2006) 37 Case Western Reserve Journal of International Law 231, 238; Kremnitzer (n 47) 254–57. 59 M Dan-Cohen, Harmful Thoughts: Essays on Law, Self, and Morality (Princeton, Princeton University Press, 2002) 233–35; M Gur-Arye, ‘Justifying the Distinction between Justification and Power’ (2011) 5 Criminal Law and Philosophy 293.
460 Barak Medina government finds it impossible to openly justify its activities, it should not be shielded from scrutiny by referring to arguments of the sort of ‘acoustic separation’.60 If the ‘noise’ is troubling, the government should not make it, not merely conceal it behind ‘acoustic’ walls. Preventing effective address of the threats. As indicated, supporters of the first model also point at the possible ‘cost’ of insisting on legislative authorization – the arguable concern is that enforcing the requirement of authorised legislation might prevent the government from effectively addressing the threats. This concern is difficult to sustain. The doctrine under consideration does not require the government to obtain the legislature’s approval to employ certain measures (or, for that purpose, to declare ‘war’ or a state of emergency). It merely imposes on the Executive Branch the task of convincing the legislature that certain measures should be included in the arsenal of actions that can be used, in a given set of circumstances, in the fight against terrorism. Indeed, when the circumstances are exigent, the Executive Branch may act even without legislative authority, based on some form of ‘emergency powers’, or to take immediate measures as needed in cases of necessity. The government is required, however, to establish that the circumstances are indeed exigent. But as a general matter, the fight against terrorism as such should not be regarded as constituting a state of emergency, as the measures that governments consider as essential to pursue this fight are well-known. It is the Executive Branch’s responsibility to initiate legislative proceedings to obtain the authority it deems necessary. VI. CONCLUDING REMARKS
Measures which are directed against suspected terrorists are measures which infringe upon basic human rights, irrespective of the nationality of the targeted persons or the place in which the governmental activity takes place. Such measures can be employed only subject to legislative authorization (or when the circumstances are exigent). Legislation provides the required authority only when it explicitly addresses specific measures and details the circumstances in which each of them can be employed. Insisting on legislation, which requires the government to act on the basis of general, predefined norms, is expected to induce several benefits: it imposes on the representatives a duty to take direct and formal part in a deliberation on the relevant aims and means, which may have a substantial effect on actual outcomes and serves essential intrinsic values. Democracies should thus follow this path in determining their responses to security threats.
60 M Dan-Cohen, ‘Decision Rules and Conduct Rules: On Acoustic Separation in Criminal Law’ (1984) 97 Harvard Law Review 625.
30 Of Law, Constitutions and Security ADAM TOMKINS
I. INTRODUCTION
O
NE OF MY favourite books as a young boy was Douglas Adams’s sci-fi comedy, The Hitchhiker’s Guide to the Galaxy. The Guide had the words ‘Don’t Panic’ emblazoned on its cover, as did my copy of the book. The instruction was meant for inter-galactic space travellers seeking to get about the planets on the cheap. But, in the real world, it is a motto that we would all do well to live by when we are confronted with threats to our security. Sometimes quick and decisive action is needed, for sure. Sometimes a loud call for help is required. And sometimes we will have no option but to reach for coercive powers and, in extremis, for the use of lethal force. But what we should never do is panic. Yet panic was for many years the routine reaction of British Governments when confronted with threats to security. From the 1970s the United Kingdom has faced a range of terrorist threats. Many of these are ongoing. At the time of writing the current level of threat from international terrorism in the UK is assessed by the authorities to be ‘substantial’, meaning that ‘a terrorist attack is a strong possibility’. And the current level of threat from terrorism associated with Northern Ireland is assessed in Northern Ireland to be ‘severe’, meaning that ‘a terrorist attack is highly likely’. Between 1972 and 1992 some 3,000 deaths and over 40,000 terrorist incidents were attributed to terrorism in Northern Ireland. (The population of Northern Ireland is about 1.7 million people.) In July 2005 52 people were killed when a series of bombs exploded on London’s public transport network, in a coordinated set of suicide attacks perpetrated by Islamic extremists. The terrorism that the United Kingdom has experienced is nowhere near to being on the scale of that which Israel has confronted since its founding in 1948 and, unlike Israel, the security threat to the UK is not existential in nature. Britain’s neighbours are not aggressive towards the UK and, as Lord Hoffmann observed in his infamous opinion in the Belmarsh case,1 while ‘fanatical groups of terrorists [have the ability] to kill and destroy’, they cannot ‘threaten our institutions of government or our existence as a civil community’.2 Such a ‘public emergency’ as the UK faces is not such as to constitute a threat to the ‘life of the nation’.3 A v Secretary of State for the Home Department [2004] UKHL 56, [2005] 2 AC 68 (hereinafter: Belmarsh). ibid para 96. See further on this case below. 3 In this view Lord Hoffmann was alone among the nine Law Lords who heard the appeal in Belmarsh. Eight 1 2
462 Adam Tomkins This is not so with Israel. This chapter reflects on those written for this book by Daphne Barak-Erez and Barak Medina. It does so by contrasting the stories they tell with the analogous story in the United Kingdom. It must be understood, however, that UK–Israeli comparisons in the arena of national security are neither equivalents nor parallels. The security contexts of the two countries are vastly different. Yet, as we shall see, the ways in which the British and Israeli constitutional orders have developed their responses to terrorism and related security threats have much in common. It may not be on a Middle Eastern scale, but terrorism in the UK has been more than serious enough to send British Governments of all parties into spasms of panic. For 35 years the standard governmental response to a terrorist outrage was to call for the parliamentary draftsman, so that more pages could be added to the already burgeoning statute book. And, for 35 years the legislation that would follow on from an act of terrorism would be rushed through Parliament in what can only be described as a fit of legislative panic. Even after 9/11 we saw the by now familiar routines revisited, as the 129 sections and eight schedules of the Anti-terrorism, Crime and Security Act 2001 were taken through the House of Commons with only 16 hours of debate. This, despite the fact that in the Terrorism Act 2000 the United Kingdom already possessed the ‘most rigorous’ anti-terrorism legislation ‘in Europe’.4 The courts, too, were prone to panic when questions of national security came before them. Judicial panic took a different form from that seen in the legislature. In Parliament the symptom of panic was unseemly haste. In the courts it was denial. When governments sought to make arguments based on national security, all too often the courts just did not want to know anything more. National security being too hot a political potato, the judges panicked, and threw it back directly at government. Thus, if ministers claimed that it was necessary in the interests of national security to withhold evidence from disclosure in legal proceedings, so be it. The minister’s say-so was good enough for the judges: not for them to investigate to see if security considerations really did require the evidence to be withheld.5 Likewise, if ministers deemed it necessary in the interests of national security to deport someone from the UK, so be it. The courts would not intervene.6 Nor did they intervene to stop internment during the two World Wars, despite powerful, indeed ringing, dissenting opinions from Lord Shaw in 1917 and Lord Atkin in 1942.7 This is not an attractive picture. In the twentieth century terrorism or other perceived threats to the security of the state elicited over-reaction from the government of the day, appeased by legislators who were in such a rush that they had no time to reflect on what they were being asked to do, and left unchecked by the courts, whose judges preferred to look the other way until normalcy returned.
of their Lordships ruled that the UK did face a public emergency threatening the life of the nation. A Grand Chamber of the European Court of Human Rights subsequently agreed with the majority of the House of Lords: see A v United Kingdom (2009) 49 EHRR 29. 4 The verdict of Parliament’s Joint Committee on Human Rights: see its 2nd report (2001–02, HL 37, HC 372) para 30. 5 Duncan v Cammell Laird [1942] AC 624. This rule held until 1968, when a degree of judicial oversight was introduced: see Conway v Rimmer [1968] AC 910. 6 R v Secretary of State for the Home Department, ex p Hosenball [1977] 1 WLR 766 (CA) and R v Secretary of State for the Home Department, ex p Cheblak [1991] 1 WLR 890 (CA). 7 R v Halliday, ex p Zadig [1917] AC 260 and Liversidge v Anderson [1942] AC 206.
Of Law, Constitutions and Security 463
Unhappily, in the twenty-first century we continue in the United Kingdom to make significant quantities of new law in the national security and counter-terrorism fields. Equally unhappily, we continue to witness an apparently ever growing volume of serious appeal court litigation in national security law, some of it raising allegations of the most egregious abuses, including the return to Britain of the dread and disgrace of torture.8 Less unhappily, in their most recent attempts to grapple with security and terrorism, both the legislature and the courts in the United Kingdom appear (at least for the time being) to have stopped panicking, as they have started to approach their constitutional functions with more maturity. II. THE LAW-MAKING PROCESS
In Parliament, our most recent counter-terrorism and national security legislation has not been fast-tracked through its legislative process. Since the Anti-terrorism, Crime and Security Act 2001, only two terrorism Bills have been fast-tracked and both of these have since been replaced with other legislation that was not fast-tracked.9 Neither the Terrorism Act 2006 nor the Counter-Terrorism Act 2008 were fast-tracked through Parliament: the former took five months to pass through Parliament and the latter took 10 months. The legislation to replace control orders was not fast-tracked through Parliament,10 and neither was the legislation that placed terrorist asset-freezing on a permanent statutory footing.11 At the time of writing the Justice and Security Bill is before Parliament: it is expected to pass into law in 2013. It has not been fast-tracked. But it is not just the speed of the legislative process that is important. So too is its quality. And here, too, Parliament has started to take its constitutional responsibilities far more seriously. In order to understand this, we need to put it in context. In the British system the majority of legislation passed by Parliament is government legislation. The Government12 do not have an exclusive right of legislative initiative – backbench MPs and Peers in the House of Lords may introduce bills, in certain circumstances. But the Government have effective control of the parliamentary timetable, at least in the House of Commons, and this means that very little legislation is passed by Parliament without government approval.13 Likewise, almost all of the Government’s legislation is passed.14 8 A v Secretary of State for the Home Department (No 2) [2005] UKHL 71, [2006] 2 AC 221; R (Mohamed) v Secretary of State for Foreign and Commonwealth Affairs [2010] EWCA Civ 65, [2011] QB 218; R (Mousa) v Secretary of State for Defence [2011] EWCA Civ 1334, [2012] HRLR 6. See also the report of the Baha Mousa Inquiry: www.bahamousainquiry.org. 9 The Prevention of Terrorism Act 2005 was repealed and replaced by the Terrorism Prevention and Investigation Measures Act 2011, and the Terrorist Asset-Freezing (Temporary Provisions) Act 2010 was repealed and replaced by the Terrorist Asset-Freezing etc Act 2010. 10 The Terrorism Prevention and Investigation Measures Act 2011. 11 The Terrorist Asset-Freezing etc Act 2010. 12 By ‘Government’ I mean specifically the Executive Branch – the administration that is for the time being in office – headed by the Prime Minister and the Cabinet. 13 An exception is the Hunting Act 2004, which was passed by Parliament despite the Government not being in favour of the manner in which it sought to ban hunting in England and Wales. 14 An exception is the House of Lords Reform Bill (2012–13), which was withdrawn by the Government when it appeared that no programme motion could be agreed in the House of Commons. Without a programme motion the Government would have had much less control on the amount of parliamentary time devoted to the Bill. The fear was that so much time would be consumed with this Bill that the Government would struggle to make progress with its other legislative priorities.
464 Adam Tomkins Parliament’s principal legislative job, then, is neither to generate legislation of its own initiative nor to seek to block the legislative proposals of the Government. It is, rather, to debate, amend, revise and improve the Government’s legislation. For a variety of reasons to do with party controls and the rules of parliamentary procedure, this is a function that has come to be far better performed by the unelected House of Lords than by the elected House of Commons. The Commons is unrivalled as a theatre in which ministers must account for their policies, actions and decisions, and it is still in the House of Commons that the great debates are staged, where the issues of the day are aired and where opinion is defended and tested. These are all invaluable parliamentary tasks, but as a strictly legislative assembly, the modern House of Commons is gravely limited. Routinely, it is in the House of Lords that we see the most careful and detailed attention being paid to legislative scrutiny. The Lords take seriously their constitutional function of being a ‘revising chamber’ and they devote huge energies to this. In addition, the rules of the Upper House do not permit the Government to control parliamentary time in the Lords to anything like the extent that is true for the Commons. Finally, no political party enjoys a majority in the House of Lords, and the political appointees have to work alongside ‘cross-bench’ Peers who are appointed to the House not because of their party allegiance but because of their expertise and experience. It is relevant in our context that lawyers are among the groups over-represented on the cross-benches. With this context in mind, Parliament’s current contribution to law-making in the national security and counter-terrorism fields can be illustrated with three examples. These examples span the end of the last Labour Government (in office from 1997–2010) and the beginning of the current Conservative–Liberal Democrat Coalition Government (in office since May 2010). The first example dates from the period 2006–08. The Government sought to increase the maximum length of pre-charge detention for terrorist offences to 42 days. The normal maximum period in English law for which a suspect may be detained by the police before being charged with a criminal offence is 24 hours. In the most serious cases this limit can be extended to a maximum of 96 hours. For terrorist offences the limit was set at five days (from 1974–2000), seven days (from 2000– 03), 14 days (from 2003–06) and 28 days (from 2006). Ministers indicated that they wished the limit to be increased further to 42 days. Parliamentary committees responded by seeking evidence as to whether this further increase was necessary.15 They failed to find any and, indeed, they found authoritative evidence that it was not necessary. Regardless, ministers pursued the matter. When the issue came to a vote in the House of Lords, the Government were defeated by 309 votes to 118,16 a loss so crushing that the Government abandoned their plan and did not seek to revive it in the Commons. Our second example concerns terrorist asset-freezing. Asset-freezing is a key com ponent of effective counter-terrorism, and is mandated by a series of United Nations Security Council Resolutions, including UNSCR 1267 and UNSCR 1373 (as amended). The United Kingdom gave domestic legal effect to these Security Council Resolutions through Orders in Council made by ministers under the United Nations Act 1946. In Ahmed v HM Treasury the UK Supreme Court ruled that the Orders in Council were 15 To this end the work of the House of Commons Home Affairs Committee and of the Joint Committee on Human Rights was invaluable. For a detailed account, see A Tomkins, ‘Parliament, Human Rights, and CounterTerrorism’ in T Campbell, K Ewing and A Tomkins (eds), The Legal Protection of Human Rights (Oxford, Oxford University Press, 2011) 20–28. 16 See HL Deb 13 October 2008, vol 704, cols 491–544.
Of Law, Constitutions and Security 465
ultra vires.17 Powers as intrusive of fundamental human rights as those contained in the Orders required express authorisation in primary legislation, the Court ruled: general Order-making powers such as those contained in the United Nations Act could not be relied on as authority for making Orders as coercive as these. Operative provisions of the Orders were therefore quashed. The Government could not allow the UK’s regimes of terrorist asset-freezing to lapse, so fast-track legislation was rushed through Parliament in order to give emergency legislative authority to that which the Supreme Court had ruled unlawful. This legislation was time-limited and before the end of 2010 was replaced by permanent legislation. In the making of the Terrorist Asset-Freezing etc Act 2010 Parliament amended the Government’s proposals in a number of ways. Perhaps the most important two were as follows. The Government had proposed that the Treasury should be empowered to freeze the assets of persons reasonably suspected of involvement in terrorist activity. The House of Lords persuaded the Government to change this test of reasonable suspicion to one of reasonable belief.18 Secondly, the Government had proposed that persons wishing to challenge a decision by the Treasury to freeze their assets would be able to go to court but that the court would be able only to review the lawfulness of the decision. Applying the ordinary rules of English administrative law this would mean that the court would be able to quash the decision only if it ruled that the decision was so unreasonable that no reasonable decision-maker could have taken it.19 Given the gravity of the decisionmaking we are talking about here, this would have been a very limited jurisdiction. Again the House of Lords persuaded the Government to amend the legislation, so that it now provides for a full right of appeal to the courts.20 Our final example is the Justice and Security Bill which, at the time of writing, remains before Parliament. A recurrent problem in national security law is how to make litigation fair. Fundamental rules of open and natural justice require that parties to litigation are able to see and to challenge the evidence of other parties in the case. On occasion it is extremely difficult to square these rules with the need to keep operationally sensitive intelligence material secret. A classic example is when the Government determines that the ongoing presence of a foreign national in the United Kingdom is no longer conducive to the public good and, in the interests of national security, that the person should be deported. The deportee challenges the decision to deport him. The Secretary of State responds by explaining that the case against the deportee comprises sensitive secret intelligence, which cannot be disclosed to him. How can a case such as this be conducted fairly? In 1996 the European Court of Human Rights (ECtHR) ruled that the procedures then used for such cases in the United Kingdom were unfair.21 In response a new procedure was introduced into UK law: closed material procedure (CMP). Under CMP evid ence that cannot be served on parties in litigation owing to its sensitivity or secrecy is served instead on a special advocate (SA). The SA is a security-cleared lawyer appointed to represent the interests of a party to litigation in closed process. Once the closed mater ial has been served on the SA, they are prohibited from communicating directly with the Ahmed v HM Treasury [2010] UKSC 2, [2010] 2 AC 534. See the Terrorist Asset-Freezing etc Act 2010, s 2(1). 19 This is the test of Wednesbury unreasonableness: see Associated Provincial Picture Houses Ltd v Wednesbury Corporation [1948] 1 KB 223 (CA). 20 See the Terrorist Asset-Freezing etc Act 2010, s 26. 21 Chahal v United Kingdom (1997) 23 EHRR 413. 17 18
466 Adam Tomkins party on whose behalf they act. When a court goes into a closed process everyone is required to leave the courtroom except for the judge, the Government’s lawyers and the SA. Other parties, their lawyers and members of the public must wait outside until the court re-opens its doors. Matters dealt with in closed process are determined by the court in a closed judgment. Closed judgments remain secret: only the Government and the SA in the particular case will be able to read them. CMP was first used in the UK for deportation and other related immigration hearings.22 Subsequently the use of CMP and SAs grew to control orders cases,23 to assetfreezing cases,24 and to other counter-terrorism cases, such as under the Terrorism Prevention and Investigation Measures Act 2011.25 In Al Rawi v Security Service,26 Al Rawi and certain other former detainees at Guantanamo Bay sued the UK’s security and secret intelligence services for damages. The claimants alleged that the security and secret intelligence services, as well as other departments and agencies of the UK Government, were complicit in the torture, illegal rendition and other ill-treatment they had suffered at and en route to Guantanamo. The defendants argued that so much of the material they wished to serve in their defence was sensitive and secret that a fair trial of the action could proceed only under CMP. This was the first occasion on which it was suggested that a damages action should proceed under CMP. Hitherto all use in the UK of CMP had been in litigation testing the legality of Government decision-making rather than the Government’s liability. Both the Court of Appeal and the Supreme Court ruled that the courts possessed no inherent power to order that a damages action proceed under CMP. Closed material procedure, ruled the appeal courts, was such an incursion into the common law principles of open and natural justice that it required express statutory authorisation. The Government settled Al Rawi’s claims in a confidential out-of-court settlement later asserted to be in the region of £20 million plus costs. Plainly frustrated with having to pay millions of pounds of damages to people the Government considered to be highly dangerous, ministers announced their intention to legislate so that, in the future, claims such as Al Rawi’s could be heard in accordance with CMP. This, the Government contended, was a necessary step in order for such claims to be fairly defended in effective litigation. The result was the Justice and Security Bill, introduced into Parliament in May 2012. The Government’s desire to legislate in this manner is readily understandable given the result in Al Rawi. But the means by which the Government proposed that CMP should be available in ordinary civil litigation left a lot to be desired. Under the Government’s proposals, the Justice and Security Bill would have provided for CMP to be used in civil proceedings on the application only of the Secretary of State. The court would have been compelled to grant such an application where the proceedings in question would otherwise have required any party to disclose material whose disclosure would be damaging to national security. These provisions were amended during parlia Under the Special Immigration Appeals Commission Act 1997. Under the Prevention of Terrorism Act 2005. 24 Under the Counter-terrorism Act 2008 and the Terrorist Asset-Freezing etc Act 2010. 25 For a thorough account of the position down to 2009, see E Metcalfe, Secret Evidence (London, Justice, 2009). A more recent account is Amnesty International, Left in the Dark: The Use of Secret Evidence in the UK (London, AI Publications, 2012). 26 Al Rawi v Security Service [2011] UKSC 34, [2012] 1 AC 531. 22 23
Of Law, Constitutions and Security 467
mentary debates on the Bill in the House of Lords. As amended, the Bill provided that CMP may be used in civil proceedings on the application of any party to those proceedings or on the court’s own motion. The court would have a discretion as to whether to grant an application that proceedings go into a closed process. In exercising that discretion the court was to consider: (a) whether the degree of harm to national security caused by any disclosure of material outweighs the public interest in the fair and open administration of justice; and (b) whether a fair determination of the proceedings is possible by any other means. As with the Lords’ earlier assault on the Government’s plan over 42-day detention, the amendments to the Justice and Security Bill were the work principally of cross-bench and other lawyers. The amendments made to this scheme had three objectives: first they sought an equality of arms that was missing from the original Bill; secondly, the amendments sought to build into the CMP process an element of judicial balancing, such that adopting a CMP in any trial would occur only where, on balance, the damage to national security in not doing so outweighs the damage to the principles of open and natural justice that inevitably arises in doing so. Thirdly they sought to ensure that conducting a trial under CMP is genuinely a measure of last resort. Most, but not all, of these Lords amendments survived, in one form or another, in the House of Commons, but a full account is beyond the scope of this paper. These, then, are my three examples – pre-charge detention, asset-freezing and closed material procedure – of improved legislative scrutiny in the UK Parliament of matters of national security and counter-terrorism law. It is notable that the story outlined here of more active parliamentary involvement in the making of national security law bears a striking resemblance to the story Barak Medina tells of the increased, if still partial, activism of the Knesset in the same arena. Some of the parliamentary victories here are more significant than others: stopping 42 days was a bigger deal than substituting reasonable belief for reasonable suspicion, for example. But what each of these examples has in common is that Parliament was both willing and able to scale back the extent of the powers that the Government was claiming they needed. Thus, the Government asserting that a particular power is necessary in the interests of national security is no longer any guarantee that the power will be legislated for. Parliament will make its own judgement of what is needed. III. THE ROLE OF THE COURTS
Surveying a number of the cases mentioned above – including Halliday, Liversidge v Anderson and Hosenball – David Dyzenhaus remarked that ‘the judicial record in enforcing the rule of law’ in cases of national security ‘is at worst dismal, at best ambiguous’.27 The twentieth-century record was indeed dismal, culminating in Lord Diplock’s famous dictum in the GCHQ case that national security is ‘par excellence a non- justiciable question’.28 The judicial record in the twenty-first century, while patchy, is certainly a lot less dismal than that of the twentieth. As I did in the previous part, so too here I will give three examples. D Dyzenhaus, The Constitution of Law (Cambridge, Cambridge University Press, 2006) 17. Council of Civil Service Unions v Minister for the Civil Service [1985] AC 374, 412 (HL).
27 28
468 Adam Tomkins The first is the decision of the House of Lords in the Belmarsh case.29 Among other matters, the Anti-terrorism, Crime and Security Act 2001 permitted the Secretary of State to certify that a person is a terrorist or that a person’s presence in the UK is a risk to national security. Persons so certified were liable to be detained indefinitely without charge or trial. In Belmarsh an eight-to-one majority of the Law Lords ruled that this scheme was unlawful. The Government had accepted that the scheme was incompatible with Article 5 of the European Convention on Human Rights30 but had sought to derogate from this provision owing to the ‘public emergency’ that the UK faced after 9/11. While the House of Lords upheld the legality of the derogation (Lord Hoffmann dissenting, as we saw above), measures taken in pursuance of a derogation must be ‘strictly required by the exigencies of the situation’.31 Their Lordships ruled that the scheme of indefinite detention without charge – a variant of internment – did not meet this test. In Chahal the ECtHR had ruled that it is unlawful for a state party to the Convention to deport someone to a country where they would face a real risk of torture.32 Whilst this is a valuable and powerful reminder of the unqualifiedly evil nature of torture, the ruling created a problem. What should a European state do with people whom it would like to deport for reasons of national security but who may not be deported because of the rule in Chahal? The UK’s answer post 9/11 was the scheme of indefinite detention without charge legislated for in the Anti-terrorism, Crime and Security Act. Because the problem to which the scheme was an answer related only to foreign nationals (no state may lawfully deport its own nationals), the scheme applied only to persons subject to immigration control: British citizens could not be detained under the Act. Yet, the House of Lords found, the security threat to the UK post 9/11 came not only from foreign nationals resident in the UK but also from British nationals. If indefinite detention without charge was not necessary for British nationals, how could it be ‘strictly required’ for persons subject to immigration control? This was the flaw that fatally undermined the scheme, in the judgment of the Law Lords. Our second example relates back to closed material procedure (CMP) and special advocates (SAs). We saw above that when CMP is used, evidence that the Government deems to be sensitive is not served on the other parties to litigation, but is shown only to the court and to the SAs. In Secretary of State for the Home Department v MB33 the House of Lords ruled that this procedure would have to be ‘read down’ so as to ensure that it was compatible with the Convention right to a fair trial under Article 6 of the ECHR. Lower courts struggled to know what this process of ‘reading down’ would entail in any particular case and the matter returned to the House of Lords in Secretary of State for the Home Department v AF.34 Immediately before AF was argued in the House, the ECtHR handed down its judgment in A v United Kingdom.35 In this case, Strasbourg ruled that, in order to comply with the right to a fair trial, a party repre Belmarsh (n 1). The Convention for the Protection of Human Rights and Fundamental Freedoms (European Convention on Human Rights, as amended) (ECHR). 31 ibid Art 15. 32 Chahal is one of a number of cases in which the ECtHR has played a decisive role in reshaping British national security law. That European and international law have had this influence is a feature that the UK shares with Israel, as Daphne Barak-Erez explains towards the end of her chapter. 33 Secretary of State for the Home Department v MB [2007] UKHL 46, [2008] 1 AC 440. 34 Secretary of State for the Home Department v AF [2009] UKHL 28, [2010] 2 AC 269. 35 A v United Kingdom (2009) 49 EHRR 29. 29 30
Of Law, Constitutions and Security 469
sented by a SA must be provided with ‘sufficient information about the allegations against him to enable him to give effective instructions to the special advocate’.36 The House of Lords relied on this judgment when ruling in AF that, as long as this requirement is satisfied ‘there can be a fair trial notwithstanding that the [individual] is not provided with the detail or the sources of the evidence forming the basis of the allegations’.37 This is sometimes known as the AF ‘gisting’ requirement. This was a level of disclosure that the Government felt they could not always live with and, in a number of instances, the Government elected to abandon proceedings rather than to disclose the material that would be necessary in order to comply with the rulings in A v United Kingdom and AF. In a series of cases the Government has sought to restrict the scope of the gisting requirement. They were successful in Tariq v Home Office38 in persuading the Supreme Court to rule that this level of disclosure was not required in a race discrimination claim brought by a former civil servant. Our final example concerns deportation. The British Government has sought several routes out of the problem that was set by the ECtHR in Chahal. One of their strategies has been to seek diplomatic assurances that foreign nationals whom the UK would like to deport will not be tortured upon return to their country of origin (or country of citizenship). In a number of instances, courts have ruled that the assurances obtained by the British Government are insufficiently robust to render deportation lawful: despite the assurances given, there remains a real risk of treatment contrary to Article 3 ECHR. The Court of Appeal came to this conclusion in respect of assurances obtained from Gaddafi’s regime in Libya.39 The House of Lords upheld the legality and robustness of similar assurances that had been obtained from Algeria and Jordan40 but, in the latter case, the ECtHR ruled that the person concerned (Abu Qatada) could not be deported to Jordan because he would face the prospect of being tried there on the basis of evidence obtained by torture.41 Further assurances subsequently obtained from the Jordanian authorities were eventually held to be sufficient to guarantee that any trial Abu Qatada could face would be free of the taint of torture. He was deported to Jordan on 2013. As with our parliamentary examples, what the Belmarsh, AF and Abu Qatada cases each illustrate is that the courts are no longer anything like as reluctant as they formerly were in seeking to subject to searching scrutiny the Government’s claims as to what is necessary in the interests of national security. This does not mean that the Government will always lose (any more than the case law of the Israeli Supreme Court means that the Israeli Government will always lose in that Court). But it does mean that the Government no longer always wins merely by virtue of playing the national security card. IV. CONCLUSIONS
Two related conclusions may be drawn from these various legislative and adjudicative episodes. Each of them is also at least implicit in the papers published here by Daphne ibid para 220. AF (n 34) para 59. 38 Tariq v Home Office [2011] UKSC 35, [2012] 1 AC 452. 39 AS (Libya) v Secretary of State for the Home Department [2008] EWCA Civ 289, [2008] HRLR 28. 40 RB (Algeria) v Secretary of State for the Home Department [2009] UKHL 10, [2010] 2 AC 110. 41 Othman (Abu Qatada) v United Kingdom (2012) 55 EHRR 1. 36 37
470 Adam Tomkins Barak-Erez and Barak Medina. The first is that, properly understood and despite all too frequent protests to the contrary, matters of national security and counter-terrorism are not for the executive alone. Whilst governments, ministers and the civil servants and military officers who are responsible to them will have the lead role in determining what measures are necessary in the name of national security, they do not act alone. Rather, national security is the shared responsibility of all the branches. Secondly, ‘emergency constitutionalism’ is the wrong paradigm for assessing matters of national security and counter-terrorism law. This has been a dominant narrative in the United States in the decade since 9/11 but, as British and even more so as Israeli experience shows, threats to security may be more or less permanent, not exceptional. The more we can do to normalise our response to such threats, and the less we reach for the law of the exception as a justification for the otherwise unjustifiable, the more likely we are to ensure that our cherished ideals of constitutional justice are not sacrificed. This, I think, may be Israel’s greatest contribution to the debate. Even in a country that has lived under a permanent state of emergency, considerable efforts continue to be made within the Knesset and by the Supreme Court to normalise national security law. In this way may constitutional responsibility for national security law be shared in Israel. And if Israel can do this, given the extraordinary threats it has faced throughout its existence, so surely can all democracies that are committed to the rule of law. In the United Kingdom we have come to see this only recently. In the twentieth century the parliamentary and judicial records alike were dismal. In the twenty-first, conscious that the threats we face may be with us for a while, and under the close watch and scrutiny of international human rights law, we seem to have stopped panicking about national security and to have started to act with at least a degree of maturity. Reflective deliberation on what powers governments should have and robust adjudication of disputes about the exercise of those powers are the twin hallmarks of democratic constitutions committed to the rule of law. Both Britain and Israel face ongoing challenges (as do many other countries around the world) in ensuring that in the national security field these hallmarks are adhered to. The papers by Daphne Barak-Erez and Barak Medina give us reason to hope that in Israel’s case the challenge can be met. In the end, our security and counter-terrorism strategies must conform to the constitutional values we hold most dear. Responsibility for ensuring that this happens is shared by us all.
31 Jewish and Democratic: Three Zionisms and Post-Zionism* CHAIM GANS
I. INTRODUCTION
T
HERE ARE THREE defining tenets that constitute the common denominator for all the versions of Zionism. The first is that the Jews, apart from sharing a religion, also form an ethno-cultural nation. The second is the ethno-cultural principle. According to it, members of groups sharing a common history and culture have fundamental, morally significant interests in adhering to their culture and in sustaining it for generations. Furthermore, these interests warrant political recognition and support, primarily by means of the right to national self-determination.1 The third tenet of Zionism is that the Jews should realise their right to ethno-national self-determination in Zion, namely, in the historical homeland from which the Jewish people had been cut off since antiquity. Post-Zionists reject all these tenets.2 Here I will concentrate on their rejection of the Zionist principle supporting the realization of the Jewish right to self-determination in Israel. According to one major version of post-Zionism, the ‘ethnic nationalism dominant in Israel, according to which the people as an ethno-cultural entity is identical with the “state” as an administrative-constitutional entity’, must be replaced with ‘territorial nationalism, according to which it is possible to separate nation and state in such a way that communal (or cultural, religious, ethnic etc) belonging will not be constitutionally protected’.3 According to some post-Zionist writers, at the constitutional level of the * This chapter is based on my forthcoming book A Political Theory for the Jewish People: Three Zionist Narratives (Haifa, Haifa University Press, 2013). It summarizes the gist of the argument in that book concerning the status of the Jews in Israel relative to that of the Palestinians. The book also contains parts relating to the status of the Jews in Israel relative to the Jews living outside it, and the Zionist historiography of Judaism. 1 For a detailed formulation of this interpretation of ethno-cultural nationalism, see C Gans, The Limits of Nationalism (Cambridge, Cambridge University Press, 2003) 7–38. 2 Some scholars distinguish between ‘rejectionist post-Zionism’ and ‘affirmative post-Zionism’. Post-Zionists belonging to the first group are anti-Zionist. Those belonging to the second group are not anti-Zionists. I think that this distinction is misconceived, and that the notion of post-Zionism should only refer to anti-Zionists. For a detailed discussion of this issue see the introduction to my book, C Gans, A Political Theory for the Jewish People: Three Zionist Narratives (Haifa, Haifa University Press, 2013). 3 U Ram, The Time of the ‘Post’: Nationalism and the Politics of Knowledge in Israel (Tel Aviv, Resling, 2006) 17 (in Hebrew). Ram is the clearest and the most detailed spokesman among writers who could be considered as belonging to this brand of post-Zionism, such as S Sand, The Invention of the Jewish people (New York, Verso, 2010); O Yiftachel, Ethnocracy: Land and Identity Politics in Israel/Palestine (Philadelphia,
474 Chaim Gans State of Israel neither world Jewry nor even Israeli Jewry itself ought to be protected. Like the Jews in many other places in the world, and like non-Jewish groups in Israel, the Jews in Israel are free to form their own associations. The liberal principle of freedom of association allows them to do so in the very same way that it allows soccer teams and people who share religions to form their own associations. Post-Zionist writers present three main arguments to support the above thesis. First, they assert that there is a fundamental contradiction in Israel’s perception of itself as a state which is both ‘democratic and Jewish’.4 Secondly, they claim that the liberal ideal of state neutrality requires separating (ethno-cultural) nationality and state for the same reasons that it requires a separation of Church and state.5 Thirdly, they invoke the alleged moral superiority of civic-territorial nationalism – such as, the type of nationalism embodied by the United States or France – over ethno-cultural nationalism, such as that of Germany or Serbia.6 Two major normative concerns underlie these arguments against the continued Jewish self-determination in the Land of Israel.7 One is a concern for equality. Another is a concern for freedom. In responding to this post-Zionist critique I will distinguish between three major interpretations of the Zionist idea: the proprietary, the hierarchical, and the egalitarian. I shall argue for three theses regarding the above post-Zionist critique. First, I will show that it is rather a feeble critique with relation to the conception of Zionism which is the most popular among Israeli – and perhaps not only Israeli – Jews, the proprietary conception of Zionism. Second, I will show that the above post-Zionist critique is at least partly justified with regard to a second conception of Zionism, ‘hierarchical Zionism’, which is common mainly among academics and professionals in fields related to polit ical thought and law. Third, I will show that with regard to a third understanding of Zionism, the one I argued for in A Just Zionism8 and which I shall call here ‘egalitarian Zionism’, the post-Zionist critique completely loses its force. In part II of this chapter I Pennsylvania UP, 2006); B Kimmerling, ‘Religion, Nationality and Democracy in Israel’ (1994) 50–51 Zmanim 116 (in Hebrew); J Shapira, ‘The Secular Politicians and the Place of Religion in Israel’ in M Mautner, A Sagi and R Shamir (eds), Multiculturalism in a Democratic and Jewish State (Tel Aviv, Ramot, 1998); Y Peled, ‘Strangers in Utopia: The Civic Status of Israel’s Palestinians’ (1993) 3 Theory and Criticism 21 (in Hebrew). This brand of post-Zionism is civic. According to it, Israel should be the state of the Israeli nation, a nation that will consist of both its Jewish and Arab citizens. Another version of post-Zionism could be branded ‘multicultural’. Its proponents are mainly post-colonial multiculturalists. As such, they believe that only Arabs and Mizrahi Jews, groups that suffered under Ashkenazi Zionism, should enjoy legal protections by the State of Israel. They oppose, together with the civic post-Zionists, Israel’s conception of itself as a Jewish state. The main proponents of multicultural post-Zionism are Y Shenhav and Y Yona, What is Multiculturalism? On the Politics of the Difference in Israel (Tel Aviv, Bavel, 2005). 4 Ram, The Time of the ‘Post’ (n 3) 187; Basic Law: Human Dignity and Liberty, s 1A; Basic Law: Freedom of Occupation, s 2. 5 For these reasons, see Ram, The Time of the ‘Post’(n 3) 91. 6 ibid 188; Sand, The Invention of the Jewish People (n 3). 7 The term ‘Land of Israel’ (‘Eretz Yisrael’ in Hebrew) does not merely refer to the land or the territory of the State of Israel. It is the land promised to the Jewish people in the Old Testament. Jews lived and were politically dominant in many parts of it mainly in the first millennium BC until the destruction of the second temple in 70 AD. The term denotes most of the land which today comprises the State of Israel and the Kingdom of Jordan. Since the 1920s, it has mainly denoted the territories under the British Mandate; that is, the land between the Jordan River and the Mediterranean. It thus overlaps with historic Palestine. Another name for the Land of Israel in Jewish tradition (which appears frequently in the Old Testament and in liturgy) is Zion – also a traditional name for the Holy City of Jerusalem. This of course explains why the ideology discussed in this chapter is called Zionism. 8 C Gans, A Just Zionism: On the Morality of the Jewish State (Oxford, Oxford University Press, 2008).
Jewish and Democratic: Three Zionisms 475
will briefly present the proprietary, hierarchical and egalitarian interpretations of Zionism. In part III I will argue for the three theses just listed regarding the post-Zionist critique. II. THREE ZIONISMS
A. Proprietary Zionism According to proprietary Zionism, the Jewish people have a right of ownership as it were, over the Land of Israel and its political institutions, that is rooted in antiquity. This interpretation resonates in the Israeli Declaration of Independence, the first section of which states that [t]he Land of Israel [is] the birthplace of the Jewish people. Here their spiritual, religious and political identity was shaped. Here they first attained to statehood, created cultural values of national and universal significance and gave to the world the eternal Book of Books.9
To be sure, this section tells the (true) story of the primacy of the Land of Israel in Jewish history, emphasizing that it was from the Land of Israel that the Jews bequeathed the world with the Bible. However, Jewish Israelis adhering to mainstream Zionism and its politicians usually reverse what is stated in the first section of the Israeli Declaration of Independence.10 Instead of emphasizing what it actually asserts, namely, the primacy of the Land of Israel in Jewish history and identity, they usually stress the primacy of the Jews in the history of the Land of Israel. And instead of noting that the Jews had bequeathed the Bible to the world when they resided in the Land of Israel, they speak of the Bible as bequeathing the Land of Israel to the Jews. This reversal serves to justify their proprietary claims to the Land of Israel. The second and third sections in the Declaration of Independence encourage this transposition: After being forcibly exiled from their land, the people kept faith with it throughout their Dispersion and never ceased to pray and hope for their return to it and for the restoration in it of their political freedom. Impelled by this historic and traditional attachment, Jews strove in every successive generation to re-establish themselves in their ancient homeland.11
These statements in the Declaration of Independence, which according to historians associated with mainstream Zionism (such as Anita Shapira12 and Yisrael Bartal13) are false, are used in order to reject any possible claims on the limitation of the Jews’ ancient proprietary rights to the Land of Israel. According to the second and third paragraphs in the Declaration of Independence, the Jewish people were forcibly expelled from the Land of Israel, and from the time of their expulsion never ceased striving to return to it. Therefore, it is only their physical possession with the land that they have lost, not their right of ownership over it. Declaration of the Establishment of the State of Israel, 1 LSI 7 (1948). The major characteristic of mainstream Zionism is the attribution of privileged status to the Jews in Israel relative to all other groups living in the country. 11 Declaration (n 9). 12 A Shapira, ‘Review Essay: The Jewish-people deniers’ (2009) 28 Journal of Israeli History 63. 13 Israel Bartal, ‘Inventing the Invention’, Haaretz, 27 May 2008: www.haaretz.co.il/literature/1.1327315 (in Hebrew). 9
10
476 Chaim Gans The current interpretation of Zionism, which is allegedly justified by the ancient Jewish rights over the Land of Israel functions not only as a justification for the return of the Jews to the Land of Israel upon the inception of Zionism. It also justifies their ongoing superior status in it relative to the Arabs. If the Jewish people are the owners of the Land of Israel from time immemorial, and its ownership has not lapsed, then it follows from this interpretation of Zionism that the Arab nation is the usurper of the Land of Israel. It must be noted that this interpretation presupposes a collectivist moral ontology in which the basic subjects of political morality are nations and not individuals. The combination of proprietary justice and collectivist moral ontology underlying the current version of Zionism provides possible explanations for two of the most important components of mainstream Zionism regarding the status of Arabs in Israel. On the one hand, this combination provides an explanation for the consensus existing among most Jewish political parties in Israel, according to which any collective rights should not be granted to the Arabs in the Jewish state, at least not rights of a political and territorial nature that express recognition of the Arab collectivity as an historical collectivity with respect to Palestine/the Land of Israel. On the other hand, the combination of proprietary justice and collectivist moral ontology can provide an explanation of the fierce disputes that often arise in Israel over issues concerning the individual rights of the Arabs in the Jewish state. On the one hand, since according to proprietary Zionism, the Arabs as an historical collectivity must be perceived as the plunderers of the country, it is totally unthinkable – from the perspective of this interpretation – that they should be granted collective rights there. For granting them such rights would be tantamount to granting a thief rights to objects he had stolen on the mere grounds that they had been stolen by him. On the other hand, the fact that the Arabs as a collective and not as individuals could be considered as usurpers of the land of the Jews provides an explanation for why the question as to whether they deserve individual rights, or which individual rights they deserve, is under dispute among the different sections of mainstream Zionism. The limiting right-wing response to this question would be that of the racist right-wing Zionism inspired by Rabbi Meir Kahana that calls to remove all Arab individuals from the Land of Israel. The limiting ‘left-wing’ response of proprietary Zionism would be that of Revisionist Zionism inspired by Zeev Jabotinsky, the current main spokesman of which is Moshe Arens. It proposes the removal of the Arabs as a collective from the Land of Israel, while recognizing their status as individual citizens. This status grants them full civil and political rights. In addition, there are many other possibilities between these two poles. If members of the public at large and its politicians make proposals that express views to the left of Arens’s position, they usually do so for tactical reasons or at most for pragmatic reasons, and not for principled ones. Examples are Ben-Gurion’s acceptance of the Peel Partition Plan in 1937, or the concession Netanyahu made to Obama in his 2009 Bar-Ilan speech by agreeing to the establishment of a Palestinian state within the area dominated by Israel since 1967.14
14 Ben-Gurion was explicit in indicating that his reasons for supporting the Peel Partition Plan were tactical. See the excerpt from a letter he wrote to his son Amos in 1937. D Ben-Gurion, Letters to Paula (Pittsburgh, University of Pittsburgh Press, 1968) 153.
Jewish and Democratic: Three Zionisms 477
B. Hierarchical Zionism
In addition to the historical argument, the Israeli Declaration of Independence justifies itself by invoking two additional arguments pertaining to political morality – one is based on the persecution of the Jews and the other on the universal right to national selfdetermination.15 The latter is interpreted by some leading representatives of mainstream Zionism as entailing a right to a nation-state within which the Jews exercise hegemony. According to Ruth Gavison, the state declared in the Declaration of Independence ‘is the state in which the Jewish people realises its right to self-determination, or in other words: Israel is the nation-state of the Jewish people’.16 She further notes that a nationstate is a state in which there is a connection between the state institutions and a particular national culture, [one that] grants important advantages to the nation with which the state is identified and places a great burden on citizens of the state who are not members of this nationality.17
While the proprietary argument for a hierarchy between Jews and non-Jews in Israel is mainly popular among the public at large and many politicians representing mainstream Zionism, the argument based on the hegemonic interpretation of the universal right to self-determination is mainly invoked by academics from the disciplines relevant to politics and members of academic professions that serve mainstream policy.18 Their discussion of the status of the Jews compared to the status of the Arabs in the Land of Israel is mainly based on this argument. They treat the historical arguments invoked by the Declaration of Independence – namely, the argument based on the historical connection between the Jews and the Land of Israel and the argument concerning the history of the persecutions of the Jews – as secondary and as playing an auxiliary role in support of the self-determination based argument. The argument for hegemony based on self-determination is an a-historical argument. Unlike proprietary–historical rights, the right to national self-determination is a right that groups have by virtue of belonging to a general category to which other groups also belong – their being peoples – rather than by virtue of certain events in their particular history. The current argument for hegemony also differs from the historical argument in 15 The Declaration also contains legal arguments listing international declarations in support of the Zionist enterprise. These include the 1917 Balfour Declaration, the 1922 British Mandate for Palestine, and the UN General Assembly Partition Plan of 1947. 16 R Gavison, Israel as a Jewish and Democratic State: Tensions and Possibilities (Jerusalem, Van Leer Institute and Hakibbutz Hameuchad, 1999) 26. 17 R Gavison, ‘The Jewish State: The Principle Justification and the Desirable Character’ (2002) 13 Tekhelet 50, 54 (in Hebrew). 18 The most important among this latter group are mainly judges and jurists. Thus, we find the former Supreme Court Judge Menachem Elon asserting that ‘the state of Israel is the state of the Jewish People and only the Jewish people’ (EA 2/88 Ben-shalom v Central Elections Committee for the Twelfth Knesset PD 43(4) 221, 272 [1989] (in Hebrew)). Prominent examples of the former group are Shlomo Avineri, Ruth Gavison, Amnon Rubinstein and Alex Yaakobson. Avineri frequently publishes opinion editorials in the daily Haaretz which presuppose this conception of Jewish self-determination in Israel: S Avineri, ‘The Hope Will Not Die’, Haaretz, 20 October 1995; ‘Yes, Jewish and Democratic’, Haaretz, 3 November 2003: www.haaretz.co.il/ misc/1.921631 (in Hebrew). Gavison, Rubinstein and Yaakobson published full scale books and articles defending this interpretation: R Gavison, ‘The Jews’ Right to Statehood: A Defense’ (2003) 15 Azure 70, 74–75; A Yakobson and A Rubinstein, Israel and the Family of Nations: Jewish Nation-State and Human Rights (London, Routledge, 2008).
478 Chaim Gans that it does not derive from proprietary justice, which grants rights mainly on the basis of the legitimacy of acts of acquisition. Rather, it is rooted in distributive justice. According to distributive justice, resources must be allocated mainly on the basis of the principles of equal respect and concern for individuals and groups in need of the resources. There are two reasons why the moral implications of Jewish hegemony that is based on self-determination are much less terrifying than those of hegemony based on the proprietary–historical argument. First, hegemony justified by self-determination only has implications concerning the institutional dimension of Jewish rule in the Land of Israel, and does not have any implications regarding its territorial dimension. The hegemony justified by self-determination does not necessarily cover the entire territory of the Land of Israel, and therefore does not necessarily threaten all the Arabs (and other non-Jews) inhabiting the Land of Israel, but only those within the borders of the area in which the Jews enjoy self-determination. Secondly, this kind of hegemony cannot imply committing wrongs as horrendous as those entailed by the proprietary argument. This is so because unlike hegemony deriving from the notion of ownership, the hegemony entailed by the right to self-determination also includes many constraints and limitations which are usually related to the exercise of governmental powers, such as the constitutional constraints of human and civil rights. Basing Jewish hegemony in the Land of Israel on the right to national self-determination does not easily lend itself to allowing the expulsion of Arabs from the Land of Israel, violating their rights to private property, making their political rights conditional upon their loyalty to the Jews, and similar types of gross wrongs, some of which are actually carried out by Israel and others proposed time and again by political parties in Israel. From the viewpoint of the proprietary–historical argument for the Jews’ rights in the Land of Israel, the aforementioned acts are not even considered wrongs. If lands in the Land of Israel belong to the Jews, taking it away from Arabs means redeeming it, not plundering it. On the other hand, from the viewpoint of those who believe in hegemony, because they interpret the right to self-determination as the right to political hegemony, most acts of these types are considered unthinkable. Yet, hierarchical Zionists are convinced that certain inequalities regarding the Arabs derive from the right to self-determination. Their view is that this right means the right to a Jewish nation-state, a state identified as solely Jewish in its public sphere, its symbols, language, anthem, the national/religious affiliation of those immigrants allowed to enter the country and become citizens, ‘and the right to the monopoly [that the Jewish people has] over the entire public and symbolic space of the state’.19 Thus, even though hegemony entailed by self-determination cannot approach the intensity of the oppression which is implied by proprietary Zionism, that justification, too, condemns Israel to upholding a regime of inequality between Jews and Arabs, by excluding the latter – as a matter of constitutional principle – from presence in the public, symbolic, and essentially the political spheres of Israel, at least as a collective.20
Gavison, ‘The Jewish State’ (n 17) 28. I have critiqued at length the arguments supporting the current conception of self-determination in general, and specifically, the arguments of its Israeli spokesmen in Gans, The Limits of Nationalism (n 1) 70–83; Gans, A Just Zionism (n 8) 53–80. 19 20
Jewish and Democratic: Three Zionisms 479
C. Egalitarian Zionism This Zionism is the one which I argued for in A Just Zionism. Like proprietary Zionism and hierarchical Zionism, it is based on the three arguments for the establishment of the state of Israel that are cited in its Declaration of Independence: (a) the historical link of the Jews to the Land of Israel; (b) the right to national self-determination; and (c) the Jewish history of persecution. Yet, the roles and meanings of each of these arguments within the framework of egalitarian Zionism differ from those in the other versions of Zionism. Unlike proprietary Zionism, but as in the framework of hierarchical Zionism, the right to national self-determination – rather than the historical link between the Jews and the Land of Israel – constitutes the core of the egalitarian interpretation. Unlike hierarchical Zionism, however, egalitarian Zionism does not conceive of the right to national self-determination as implying a hierarchy between homeland groups. The point of departure of the right to national self-determination is equality between homeland nations within the framework of the states ruling their homeland territories. Any digression from this equality can only be justified if there are reasons which themselves derive from the requirements of equality, for example, due to differences in the population size of each of the homeland groups in the state that rules over areas of this homeland, or due to special needs of a group compared to other group/s. Egalitarian Zionism interprets the historical ties between the Jews and the Land of Israel in the way they are actually described in the Declaration of Independence – as pertaining to the primacy of the Land of Israel in Jewish identity rather than to the primacy of the Jewish people in the history of the Land of Israel. The latter interpretation is the one commonly held among those supporting proprietary Zionism. Moreover, within the framework of egalitarian Zionism, the historical ties constitute a basis just for determining the geographical site for the realization of the universal and a-historical right to self-determination. They do not constitute a basis for the Jews’ proprietary rights over the whole Land of Israel. As for the persecutions of the Jews: they serve within egalitarian Zionism as the basis for the realization of Jewish self-determination in the Land of Israel despite the fact that Arabs were already living there. They are interpreted as a source for a necessity which justified – or at least can serve as an excuse for – the establishment of a Jewish colony in Palestine. As such, they set the limits to the price the Arabs have been forced to pay for the realization of Zionism, namely, giving up territories which they inhabited and over which they could have had exclusive control. The necessity which these persecutions created is part of the justification for the realization of the Jewish right to self- determination in Palestine. These persecutions, which the Jews suffered over the centuries, did not end despite their emancipation in the nineteenth century, and culminated in the Holocaust. However, the urgent necessity that these persecutions created by the 1930s and 1940s, gradually diminished by the late 1940s with the establishment of the State of Israel, and June 1967, when the state achieved a stunning victory in the Six-Day War. The expansion of Jewish settlement in the Land of Israel after that year, and beyond the borders determined between 1949 and 1967, cannot be justified in the absence of such extensive persecution of the Jews. By perceiving the historical connection between the Jews and the Land of Israel as the basis only for determining the geographical site for the realization of the Jews’ universal
480 Chaim Gans right to self-determination within the Land of Israel, and not as a basis for their proprietary rights over the Land of Israel as a whole, the egalitarian conception of Zionism avoids the potentially disastrous moral implications entailed by the proprietary conception of Zionism. By perceiving the persecution of the Jews as the basis for the realization of Jewish self-determination in the Land of Israel despite the fact that Arabs were already living there, egalitarian Zionism sets limits, as I have just explained, to the price the Arabs have been forced to pay. However, the core of egalitarian Zionism consists in its interpretation of the right to self-determination. According to its egalitarian interpretation, this right should be viewed as consisting of self-government and cultural preservation rights that should be granted equally to all homeland groups within the states dominating their homelands. It is only when equality itself requires differences in rights allocated to homeland groups as a result of relevant non-normative differences between them (such as their size, or their needs), that such different rights should be granted. Equality, as we know, does not mean equal arithmetic division of goods but rather a division which demonstrates equal concern and respect for the parties involved. Granting equal child benefits to families means granting more benefits to families with many children than to families with few children. Similarly, in a state in which there are two homeland nations, if one of them has many members while the other is smaller, it is fitting that the larger group enjoy more resources for educating its younger generation, more massive presence in the institutions of civil society, politics and the economy of the state, more of a presence in the public sphere, and larger representation in the symbols of the state. Yet, according to the egalitarian conception, there is no justification for the group with a larger number of members – as supporters of hierarchical Zionism think – having ‘a right to a monopoly over all the public and symbolic space of the state’ (emphasis added).21 As we know, it is justified not to distribute goods in arithmetically equal portions not only because of differences in size among the relevant groups, but because of differences in needs that justify such deviations. Equality in distribution of medicines does not mean that an equal amount of medicines is distributed among the ill and the healthy, but rather the prioritizing of ill people who need medicines over healthy individuals who do not require them. On Indian reservations in North America, various special privileges are granted to Native Americans over other citizens of these states owing to the needs of the former for the special protection of their cultural interests on their reservations. Perhaps the Jewish right to self-determination in the Land of Israel also requires such protection. I believe that the Jewish right to self-determination in the Land of Israel does indeed require such protection owing to the history of the Jewish–Arab conflict as well as due to the fact that the Jews are a small minority in the Middle East. However, one must be aware that justifications such as these for deviations from the arithmetically equal distribution of rights and goods must be limited to those areas to which these special needs apply. In the case of the Jews in Palestine, they only apply to the spheres of security and demography, and do not comprise all other issues. Moreover, the priorities, special rights and other means which are granted must be constrained by the requirements of human rights. For example, preserving the Jewish demographic majority in Israel by prohibiting Palestinian family unification would be unacceptable, because it violates Article 1 of the International Convention on the Elimination of All Forms of See Gavison, ‘The Jewish State’ (n 17) 28.
21
Jewish and Democratic: Three Zionisms 481
Racial Discrimination.22 According to this Article, the term ‘racial discrimination’ applies to any distinction, exclusion, restriction or preference based on race . . . or national or ethnic origin which has the purpose or effect of nullifying or impairing the recognition, enjoyment or exercise, on an equal footing, of human rights and fundamental freedoms (emphasis added).23
III. THE POST-ZIONIST CRITIQUE
A. Jewish Self-Determination and Equality When the post-Zionist writers argue that Israel’s Jewishness conflicts with its claim to being a democratic state, they do not mean to argue that the principle of majority rule is being violated in Israel, or that its citizens have not been granted rights such as the right to vote or to be elected, the right to assembly, or the right to free speech. What they do mean to say is that in the case of the Arab citizens of Israel, these rights have to a great extent been emasculated, while Jewish citizens enjoy immunity from any Arab intervention in their pursuit of their interests as Jews. The Arabs in Israel have the right to vote and to be elected, but they are constitutionally prevented from advancing their collective interests as Arabs, since they cannot vote and be elected for the purpose of promoting their group interests as a homeland ethno-cultural group in Israel. Only the Jews in Israel have this right.24 The alleged conflict between the Jewishness of the state and democracy thus refers to the unequal distribution of political power between Jews and Arabs. This critique, if perceived as a critique of mainstream Zionism, is simultaneously too feeble and too strong with regard to this kind of Zionism. It is too weak to the extent that it is targeted at the proprietary version of mainstream Zionism. Yet, it is too strong in relation to the hierarchical conception of the right to self-determination. The post-Zionist critique is too feeble in relation to proprietary Zionism, since this Zionism condemns Israel not only to inequality between its Jewish and Arab citizens, but also condemns it to the perpetual violation of the human rights of its Arab citizens. In his book The Law of Peoples, John Rawls distinguishes among various types of peoples regarding their moral perfection. At the top of the ladder, he positions ‘liberal’ peoples – those who maintain democracy and equality among their members. After them, he ranks peoples that he calls ‘decent’ – the type that does not maintain democracy and equality but instead has a hierarchy of rights pertaining to different groups and communities. However, these peoples at least protect the human rights of those under their rule. Rawls’s third category, which is important for our discussion, is that of ‘outlaw’ states – those states that threaten peace by attempting to expand their spheres of influence and by their violation of the basic human rights of the people inhabiting their territories.25 22 International Convention on the Elimination of All Forms of Racial Discrimination (adopted 21 December 1965, entered into force 4 January 1969) 660 UNTS 195. 23 For further discussion, see C Gans, ‘Individual’s Interest in the Preservation of Their Culture: Its Meaning, Justification and Implications’ (2007) 1 Journal of Law and Ethics of Human Rights 6. 24 This is especially due to the Basic Law: The Knesset, s 7A. This section allows the disqualification of political parties that have a racist platform and of political parties that deny Israel’s right to exist as the state of the Jewish people. This section of the law has been amended over the years and today requires the disqualification of parties which deny Israel’s right to exist as a Jewish and democratic state. 25 J Rawls, The Law of Peoples (Cambridge, Harvard University Press, 1999) 23–25, 63–66, 80–81.
482 Chaim Gans If what I said above about the role played by proprietary Zionism in Israeli politics is correct, Israel should not only be categorised as a country that is neither liberal nor egalitarian, but as a state that is not even ‘decent’. If the theory underlying Israeli politics is historical–proprietary Zionism – and in view of the settlement policy, there is no other way to interpret the theory underlying current Israeli politics – it seems that Rawls’s third category, not necessarily the second, is a more apt characterization of present-day Israel. To make do with the post-Zionist critique that there is a contradiction between Israel’s Jewishness and its democratic nature in a way that turns it into a non-egalitarian society is to voice too feeble a claim and to miss criticizing what most needs to be criticised. Conversely, if the post-Zionist critique is aimed at the hierarchical interpretation of Zionism, the type that relies on the hegemonic conception of self-determination, this critique is exaggerated. Although its premises are well-founded, the practical conclusion which the post-Zionist writers draw from these premises does not follow. Israel, according to the conception held by Ruth Gavison, Amnon Rubinstein, Alex Jacobson, and others, is by their own admission a non-egalitarian, hierarchical society, a country that ‘grants important advantages to the nation with which the state is identified and places a great burden on citizens of the state who are not members of this nationality’.26 In this chapter I have clarified why such inequality cannot be justified. However, the practical conclusion that the post-Zionist writers draw from this critique, namely, that Jewish self-determination in Israel should be annulled, does not follow from the condition of inequality implied by the hegemonic conception of self-determination. What does follow is that the hegemonic conception of self-determination must be replaced by its egalitarian conception. According to this conception, since both Jews and Arabs in Israel are homeland groups, they should both, as a matter of principle, be granted hegemony compared to communities of immigrants but not compared to each other. On the other hand, for circumstantial considerations the Land of Israel must be divided between Jews and Arabs into two states, one that is mainly Jewish while the other is principally Palestinian. Jewish dominance in one of these states, and Arab dominance in the other, would be justified because of numbers and special needs, which are equality-based reasons for inequality, and not because there is something about the principle of selfdetermination which makes it hierarchical as a matter of principle. B. Jewish Self-Determination and Freedom In actuality, and despite the opposite impression they create, post-Zionist writers are aware of the fact that the inequality forged by the hegemonic interpretation of the right to self-determination does not entail the need to totally remove the Jewish presence from the constitutional level of the state. In response to the possibility of a bi-national state, Uri Ram says that one should not confuse between the current situation, in which there might be a justification for demanding national equality, and the ideal state of affairs from the post-Zionist perspective, in which nationality will have no constitutional or statist standing. It may be that at the present state of affairs, in a country that is bi-national by the composition of its population, bi-national governance would Gavison, ‘The Jewish State’ (n 17) 26.
26
Jewish and Democratic: Three Zionisms 483
be more democratic than mono-national governance. Yet, there is no need to promote a solution that does none other than double by two the mono-nationality and in fact perpetuate the suppressive nationalist structures of control each by itself and in contrast to the other.27
In other words, the problem of inequality is a problem that could be solved by granting constitutional status not only to the Jewish group but also to the Arab. Nevertheless, according to Ram, this still leaves the problem of freedom. He believes that ethno- cultural governmental structures are ‘oppressive structures’ in relation to the individuals under their rule, even if they maintain equality between different national groups. The fact that Ram is concerned with impeding freedom rather than equality is underscored in many of the statements he makes in his discussion of the Israeli Supreme Court (sitting as the High Court of Justice) decision Ornan v Minister of Interior.28 In this case, the Court considered a petition by many Israeli celebrities to instruct the state to register them as Israeli under the rubric Nationality on their identity cards rather than as Jewish. Ram describes this petition as expressing the post-Zionist conception that seeks to replace ethnic nationality (the EasternEuropean and German models), according to which Israel is the state of the Jewish people, with civic nationality (the French–American model), according to which Israel is a state of its citizens (and inhabitants) and only theirs.29
When this petition seeks ‘to ground the principle of nationality on the principle of citizenship’, says Ram, ‘it is endeavoring to apply to national identification the basic principles of democracy – personal choice (and not state coercion) and recognition of a range of alternatives (and not coercion of given possibilities)’.30 That is, the problem in the Jewishness of Israel is not simply in its Jewishness, but in this Jewishness being the realization of ethno-cultural nationalism, a nationalism that will remain a problem even if Israel would cease to be mono-ethno-nationalist and would become bi-ethno-nationalist. ‘The bi-national principle sanctifies nationalism instead of proposing a worthy alternative to it’ – Ram says.31 ‘From the democratic point of view’, he adds, ‘it is fitting for a state to represent analytically abstract citizens rather than cultural communities’.32 But is such a state possible? And will such a state actually serve the freedom of its citizens? These questions took centre stage in the debate between the major political theories of neutralist liberalism that aroused great interest from the 1970s onwards, in particular John Rawls’ theory of justice, the non-liberal responses to neutralist liberalism called communitarianism, as well as the intra-liberal responses to them, namely, those that have been called liberal multiculturalism and liberal nationalism. This debate is an elaborate replica of the famous clash between enlightenment and anti-enlightenment thinkers that preceded it by 200 years. However, the main novelty in the current debate is that arguments against neutralist liberalism are being invoked not only by the opponents of liberalism, but also by writers within the liberal camp: liberal multiculturalists and liberal nationalists. Their arguments – expressed in thousands of articles and hundreds of books that have appeared since the 1980s – are scarcely mentioned by Ram. Yet, these Ram, The Time of the ‘Post’ (n 3) 191. HCJ 11286/03 Ornan v the Minister of Interior (20 September 2004), Nevo Legal Database (by subscription) (in Hebrew). 29 Ram, The Time of the ‘Post’ (n 3) 196. 30 ibid 200. 31 ibid 91. 32 ibid. 27 28
484 Chaim Gans arguments justify negative replies to the two questions at the beginning of this paragraph. Even if ‘it is fitting for a state to represent analytically abstract citizens rather than cultural communities’, such a state is humanly not really feasible. The alternatives that one might consider as approximating it mostly would not serve human freedom, at least no more than states that serve citizens belonging to cultural communities. A number of writers have already noted that states cannot be neutral with respect to their support of culture in the same way as they can perhaps be neutral with respect to religion. The reason is that it is impossible to avoid giving preference to a particular language or particular languages spoken by their citizens. Preferring certain languages over others is an unavoidable practical necessity. As a result, in a multi-ethnic and multinational state, the state cannot be neutral with regard to its citizens’ interests in adhering to their original languages and cultures. In actual practice, the state’s choice of one particular language over other languages means demonstrating preference for the cultural group that speaks that language. The fact that their language has been chosen makes it possible for members of that group to adhere to their culture and to sustain it for generations, while the same possibilities are denied to other citizens. On the practical level, there is no way out of this predicament. The ideal that states should be neutral in the cultural sense is therefore necessarily unattainable, and it has not and cannot be implemented by any state. In this context, Will Kymlicka noted the language policy implemented by the United States, which is ‘the allegedly prototypically “neutral” state’. In the United States, there is a legal requirement for children to learn English in schools. Knowledge of English is a condition imposed on immigrants for receiving citizenship, and it is also a condition for employment in government. Kymlicka further noted that the borders of the states in the United States and the dates on which new states joined the Union were intentionally determined in a manner that would ensure an English-speaking majority in these states. According to him, these requirements and decisions ‘have played a pivotal role in determining which ethno-linguist groups prosper and which ones diminish’.33 As for France, the situation is even more striking: the aspiration to preserve and promote French culture is manifestly a part of the political agenda of this civic nation-state; it is an aspiration that guides its political and budgetary decisions. France is a country that enacts regulations to protect its singers’ chansons by placing quotas on the broadcasting of foreign music by its radio stations. Similarly, it protects its film-makers by providing them with massive subsidies. France does this for cultural reasons, not economic reasons. France could relinquish this cultural protectionism, but it could not refrain from adopting a language, or a small number of languages as its language/s. This means that cultural neutrality is a possibility that countries cannot really opt for. Moreover, states whose citizens are divested of their culture and which allow human freedom to be realised in a manner disconnected from actual concrete cultures are not only impossible – they are also undesirable. The closest approximation of the ideal of the cultural neutrality of states that one might realistically think of is the adoption of a single language and culture by all the countries in the world. If all humankind had one common culture, inequality among people belonging to different cultures within the 33 W Kymlicka, ‘Western Political Theory and Ethnic Relations in Eastern Europe’ in W Kymlicka and M Opalski (eds), Can Liberal Pluralism be Exported?: Western Political Theory and Ethnic Relations in Eastern Europe (Oxford, Oxford University Press, 2001) 13, 17.
Jewish and Democratic: Three Zionisms 485
framework of the various existing states would not arise, and people would then not be limited culturally from moving from one place to another the world over. Yet, obviously such a reality, which in certain respects would enhance the freedom of individuals, would diminish it in other respects: first and foremost – relative to the prevailing state of affairs today – it would reduce the range of options open to people regarding the choice of ways of life and lifestyles. Yet, beyond that, creating a reality in which there is only one culture and language in the world would block peoples’ freedom to live within a framework of cultures with which they identify and within which they have a sense of historical belonging. The interest that masses of people everywhere currently have to cling to their original culture as well as their interest in maintaining it for generations, should they wish to do so, is for these people an interest at least as important as the interest of people belonging to a particular gender to continue to belong to that gender, if they wish to do so, or the interest of people with this or that sexual orientation to continue living according to it, should they wish to do so. These are pivotal human com ponents of identity. For those persons for whom these components of identity are important, politics and law must maintain political and legal realities that allow them to continue to cling to these components of their identity. The existence of political and legal realities that do not allow people to adhere to identity and that have decisive influence on their welfare means the oppression of those people, in effect, a violation of their freedom, dignity, and their welfare. Just as no political or legal reality should be maintained that threatens the ability of people whose sexual orientation is homosexual to realise this orientation, there should be no political or legal reality that threatens the ability of people whose cultural identity is Jewish national to realise that identity. Relinquishing Jewish self-rule in the Land of Israel – and for reasons I listed earlier, even relinquishing Jewish majority – will pose a threat to the ability of all those masses of Jews who are interested in preserving their national identity to realise this interest. It seems to me, therefore, that not only would the post-Zionist proposal to waive this constitutional protection not promote freedom, but it would even critically violate the freedom of many Jews. Of course, this is not to be understood as saying that any legal means is appropriate for advancing the majoritarian aim. In the previous section, I mentioned several means which to my mind are inappropriate. Moreover, I am convinced that refusing to register citizens who wish to be listed as Israelis as their nationality and not Jewish is indeed an improper legal means to preserving Jewish identity on the political and constitutional level in Israel. Such means harms freedom. But as I have mentioned a number of times, one must not confuse the issue of the justness of the goal of maintaining Jewish self-determination in Israel with the issue of the justness of one means or another for promoting this goal. The justness of the goal does not mean that all means proposed to advance it are just, and the impermissibility of this or that means for the advancement of a given goal does not mean that the goal is unjustified. I therefore return to the thesis I voiced at the beginning of current part of this chapter. The post-Zionist critique of Zionism is simultaneously too feeble and too strong. It is too weak to the extent that it is targeted at the proprietary version of mainstream Zionism, whereas it is too strong in relation to the interpretation of Zionism by mainstream academics and professionals, namely, the one based on the hierarchical conception of the right to self-determination. Zionism need not be abandoned because of the post-Zionist critique because there is a version of Zionism, its egalitarian version, which cannot be subjected to the post-Zionist criticisms.
32 A Jewish Nation-State: A Discussion in Light of the Family Reunification Case AVIAD BAKSHI AND GIDEON SAPIR
I. INTRODUCTION
T
HE ISRAELI DECLARATION of Independence defines Israel as a ‘Jewish state’.1 The Israeli Supreme Court determined that the Jewishness of the state ‘is a constitutional fact which lies at the base of the activity of the state authorities’.2 Israeli law has internalised this basic constitutional element in many areas. For example, Israel’s immigration laws grant the right to nearly every Jew to become its citizen almost immediately.3 For over 50 years the State of Israel customarily allotted land for settlement intended solely for Jews.4 Israel anchors commemoration of the Holocaust in legislation,5 cares for reparations for its victims6 and for bringing Nazis to trial for their crimes,7 and applies the Penal Law of the State of Israel to crimes ‘against a Jew for being a Jew,’ even if perpetrated outside its territory and against people who are not citizens of Israel.8 Over the course of years, the Israeli Supreme Court invalidated several arrangements of this type. In the past decades, the Court intensified that trend,9 while at the same time reemphasizing that Israel’s very definition as a Jewish state is legitimate from the democratic Declaration of the Establishment of the State of Israel, 1 LSI 7 (1948). EA 1/65 Yardor v Chairman of the Central Elections Committee to the Sixth Knesset 19(3) PD 365 [1965] (in Hebrew). 3 Law of Return, 5710-1950, s 1 (hereinafter: Law of Return); The Nationality Law, 5712-1952, s 2. 4 HCJ 114/78 Burkan v Minister of Finance 32(2) PD 800 [1978] (in Hebrew). For academic review of this policy, see, eg Y Benkler, ‘Non-discrimination in Housing: Content, Scope, and Assistance’ (1991) 17 Tel-Aviv University Law Review 131 (in Hebrew). 5 Martyrs’ and Heroes’ Commemoration Law – Yad Vashem, 5713-1953; Martyrs’ and Heroes’ Remembrance Day Law, 5719-1959; Denial of Holocaust (Prohibition) Law, 5746-1986; State Education Law (Amendment No 5), 5760-2000, s 2(4). 6 Nazi Persecution Disabled Persons Law, 5717-1957; Victims of Holocaust Property Law (Restoration to Successors and Dedication to Support and Perpetuation), 5765-2005. 7 Nazis and Nazi Collaborators (Punishment) Law, 5710-1950; CC (Jm) 40/61 Attorney General v Eichmann 27 PM 169 [1961] (in Hebrew); CrimA 336/61 Eichmann v Attorney General 17 PD 2033 [1962] (in Hebrew); CrimA 347/88 Demjanjuk v State of Israel 47(4) PD 221 [1988] (in Hebrew). 8 Penal Law, 5737-1977, s 13(b)(2). 9 Thus the Court gradually eroded the senior status of Hebrew versus Arabic and other languages. See CA 294/91 Jerusalem Community Burial Society v Kestenbaum 46(2) PD 464 [1992] (in Hebrew); CA 105/92 Reem Engineers Contractors v Nazareth-Illit 47(5) PD 189 [1993] (in Hebrew); CA 12/99 Jamal v Sabek 53(2) PD 128 [1999] (in Hebrew); HCJ 4112/99 Adalah v Tel-Aviv Municipality 56(5) PD 393 [2002] (in Hebrew). 1 2
488 Aviad Bakshi and Gideon Sapir point of view.10 This position of the Court is attacked from two opposing directions. There are those who argue that the Court’s policy turns the Jewishness of Israel into a defunct symbol.11 In contrast, others claim that the Court’s steadfastly clinging to the definition of the State of Israel as a Jewish national state, as poor as its content may be, conflicts with basic liberal foundations.12 This chapter has two aims. The first is to reinforce the criticism against the trend toward limiting the components of national identity in Israel. We shall argue, basing ourselves on liberal premises, that the Jews have the human right to live in their own nation-state. Recognition of the right for self-determination in a nation-state does not make it immune from limitation. Like any other human right, this one too is not absolute, and on certain issues it must recede in the light of other rights or reach a compromise with them. Yet, from the moment we recognise the constitutional status of the right, we must ensure that it will not remain an empty slogan. Our second aim is to examine what is the proper weight to be given to the afore mentioned right within the framework of the Israeli immigration policy. We shall discuss the Citizenship and Entry into Israel Law (Temporary Provision)13 (hereinafter: Temporary Provision Law) that limits the entrance to Israel and the acquisition of status by residents of the Palestinian Authority, and two Supreme Court rulings that examined the constitutionality of this Law. Two short clarifications are called for: 1. The concept ‘Jewish’ in the term ‘Jewish state’ may be interpreted in two ways: religious and national. There are those who feel that the State of Israel must act not only to effectuate the national yearnings of its Jewish citizens but also to realise the objectives of the Jewish religion.14 In contrast, there are some who argue that the Jewish nation has an independent existence separate from the Jewish religion and that the State of Israel as a Jewish state is obliged only to the former.15 This chapter does not enter the thicket of this debate and will focus solely on the national aspect, on the assumption that it is possible to differentiate between the two meanings. Thus, our discussion will focus on questions of relations between a state and a nation and not issues of religion and state. 10 See, eg EA 11280/02 Central Elections Committee to the Sixteenth Knesset v Tibi 57(4) PD 1 [2003] (in Hebrew); EA 2/88 Ben-Shalom v Central Elections Committee to the Twelfth Knesset 42(4) PD 749 [1988] (in Hebrew); HCJ 6698/95 Ka’adan v Israel Land Administration 54(1) PD 258 [2000] (in Hebrew); Kestenbaum (n 9). 11 See, eg U Struzman, ‘High Court of Justice: The Concept, the Jewish Agency, and the Settlement Katzir’ (2000) 140 Ha-Umma 14 (in Hebrew); E Shochetman, ‘The Legality and Constitutionality of Jewish Settlement in Erez Israel’ (2001) 6 Law and Government 109 (in Hebrew); Y Cohen, Who’s Afraid of a Jewish State (TelAviv, The Israel Bar-Publishing House, 2001) (in Hebrew). 12 See, eg A Bisharah, ‘Between People and Nation: Thoughts on Nationality’ (1995) 6 Theory and Criticism 19 (in Hebrew); Y Peled, ‘Ethnic Democracy and the Legal Construction of Citizenship: Arab Citizens of the Jewish State’ (1992) 86 American Political Science Review 432; O Yiftachel, ‘Ethnocracy: The Politics of Judaizing Israel/Palestine’ (1999) 6 Constellations 364; B Kimmerling, ‘Religion, Nationalism and Democracy in Israel’ (1999) 6 Constellations 339; A Shlaim, Collusion Across the Jordan: King Abdullah, the Zionist Movement, and the Partition of Palestine (New York, Columbia University Press,1988). 13 Citizenship and Entry into Israel Law (Temporary Provision), 5763-2003, SH No 1901, p 544. 14 See, eg Rabbi YI Herzog, A Constitution for Israel According to Tora (Jerusalem, Mosad HaRav Kook, 1989) (in Hebrew). 15 See, eg M Kremnitzer, ‘The Image of the State of Israel as a Jewish and Democratic State’ in A Ravitsky and Y Stern (eds), About the Jewishness of a Democratic State (Jerusalem, The Israel Democracy Institute, 2007) 395 )in Hebrew).
A Jewish Nation-State 489
2. The concept ‘nation’ in the expression ‘nation-state’ can also be interpreted in two ways: politically and ethno-culturally.16 Political nationalism is defined solely in terms of citizenship. In contrast, ethno-cultural nationalism is based on a common culture, on a narrative of common ethnic origin, or on another component of identity that is not dependent upon common citizenship. Our discussion will focus on the second meaning of the concept, that is, on nationalism in the ethno-cultural sense. II. A JEWISH NATIONAL STATE AS A HUMAN RIGHT
A. Justifications from the Right to Culture In recent decades many liberal thinkers have begun to suggest liberal justifications for the right to culture. The various proposals can be usefully divided into four main formulations: Kymlicka holds that life within the framework of a certain culture serves as an imperative condition for affecting the right to autonomy.17 Taylor feels that a person’s right to preserve his specific culture derives from the right to dignity. In his opinion, a person’s human need to be respected by his environment deserves recognition.18 In this framework the uniqueness of each person, which is expressed through his culture, must be honoured.19 Halbertal and Margalit are convinced that a person’s culture constitutes a fundamental component in his self-identity, and therefore the right to culture is a derivative of the right to identity.20 Finally, Gans feels that each person has the right to act for the realization of significant dreams and considers a person’s cultural choices as the effectuation of such a dream.21 In common with all the approaches described is the conclusion that the right of human beings to live within the framework of their culture should be recognised. Does the right to culture draw in its wake a right to self-determination in a nation-state? Two arguments are likely to lead to a negative conclusion. First, it can be claimed that it is possible to realise the right to culture to a sufficient degree in sub-state community 16 See, eg C Gans, The Limits of Nationalism (Cambridge, Cambridge University Press, 2003) ch 1. Gans calls the first type of nationalism ‘statist nationalism’ and the second, ‘cultural nationalism’. Since in the continuation of this chapter we shall also indicate justifications for a nation-state that do not rely on culture, we prefer to use the term ‘ethno-cultural nationalism’, which is accepted in research into nationalism. 17 W Kymlicka, Liberalism, Community, and Culture (Oxford, Clarendon Press, 1989) ch 8; W Kymlicka, Multicultural Citizenship: A Liberal Theory of Minority Rights (Oxford, Oxford University Press, 1995) ch 2. For a similar view, see J Raz, ‘Multiculturalism: A Liberal Perspective’ in J Raz (ed), Ethics in the Public Domain: Essays in the Morality of Law and Politics (Oxford, Clarendon Press, 1994) 155, 170. 18 C Taylor, ‘The Politics of Recognition’ in A Gutman (ed), Multiculturalism and ‘The Politics of Recognition’ (Mexico City, Fondo de Cultura Económica, 1992) 25. 19 Taylor ibid; C Taylor, Sources of the Self: The Making of the Modern Identity (Cambridge, Harvard University Press, 1989) chs 1, 2; C Taylor, ‘Reply and Re-articulation’ in J Tully (ed), Philosophy in an Age of Pluralism: The Philosophy of Charles Taylor in Question (Cambridge, Cambridge University Press, 1994) 213. Of note is that Taylor already began to develop the idea of identifying culture with dignity of man in his early teachings, even before the right to culture was formulated as such. See C Taylor, ‘What is Human Agency?’ in T Mischel (ed), The Self: Psychological and Philosophical Issues (Lanham, Rowman & Littlefield Publishers, 1977) 103. 20 A Margalit and M Halbertal, ‘Liberalism and the Right to Culture’ (1994) 61 Social Research 491. 21 C Gans, A Just Zionism: On the Morality of the Jewish State (Oxford, Oxford University Press, 2008) ch 3; Gans, The Limits of Nationalism (n 16) ch 2; C Gans, From Richard Wagner to the Palestinian Right of Return (Tel-Aviv, Am Oved Publishers, 2006) ch 5 (in Hebrew); C Gans, ‘Nationalism and Immigration’ in M Mautner, A Sagi and R Shamir (eds), Multiculturalism in a Democratic and Jewish State (Tel-Aviv, Ramot Publishing House, 1998) 341, 346–48 (in Hebrew).
490 Aviad Bakshi and Gideon Sapir frameworks, and if so, there is no justification to demand a nation-state. Second, effectuating the right to culture of a national group within the framework of a nation-state might entail infringing on the right of cultural minorities in that state, including those minorities’ right to culture, so its realization should be limited only to sub-state frameworks. No wonder then that a large part of the liberal thinkers who support the right to culture feel it should be realised as part of a multicultural state and not a nation-state.22 Yet, other liberal writers, such as Michael Walzer,23 Charles Taylor,24 and David Miller,25 feel that realizing the right to culture requires recognition of the right to a nation-state. The selection of a nation-state as a fitting framework for realizing the right to culture can be based on the claim that a nation-state grants the national culture protection to an extent and power that cannot be compared to the protection that can be given by a multicultural state. For example, the efforts of a cultural group to preserve its culture and bequeath it to the coming generations, as part of a multicultural state, are liable to fail if the environs of that culture will be inundated by members of a foreign culture. Conversely, a nation-state enables the cultural group to operate an immigration policy that will coincide with the interest in preserving the national culture. Many cultures include, among other things, common values. A certain culture might prefer personal freedom over solidarity and distributive justice. Another might stand firm for compromises and turning the other cheek to its adversaries. The values can be realised within a framework of internal policy (in the first instance) and external policy (in the second) that would be adopted in a nation-state. Members of the relevant cultural group would, indeed, be able to attempt to fashion the policy of the regime in accordance with their national values in a multicultural state as well, but the chances for their success are conditional upon their relative weight in the state and even if they gain the majority, they will be forced to compromise with the value system of other culture groups.
22 See, eg Kymlicka, Liberalism (n 17) chs 7, 9; Kymlicka, Multicultural Citizenship (n 17) chs 4, 9; J Raz, ‘Multiculturalism’(n 17) chs 6, 8; SR Perry, ‘Immigration, Justice and Culture’ in WF Schwartz (ed), Justice in Immigration (Cambridge, Cambridge University Press 1995) 94. It should be noted that in his later theory, Kymlicka presents an approach that considers the existence of nation-states as a legitimate arrangement. As opposed to Taylor and Walzer, Kymlicka does not feel that nation-states are necessary for the purpose of protecting the right to culture, but he no longer sees them as a contradiction to the idea of the right to culture. In any event Kymlicka does not provide a detailed explanation for his new approach and does not even tenaciously cling to it. For example in his book, W Kymlicka, Multicultural Odysseys (Oxford, Oxford University Press, 2007), Kymlicka does not indeed have recourse to his opposition to a nation-state but he also writes that this is not the optimal regime. 23 See M Walzer, ‘Comment’ in A Gutmann (ed), Multiculturalism: Examining the Politics of Recognition (Princeton, Princeton University Press, 1994) 99; M Walzer, ‘The National Question Revisited’ in Nation and Universe: The Tanner Lectures on Human Values (Salt Lake City, University of Utah Press, 1990) 507, 532; M Walzer, ‘The Politics of Difference: Statehood and Toleration in a Multicultural World’ in R McKim and J McMahan (eds), The Morality of Nationalism (Oxford, Oxford University Press, 1997) 245; for Walzer’s earlier support of a nation-state, before the right to culture crystallized, see M Walzer, ‘The Moral Standing of States: A Response to Four Critics’ (1980) 9 Philosophy and Public Affairs 209. 24 C Taylor, ‘Why Do Nations Have to Become States?’ in G Laforest (ed), Reconciling the Solitudes: Essays on Canadian Federalism and Nationalism (Montreal, McGill-Queen’s University Press, 1993) 40; C Taylor, ‘Nationalism and Modernity’ in R McKim and J McMahan (eds), The Morality of Nationalism (Oxford, Oxford University Press, 1997) 31; C Taylor, ‘Cross-Purposes: The Liberal-Communitarian Debate’ in NL Rosenblum (ed), Liberalism and the Moral Life (Cambridge, Harvard University Press, 1989) 159. The statements are based on the principles delineated by Taylor in his main writings. See Taylor, ‘The Politics of Recognition’ (n 18); C Taylor, ‘Sources of the Self ’ (n 19) 196–203. 25 DL Miller, On Nationality (Oxford, Clarendon Press, 1995) ch 2.
A Jewish Nation-State 491
Another area in which a nation-state has an advantage over a multicultural state, concerns the attitude toward the cultural–national dispersion. A nation-state can assign resources and operate in the international area for the protection of members of the cultural–national dispersion and to foster the culture among them. In contrast, it is doubtful whether a multicultural state can shoulder the task of assisting the dispersion of each one of the cultural groups belonging to it. In addition, the nation-state, by its very existence, constitutes a focal point that consolidates around it the consciousness of the members of the national culture the world over.26 The Israeli instance serves as an excellent example of this last point. As Chaim Gans argues, the intense protection to a culture that a nation-state can grant is particularly vital to ensure the bequeathing of the culture from one generation to the next.27 But from the few examples we gave above, we see that the gap between the degree of protection possible within the framework of a nation-state and that which can be provided by a multinational state is no less significant from the point of view of the members of a given generation and in relation to their own lives. The need for realizing the right to culture in a nation-state becomes especially vital in the modern period. An interesting phenomenon that characterises the writing on the right to culture is the use of examples from separatist and marginal groups. For example, Kymlicka bases himself on the studies by Schwartz about the failure of integration of Indians into nineteenth-century United States,28 and he frequently refers to native cultures in Canada29 or Australia.30 Halbertal and Margalit inject into the liberal discussion the ultra-Orthodox enclaves in Jerusalem.31 Others focus on the Amish.32 It seems that these writers believe that the right to culture is necessary today only for members of separatist groups, since other cultures are in various stages of waning and assimilation at the end of which the members of these cultures will adapt a universal, eclectic culture for themselves. In other words, according to this approach, most citizens of the West today have no need for the protection of the right to culture, and in any event are not eligible for it. Our take on this issue is different. The difficulty prevailing today in maintaining cultural uniqueness does not overshadow the natural human need for protection of this uniqueness and its development, even for members of those cultural groups who chose or were forced to somewhat attenuate the distinct characteristics of their culture.33 At least part of the justifications for the right to culture, noted briefly above, is also relevant in the reality of Western life. Culture is still a characteristic of identity even among groups that are not separatist (Margalit and Halbertal), holding on to it is especially significant for those who have chosen it (Taylor), and preserving it serves for many 26 See, eg the articles in M Fludernik (ed), Diaspora and Multiculturalism: Common Traditions and New Developments (Amsterdam, New York, Rodopi, 2003). 27 See Gans, nn 16 and 21 above. 28 Kymlicka, Liberalism (n 17) ch 8. 29 ibid. 30 Kymlicka, Multicultural Citizenship (n 17) ch 2. 31 Margalit and Halbertal (n 20). 32 J Spinner, The Boundaries of Citizenship: Race, Ethnicity, and Nationality in the Liberal State (Baltimore, The Johns Hopkins University Press, 1994) ch 5. 33 Ruth Gavison presents a specific argument that supports his general claim. Gavison argues that the Jewish nation-state is vital precisely for secular Jews whose Jewish identity is not protected by religious foundations. See R Gavison, ‘The Jews’ Right To Statehood: A Defense’ (2003) 15 Azure 70, 76–77; R Gavison, ‘Thoughts About the Significance and Implications of “Jewish” in the Expression “Jewish and Democratic State”’ in Ravitsky and Stern (n 15) 107, 152 (in Hebrew).
492 Aviad Bakshi and Gideon Sapir people as an aspiration and a dream (Gans). The combination of these two facts, the importance of the protection of distinctive characteristics of culture and the difficulty in modern life to rely, for the purpose of preserving culture, on the strength of the veteran cultural frameworks, requires and justifies strong, mainly institutionalised, protection of the right to culture. A nation-state is capable of providing the national culture with that same forceful protection. In an era in which voluntary group frameworks find it difficult to stand in the breach, state protection becomes particularly vital. The great challenge the liberal writers are taking upon themselves is to translate the human need for cultural uniqueness into liberal terms and to propose a mechanism that will allow for this distinctiveness to operate in harmony with other liberal values. Even if we have succeeded in demonstrating that a nation-state is necessary for the realization of the right to culture, there will be a need to protect, within its framework, the rights of the members of minority culture groups who live in this state, including their right to culture. As noted above, there are those who feel that such a combination is not possible in a nation-state. We think differently. In this context, one must discuss separately first generation human rights and the collective right of the minority to culture. Certain practices within the framework of the nation-state are likely to stand in conflict with this or that individual right. Tension between rights, however, is a regular reality in the discourse of constitutional rights, which developed various balancing mechanisms for dealing with it. As with our issue, too, there will be a need to strike a balance. In each instance we will have to examine whether the practice is necessary to effectuate the right to culture of the majority, the degree to which the practice infringes on the rights of the minority citizens, and the various possible alternatives for realizing the right. At the end of the day, certain practices will be disqualified or limited and others approved. The nation-state will be required to also honour the minority’s right to culture. Within this framework there will be a place to consider the granting of partial or total auto nomy to the minority group for various issues related to maintaining their culture. Indeed, the very definition of a country as a nation-state means that the culture of the minority groups will not enjoy an equal status with that of the majority, a problem that would not arise in a multicultural state. Yet, it may be that even from the point of view of the minority groups in a given nation-state, the value inherent in the existence of their nation-state, which awaits them somewhere, is greater than the damage involved in defining their country of citizenship as the state of a different nation, as long as the latter honours their individual rights and their sub-state right to culture. Chaim Gans claims that the argument we offered fails in the light of the fact that to date there are some cultural groups in the world that do not receive the right to selfdetermination in their nation-state. This lack of equality obliges us, in his opinion, to relinquish the idea of a nation-state also on the part of the cultures that benefit, or might benefit, from realization of the right.34 We however agree with Taylor, who argues that the inequality noted does not justify renouncing the right to a national state of groups who have succeeded in realizing it, and this derives from two main reasons.35 First, any social rectification takes place gradually, and one should not expect it to be effectuated all at once. Second, the fact that one group does not receive full protection of its rights does not justify detracting from the protection granted to another group.36 Gans (n 16) 74–78. Taylor, ‘Why Do Nations’ (n 24). 36 For Gans’s response to this argument, see Gans (n 16). 34 35
A Jewish Nation-State 493
B. Other Justifications
Yael Tamir37 and David Miller38 support the nation-state because of its potential contribution to the protection of civil and political rights, such as rights concerning welfare and the ensuring of personal security. Their argument is based on two premises. The first is that strong solidarity is required among all citizens and between them and the state in order to ensure a satisfactory degree of protection of first-generation human rights (civil and political) and second-generation rights (social). Only that way will citizens honour the rights of each other, agree to relinquish economic resources for the benefit of their needy fellowmen, and take upon themselves the risks involved in defending the state and its borders. The second argument is that a particularly high level of solidarity is usually found among members of an ethno-cultural group. Basing themselves on these two premises, Tamir and Miller (with slight differences between them that are not significant for the current discussion) propose using the model of a nationstate as a means for harnessing ethno-cultural solidarity for the purpose of strengthening civil solidarity. A few years ago, Amnon Rubinstein and Alex Yakobson published a book, Israel and the Family of Nations.39 The book contains a comprehensive survey of countries, whose immigration laws attest to their being nation-states. This survey is of great importance, but the claim that ‘this is what everybody does’ cannot serve as a substitute for the need for independent, normative justification for a model of a nation-state.40 Similar remarks can be made about the argument, which Ruth Gavison uses extensively, to wit the right to national self-determination was recognised in the law of nations following World War I.41 This fact is relevant in the context of international relations and international law, but it does not provide normative justification for a model of a nation-state. C. Justification from the Right to Refrain from Persecution Alongside general justifications for a nation-state, one may also present one more limited in scope, which strengthens the right of the Jewish people (and nations similar to it with the relevant characteristics) to its own nation-state. The justification, which has a number of versions, is based on the circumstances of the Jewish people as a nation that underwent severe persecution and is even today the object of hatred. In 1980, the Israeli writer AB Yehoshua published a book that describes the Holocaust of the European Jews during World War II as a critical crossroads in Jewish history.42 To be sure, Jews had suffered severe persecution in many countries and in different periods. But Yehoshua feels that the Holocaust led to a watershed after which the Jews could no longer allow themselves to exist without their own country. In light of this Y Tamir, Liberal Nationalism (Princeton, Princeton University Press, 1993) ch 6. Miller, (n 25) ch 3. 39 A Yakobson and A Rubinstein, Israel and the Family of Nations – The Jewish Nation-State and Human Rights (Jerusalem, Schocken Publishing House, 2003) (in Hebrew). 40 See G Sapir, ‘Book Review – Yakobson and Rubinstein: Israel and the Family of Nations’ (2007) 11 Democratic Culture 237. 41 See, eg R Gavison, ‘Thoughts About the Significance’ (n 33) 107, 109 (in Hebrew). 42 AB Yehoshua, For Normality (Jerusalem, Schocken Publishing House, 1980) (in Hebrew). 37 38
494 Aviad Bakshi and Gideon Sapir initial premise, Yehoshua turns to discussing normative justification for the establishment of a Jewish nation-state. Yehoshua rejects the moral validity of historical or religious justifications and turns to justification on the basis of what he calls ‘the right of existential distress’.43 He argues that whoever is under severe threat is permitted to defend himself, even on the account of legitimate interests of a third party as long as this does not lead the third party to the same lack and threat from which he is suffering. Yehoshua’s argument seeks to justify the immigration of Jews to the area in which the State of Israel exists today. He uses the argument of the persecutions to justify the nation-state post factum, but this claim is strong enough to base the right, to begin with, of a persecuted people to found for itself a nation-state. Furthermore, it is incumbent upon the family of nations to help it realise this right. Obviously a nation-state will rise in a certain area and on account of certain people, an issue which will require a dis cussion as to why precisely that area and its specific inhabitants are the ones who must pay the price for the protection of the persecuted nation. But, as noted, the right of a persecuted people creates an obligation in principle for humanity to ensure that the persecuted group will gain a safe haven. III. A NATION-STATE VERSUS FAMILY REUNIFICATION
A. The Law and the Court Rulings Israeli immigration law deals, inter alia, with the situation that is called in popular jargon ‘family reunification’, in which a citizen seeks to bring into the country his/her foreign spouse and grant that spouse citizenship. The citizenship law makes the nationalization process of the spouse easier and exempts him/her from most of the conditions an ordinary seeker of naturalization is required to fulfill.44 In fact, the Interior Minister approves these requests in the vast majority of cases. After the Oslo Accords were signed between Israel and the PLO in 1993, a steep increase occurred in the number of marriages between Palestinians who were residents of the Territories and Arab citizens of Israel. Under these circumstances, the lenient policy for the approval of citizenship for spouses of Israelis led to a most significant increase in the number of Palestinians who became citizens owing to reunification of families. According to data from the Ministry of Justice, between 1994 and 2002, 130,000–140,000 Palestinians became naturalised Israeli citizens.45 In March 2002 the Interior Minister froze the graduated process for family reunification of Israelis with Palestinian spouses. The declared motivation for the decision was security, after a Palestinian who held an Israeli identity card, given to him as a result of the process of his parents’ reunification, carried out a terror attack wounding many. In May 2002 the policy was anchored in a Government Decision.46 A number of petitions ibid 95–105. Citizenship Law, 5712-1952, s 7. On this issue there is almost no significance to the question as to whether the citizen among the spouses is a Jew; HCJ 3648/97 Stamka v Minister of Interior 53(2) PD 728 [1999] (in Hebrew). 45 From statements by the then Deputy Attorney General, Meni Mazuz, in a discussion of the Knesset’s Interior Committee on 14 July 2003 on the proposed Temporary Provision Law. The protocol of the discussion can be viewed on the Knesset website: www.knesset.gov.il/protocols/data/html/pnim/2003-07-14-01.html. 46 Government Decision 1813, taken on 12 May 2002. 43 44
A Jewish Nation-State 495
were lodged with the Supreme Court (sitting as the High Court of Justice) against this Government Decision.47 While the processes in the Court were still underway the Knesset (Israeli Parliament) passed the Temporary Provision Law,48 which anchors the limitations the government imposed.49 The Law was enacted as a temporary provision, explained by security constraints. In the period of the Law’s validity, the Interior Minister is prohibited from granting a permit to stay in Israel, a license to reside in Israel, or Israeli citizenship to residences of the West Bank, except for a very few exceptional cases specified in the Law. The Law determined a mechanism for the government, with the approval of the Knesset, to extend the period of validity of the temporary provision. The government made use of this mechanism a number of times, and each time extended the validity of the Law for a few months. A number of petitions challenging the legality of the Law were presented to the Supreme Court, which dealt with them as an expanded panel of 11 justices and rejected them by a difference of one vote (herein after: Adalah 2006).50 Taking into consideration the Court’s comments during the course of the discussion on the petition, the Knesset amended the Law, and determined a number of alleviations, even before the ruling was given.51 After the Court’s ruling the Knesset amended the Law once again (hereinafter: the third version). The third version of the Law increased the category of exceptions to the prohibition even more and in addition ordered the establishment of a committee that the Interior Minister should turn to when addressing family reunification for humanitarian reasons. The third version broadened the prohibition on reunification of families and applied it also to petitioners whose country of origin was Iran, Lebanon, Syria, or Iraq.52 A mechanism was determined that enables the government to extend the validity of the Law, with the authorization of the Knesset, for a period of up to one year for each extension. As stated, the petition was submitted against the Temporary Provision Law in its first, most extensive version. The ruling was rendered in 2006, when the second version of the Law was in effect and that enabled the Interior Minister to permit the residence in Israel of women above the age of 25, men above the age of 35, and children below the age of 14. In the judgment, which was delivered by a panel of 11 justices and which covered 263 pages, the petitions against the legality of the Temporary Provision Law were rejected by one vote.53 47 See, eg HCJ 4022/02 Association for Civil Rights in Israel v Minister of Interior (11 January 2007), Nevo Legal Database (by subscription) (in Hebrew). The appeals became redundant with the passage of the Law described here. Yet, in light of the petitioners’ arguments, a short ruling was rendered, in any event, in 2007 on these processes that rejects the petitions that attack the Government Decision and not the legality of the Law. 48 ibid n 13. 49 For the background on the enactment of the Law and a critical analysis of its legality, see G Davidov, J Yovel, I Saban and A Reichman, ‘State or Family? The 2003 Amendment to the Citizenship and Entrance to Israel Law’ (2006) 8 Law and Government 643 (in Hebrew); as well as A Rubinstein and L Orgad, ‘Human Rights, National Security and Jewish Majority: The Case of Restrictions of Family Migration’ (2006) 48 The Lawyer 315 (in Hebrew). 50 HCJ 7102/03 Gal-On v Attorney General (30 May 2005), Nevo Legal Database (by subscription) (in Hebrew); HCJ 8099/03 Association for Civil Rights in Israel v Minister of Interior (14 May 2006), Nevo Legal Database (by subscription) (in Hebrew); HCJ 8263/03 Askafi v Minister of Interior (14 May 2006), Nevo Legal Database (by subscription) (in Hebrew); HCJ 7052/03 Adalah v Minister of Interior 61(2) PD 202 [2006] (in Hebrew). 51 Citizenship and Entry into Israel Law (Temporary Provision) (Amendment), 5765-2005, SH No 2018, p 730. 52 ibid para 3d. 53 Adalah v Minister of Interior (n 50).
496 Aviad Bakshi and Gideon Sapir The minority justices thought that the petitions should be accepted. The main opinion among the minority justices was written by Court President Aharon Barak, and Beinisch, Joubran, Hayut, and Procaccia JJ concurred with him. Barak examined the issue from the viewpoint of the Israeli spouse, while leaving undecided the question of whether the foreign spouse also has relevant legal rights. Barak’s opinion is divided into three parts. In the first, he examined whether the Law infringes on constitutional rights; in the second, he surveyed whether the infringement of rights, on the assumption that this exists, meets the criteria of the Limitation Clause; and in the third, he dealt with the proper remedy, on the premise that the infringement is not justified. In the first part, Barak argued that the right to dignity anchored in the Basic Law: Human Dignity and Liberty should be interpreted as also including the right to establish a family, and this right includes the right of the Israeli citizen to receive his/her spouse in Israel and to carry on their joint life. In light of this, Barak determined that the Temporary Provision Law infringed the constitutional right to establish a family.54 Barak further determined that the Law infringed the right to equality of Arab Israeli citizens, since in practice the Law is directed almost exclusively toward them and thus discriminates against them, in comparison to Jewish citizens, in relation to their ability to realise their right to a family.55 In order for such an infringement to serve as the cause for invalidating the Law, the right must be anchored in a Basic Law. The right to equality is not included among the rights protected by the Basic Laws, as it was explicitly removed from the formulation of the Basic Law: Human Dignity and Liberty.56 But Barak determined that this aspect of equality is included within the framework of the constitutional right to dignity, and therefore the Temporary Provision Law contradicts the constitutional right to dignity also because of the detriment to equality. In the second part, Barak examined whether the infringement of the rights to a family life and equality meets the conditions of the Limitation Clause. The state argued that the purpose of the Law is to prevent terror attacks, and Barak determined that this goal is worthy.57 For our purposes, it is important to note that the state did not argue about the existence of another aim, so Barak was not required to examine other goals, including the demographic one. We shall return to this possible aim later. From here, Barak turned to examine the proportionality of the harm to the right.58 Barak determined that a rational link exists between the means adopted by the Law – sweeping limitation of the entry of Palestinians – and the goal – prevention of terror. Barak also determined that it is, of course, possible to employ less harmful means – an individual examination of requests for family reunification, but this means will not yield the same degree of security as sweeping limitation of immigration, so the Law passes the second proportionality test. Yet, in Barak’s opinion the Law fails the third proportionality test. Barak argued that the third test includes two sub-tests: one sub-test examines the benefit inherent in the Law as it is versus the damage it involves. The second sub-test looks at the additional benefit (in terms of realizing the goal) bound up with the Law in comparison to other alternatives, which effectuate the goals to a lesser degree, versus the additional harm ibid paras 31–34, 42–45. ibid para 46. 56 See G Sapir, ‘Constitutional Revolutions – Israel as a Case Study’ (2010) 5 International Journal of Law in Context 358. 57 Adalah v Minister of Interior (n 50) para 82. 58 ibid paras 84–92. 54 55
A Jewish Nation-State 497
involved in the Law in comparison to the above-mentioned alternatives.59 Barak argued that the Law does not pass the second sub-test, in the sense that the increase in security deriving from the mechanism that the Law determines, in comparison to the alternative of individual examination, is not of greater weight than the additional harm deriving from the mechanism rooted in the Law. In other words, Barak felt that it would be worthwhile to pay in terms of security so as to lower the harm to human rights.60 Concurring with Barak’s minority opinion were, as noted, Beinisch, Joubran, Hayut, and Procaccia JJ. The ruling by Procaccia J differed from those of her minority opinion colleagues over the question of the purpose of the Law. Procaccia did not accept the state’s claim according to which the motivation for the Law was only security, and she referred – as a reason for her lack of belief – to a discussion held in the Knesset as part of the legislative process in which various Knesset Members presented demographic considerations as the purpose of the Law. Procaccia refrained from expressing an unequivocal opinion over the question of whether the demographic goal could serve as a proper one, and she sufficed with the incongruence between the declared aim and the aim that could be deduced (as stated, solely in her opinion) as the reason for determining that the Law does not pass the test of worthy goal.61 Edmond Levy J was the one who tilted the scales. In principle, Levy concurred with the minority justices. He went hand in hand with Barak, leaving him only at bottom line. Levy thought that the state should be given an extension of nine months to prepare for the invalidation of the Law, and since at the time of the rendering of the judgment, the Law would in any event expire within a short time, Levy refrained from concurring in his opinion with those who proposed declaring the Law null and void. Levy added that if the validity of the Law would be extended without changes being made to it as called for by the instructions of the ruling, it would be possible to request another discussion on the constitutionality of the Law. Cheshin J wrote the main opinion of the majority. As he understood it, the Law does not infringe at all upon a constitutional right. A citizen does not have a constitutional right to absorb his foreign spouse into his country, while the right to equality is not harmed since there is a relevant difference between a person married to a subject of an enemy country and someone married to a citizen of a friendly country.62 Cheshin also opposed Barak’s conclusion on the question of proportionality. He took exception to Barak’s claim according to which a security price must be paid for preserving human rights and determined that the right to life overrides the other rights of man. A number of petitions were submitted against the third version of the Law,63 and in January 2012 the Court rejected the petitions with a majority of six to five (herein after: Gal-On 2012). Four of the minority justices, harked back to the position of the minority justices of Adalah 2006 and determined that the Temporary Provision Law did 59 For an academic presentation by Barak of the ruling in general and of his interpretation of the assessments of proportionality in particular, see A Barak, ‘Proportional Effect – The Israeli Experience’ (2007) 57 University of Toronto Law Journal 369, 383; for a critical discussion of Barak’s interpretation of the assessments of proportionality, see G Sapir, ‘Proportionality and Compromise’ in S Almog, D Beinisch and Y Rotem (eds), Dalia Dorner Book (Tel-Aviv, Nevo Publishing, 2009) 397 (in Hebrew). 60 HCJ 2056/04 Beit Sourik Village Council v Government of Israel 58(5) PD 807 [2004] (in Hebrew). 61 Adalah v Minister of Interior (n 50) paras 14, 19. 62 ibid paras 86, 92. 63 HCJ 466/07 Gal-On v Attorney General (11 January 2012), Nevo Legal Database (by subscription) (in Hebrew).
498 Aviad Bakshi and Gideon Sapir disproportional damage to the rights of family and equality. Levy J joined the minority position but differed from his colleagues with the argument that the Law did not coincide with the values of Israel as a Jewish and democratic state, and did not even meet the criterion of worthy goal.64 Similar to the majority justices in Adalah 2006, the six majority justices in Gal-On 2012 thought that even if the Temporary Provision Law harms constitutional rights, this harm meets the criteria of a Limitation Clause, in light of the security aim of the Law and in consideration of the exceptions to the Law. B. The Status of the Demographic Goal and Its Relative Weight As cited above, In Adalah 2006 Procaccia J rejected the state’s argument according to which the purpose of the Law is to guard the security interest of the state and its citizens, and she claimed that circumstances inform us that also resting at the base of the Law is the demographic aim. Yet, even Procaccia J did not express a clear opinion on the issue of the legitimacy of this aim. What did the other members of the panel think about the demographic goal? On the surface, we cannot know, since the majority of the Court’s justices accepted the state’s argument that the security goal and not the demographic one is what underlies the Law, and that being the case, they did not have to discuss the legitimacy of the demographic goal. Reinforcement for this response can be found in the statement by Cheshin J who rebuked his colleagues Procaccia and Joubran who treated the demographic issue incidentally.65 On second thoughts, however, perhaps the justices’ refraining, in Adalah 2006, from referring to the question of the legitimacy of the demographic goal can teach us something about their stance on the topic. If they were convinced that the demographic goal was legitimate and did not create problems, they would say so explicitly. Two pieces of data reinforce this observation. First, during the course of the discussion of a petition, the Court permitted a voluntary association, ‘The Jewish Majority in Israel’, to join the discussion as a respondent. This association argued that the demographic consideration was definitely at the base of the Law, that this consideration is a worthy aim, and that the Law attains this goal in proportionality. Secondly, in the state’s replies to the Supreme Court, it did indeed argue that the Law was not intended for the realization of the demographic goal. But the Attorney General added that ‘even if the dominant goal of the law was demographic – which does not seem to be the case here – then this purpose is likely to coincide with the values of the State of Israel as a Jewish and democratic state’.66 The fact that the Court chose to ignore these statements strengthens the sense that the demographic consideration hovered about the Court, but it appeared to the justices of the panel as a fear-instilling wraith, to the point that they hurried to hide themselves from it. As noted above, the Court’s rendering in Gal-On 2012 does not differ in its bottom line from Adalah 2006. The Court again held firm to the security explanation as the exclusive justification for the temporary provision. Yet, a number of incidental remarks ibid para 26. Adalah v Minister of Interior (n 50) para 135 of Cheshin J’s opinion. 66 Para 169 of the closing arguments of the state of 16 December 2003 as quoted in Adalah v Minister of Interior (n 50) para 14 of Procaccia J’s opinion. 64 65
A Jewish Nation-State 499
scattered throughout the opinions of the justices are likely to indicate judicial willingness to consider the demographic justification positively, if it will be presented explicitly in the future by the state. We will survey some of these expressions in brief. Levy J who, as noted above, determined that the security reason does not pass the test of worthy goal, points out that it is not at all clear whether the state’s refraining from basing the Law on the demographic consideration reinforces its position or weakens it.67 Levy even mentions the immigration wave by reasons of ‘family reunification’ that took place in Israel in the period 1994–2002 in a way that can be understood as alluding to the legitimacy of employing the demographic consideration.68 Court Vice President Rivlin explains that he is discussing the security rationale alone, not only because the demographic rationale was not presented by the state, but also because in his opinion the demographic rationale is not apt for the Law in its current form. Yet, Rivlin notes incidentally that the state was permitted to enact family immigration law based on the demographic aim.69 In this context Rivlin quotes from the statement by Court President Barak in a different instance,70 according to which ‘the reason for the existence of the State of Israel is in its being a Jewish state’ and that ‘it is the right of every Jew to immigrate to the State of Israel, in which the Jews will constitute a majority’ constitutes one of the core characteristics of Israel as a Jewish state. Melcer J refrained from basing his decision upon the demographic consideration in light of the state’s position, but he does determine that regarding ‘a discussion such as this, its time will come’,71 and he cites in this context the position of Ruth Gavison,72 who holds that one may justify the Temporary Provision Law (even in its original, stringent version) through demographic considerations. The only justice of the panel whose rendering can be explained as actually giving weight to the demographic rationale is that of Naor J. Naor argues that rationales other than security (ie the demographic rationale) have implications for the laws of immigration to Israel.73 From this statement it may be inferred that in Naor’s opinion – despite the fact that with the lack of the state addressing the demographic consideration, it is not possible to take it into account as part of the test of worthy goal – the demographic consideration is still relevant, as a cumulative reason for determining the need to interpret sparingly the constitutional right to family life, such that it will not include the right of a married citizen to bestow citizenship in his/her state to his/her spouse. To sum up, in both instances – Adalah 2006 and Gal-On 2012 – the Court refrained from rooting its decision on the national–demographic goal. Yet, whereas in Adalah 2006 all the majority justices distanced themselves from the demographic reason, in Gal-On 2012, some of the justices stress that their stance derives strictly from the state’s refraining from basing its position on the demographic goal, and premise in incidental remarks a foundation for taking that goal into account in the future. Is the goal of maintaining a solid Jewish majority among the citizens of the state a proper aim? In the lack of judicial reference to that question, the other natural place to look for an answer to the question is in the academic discourse. And, indeed, the law and Gal-On (n 63) para 1 of Levy J’s opinion. ibid. 69 ibid para 16 of Rivlin J’s opinion. 70 Central Elections Committee (n10). 71 Gal-On (n 63) para 31 of Melcer J’s opinion 72 ibid. 73 ibid paras 4–5. 67 68
500 Aviad Bakshi and Gideon Sapir the court’s decision stimulated great academic interest and reaped an impressive harvest of publications.74 Yet, the members of academe – except for a pair of writers in one article that argued that maintaining a demographic balance constitutes a legitimate consideration75 – referred only to the security goal and did not discuss the demographic one. We think that the demographic goal is proper. The conclusion drawn in the previous section was that members of the Jewish people have the right to political selfdetermination, whether from the right to culture, whether owing to the right to avoid persecution, or whether as a right intended to advance the interest of reinforcing the solidarity among citizens of the state. Apparently the existence of a solid Jewish majority constitutes a condition for the legitimacy of the definition of the state as a Jewish nation-state. If the members of the Jewish people cease to constitute a majority in the State of Israel, or even if they will remain a majority but lose a decidedly numerical advantage, they will lose the right to commit the state to the advancement of their culture. From this we learn that the state has a clear interest to act for the purpose of preserving the Jewish majority among its citizens and that this interest in itself is a worthy one. The conclusion that the demographic goal is worthy does not lead perforce to the conclusion that the law in one version or another is constitutional. For the purpose of the current discussion, we are ready to presume, without an in-depth discussion, that realization of the law infringes upon certain rights of members of the Arab minority in the state. In light of this premise, in order to make the law acceptable one must make certain that its arrangements find a proper balance between the worthy purpose and the harmed rights. Such an examination requires delving into the proportionality tests, while referring to the different components of the law and weighing the various alternatives that could be proposed for the purpose of attaining the demographic goal. Of course, a review such as this also involves the operation of strong discretion, which by its nature is subjective and dependent upon the worldview of the observer. We are convinced that the advantage of the researcher over the judge and the politician is not in his ability to invoke strong discretion such as this in a particularly good way but in his ability to sharpen normative considerations. Thus, we will limit our discussion to the presentation of these considerations and will refrain from taking the ultimate decisions on the question of the legitimacy of the law in this or that version. There are actions aimed at achieving demographic goals which can be justified, and there are acts for this purpose which can be peremptorily rejected. The Law of Return, which grants Jews the right to immigrate to Israel and become a citizen of the state, serves in our opinion as a good example of measures of the first kind, and indeed, as far 74 See, eg B Medina and I Saban, ‘Human Rights and Risk-Taking: On Democracy, Ethnic-Profiling and the “Limitation Clause”’ (2009) 39 Mishpatim 47 (in Hebrew); L Orgad, ‘Immigration, Terrorism and Human Rights: Admission Policy and the War on Terror in Time of National Emergency’ (2009) 25 Bar-Ilan Legal Studies 485 (in Hebrew); Y Ben-Shemesh, ‘Constitutional Rights, Immigration and Demography’ (2006) 10 Law and Government 47 (in Hebrew); Y Zilbershats, ‘Coping with the Non-Jewish Immigration to Israel (case comment)’ (2006) 10 Law and Government 87 (in Hebrew); Rubinstein and Orgad (n 49); Davidov, Yovel, Saban and Reichman (n 49) (the last two articles were written even before the decision was rendered); N Carmi, ‘The Nationality and Entry to Israel Case Before the Supreme Court of Israel’ (2007) 22 Israel Studies Forum 26; D Barak-Erez, ‘Israel: Citizenship and Immigration Law in the Vise of Security, Nationality and Human Rights’ (2008) 6 International Journal of Constitutional Law184. 75 Orgad and Rubinstein (n 49) 341–46.
A Jewish Nation-State 501
as we know, at least among the members of the Jewish majority, there is no dispute over the legitimacy of this law.76 Conversely, the vast majority of the Jewish citizens of the country would summarily reject any proposal to act to preserve the Jewish majority by negating the citizenship of non-Jewish citizens of the state, all the more so through expelling them by force from Israel. In the middle, between legitimate steps to reinforce the Jewish majority in the state and unacceptable measures, there are those whose legitimacy can be disputed. The law under discussion belongs to this intermediate group. Countering the law are the rights that have been infringed upon, but there are also considerations that should operate in its favour, considerations that either reinforce the strength of the Jewish stake or weaken the intensity of the infringement on the rights of citizens who are seeking family reuni fication. We shall enumerate a number of these considerations: first, the State of Israel has recognised the right of the Palestinians to a nation-state on part of the lands between the Jordan River and the Mediterranean. This datum reinforces the legitimacy of Israel’s tenacity to ensure the conditions necessary so that in this area there will also exist a Jewish nation-state, and these include, as noted, a clear Jewish majority in this state. It is unreasonable that while in the Territories a Palestinian nation-state is in the process of being constructed, with Israel’s agreement and support, that the Jews will be forced to see the loss of the national identity of their country. Second, the political leadership of the Palestinian minority in Israel, the Palestinian leadership in the West Bank Territories, and all Arab countries, deny the right of the Jewish people to political self-determination in any area of the Land of Israel. This fact increases the legitimacy of actions on the part of the Jewish majority whose aim is the preservation of strong numerical dominance, even if the fact of its being the numerical majority is not seriously endangered yet. Third, the immigration data of Palestinians in the period prior to the Temporary Provision Law show that it is not trivial numbers that are under discussion.77 We are talking about a staunch stream of immigration that is not waning. This fact turns protection of the Israeli majority into an especially pressing mission. Finally, the citizens of Israel belonging to the Palestinian minority, who seek family unification, can do so in the territory of the West Bank. The West Bank Territories are located quite close to Israel, and they, or a large part of them, are to be included within the area of the future Palestinian nationstate. Therefore, such a measure will not obligate them to be completely uprooted from their land or their culture. This fact weakens the intensity of the infringement on the right to family life of the Palestinian Israeli citizens who seek family reunification. We think that these considerations reinforce the legitimacy of the law. But, as noted, even if the limitation on immigration from the Territories to Israel is legitimate, this does not mean that it justifies levying a sweeping prohibition against immigration for reasons of family reunification, nor even the law in its present form, an attenuated version in comparison to its original formulation. The provisions of the law must be individually examined through the prism of proportionality tests. As stated, we do not intend to make such an examination here.
76 The members of the Palestinian minority deny this too. See, the National Committee for the Heads of the Arab Local Authorities in Israel, ‘The Future Vision of the Palestinian Arabs in Israel’: www.adalah.org/ newsletter/eng/dec06/tasawor-mostaqbali.pdf. 77 See statements by Meni Mazuz, ibid n 45.
502 Aviad Bakshi and Gideon Sapir IV. SUMMARY AND CONCLUSIONS
In this chapter we argued that the aspiration for self-determination within the framework of a nation-state is an interest that should be recognised as a human right. Furthermore, the status of this right is reinforced in the context of the Jewish people owing to the unique circumstances of this nation, which is forced to fight for its very survival against those trying to annihilate it. We pointed out the fact that the existence of a clear demographic majority of the members of a nation in a sovereign state constitutes an imperative condition for the legitimacy of its definition as the state of this nation. Therefore, maintaining the existence of this condition constitutes a worthy goal. For it to be justified, acts toward the maintenance of a demographic majority must meet the tests of proportionality. In the operation of proportionality tests, one must consider a number of relevant variables. These variables include, among others, the question as to whether the members of the minority enjoy or are likely to enjoy selfdetermination, whether the members of the minority group recognise the right of the majority to self-determination, and the degree of the threat challenging the imperative conditions for realizing it. In the present context, all three of the variables reinforce the weight of the right of the Jewish people to self-determination in its nation-state and of vigorous action for the realization of this right. At the same time, the intensity of the infringement on the rights of the members of the minority group is likely to change from one instance to the next, and the variables influencing the degree of harm must be identified and considered. In the present context, the fact that family reunification can be realised in close proximity to the present place of residence of the seeking citizens, weakens the intensity of the infringement on their right to family life if their request to realise the right within the boundaries of Israel is denied.
33 National Identity and Religion–State Relations: Israel in Comparative Perspective GILA STOPLER
I. INTRODUCTION
D
IFFERENT STATES HAVE different models of relations between religion and the state. There are diverse reasons why a state chooses to adopt a particular model of religion– state relations, one of which may be the role given to religion in creating and strengthening national identity. Israel is defined as a Jewish and democratic state and the Orthodox Jewish religion is established in the state in several ways.1 While there is an ongoing debate in Israel on whether its definition as a Jewish state should be understood as referring to the Jewish religion or merely to Jewish nationality, there can be no doubt that the partial establishment of the Orthodox Jewish religion in the state plays an important role in Jewish national identity in Israel.2 This phenomenon is not unique to Israel. Since religion and nationalism are often closely related, many times religion–state relations play a significant role in a polity’s national identity. This chapter will explore the connection between national identity and religion–state relations in Israel and will place it in a comparative perspective. It will argue that with respect to national identity religion can serve two functions: first, the unity function (an inclusive function) – to unite different individuals or groups under a unitary national identity; and second, the gatekeeping function (an exclusionary function) – to make sure that the boundaries of the national group remain clear and to exclude from the group anyone who may dilute or threaten its identity. Next, the chapter will discuss several countries in which religion is a central component of national identity, including Greece, Malaysia and Turkey. It will show how the nexus between national identity and religion finds expression in different religion–state structures, and point to the functions that it serves. Finally, the chapter will describe religion and state relations in Israel, discussing in what ways Israel’s chosen model of religion–state relations was expected to aid the strengthening of the Jewish national identity. The chapter will claim that the Israeli model has been partially successful, at best, in achieving both its unity function and its 1 D Barak-Erez, ‘Religion and the Secular State – An Israeli Case Study’ 1–3 www.juridicas.unam.mx/wccl/ ponencias/11/329.pdf. 2 Y Shapira, ‘Secular Politicians and the Status of Religion in the State of Israel’ in M Mautner et al (eds), Multiculturalism in a Democratic and Jewish State, The Ariel Rosen-Zvi Memorial Book (Tel-Aviv, Tel-Aviv University Press, 1998) 663 (in Hebrew).
504 Gila Stopler gatekeeping function and will argue that a major cause for the Israeli model’s limited success is the state’s lack of control over its own religious establishment. II. THE ROLE OF RELIGION IN NATIONAL IDENTITY
Prominent writers on nationalism, such as Anderson and Gellner, tend to ignore the role that religion often plays in the creation and preservation of nations.3 In fact, modern Western discourse has until recently maintained a dichotomy between nationalism as the product of modernization and secularization, and religion as a phenomenon of the past, which is destined to disappear with the triumph of the enlightenment.4 Nevertheless, a closer examination of the relationship between religion and nationalism shows that religion has played and still plays an important role in many national movements in both the Western and the non-Western world.5 Moreover, religion continues to play an important role not only in national movements but also in the nation-states that they create and even in the constitutional structure of these states. Thus, even among Western democracies one can find countries such as England, Ireland and Greece, in which the dominant religion, which has played an important role in the creation of the nation, is given some constitutional recognition, from the largely symbolic, as is the case in England, to the more robust as is the case in Greece.6 Giving constitutional recognition to a national religion can be done in a variety of forms and can serve various functions. I will suggest that two central functions that religion plays in national identity are the unity function and the gatekeeping function. A. The Unity Function Just as with other types of communities, the formation of a national community depends on the creation of a mutual bond around which the community can be united and sustained. Benedict Anderson famously suggested that a nation should be understood as ‘an imagined political community’.7 According to Anderson nations are imagined because their members ‘never know most of their fellow-members, meet them, or even hear of them, yet in the minds of each lives the image of their communion’.8 Nations are communities because ‘regardless of the actual inequality and exploitation that may prevail in each, the nation is always conceived as a deep, horizontal comradeship’.9 National communities, according to Anderson, ‘are to be distinguished . . . by the style in which they 3 B Anderson, Imagined Communities: Reflections on the Origin and Spread of Nationalism (London, Verso, 1991); E Gellner, Nations and Nationalism, 2nd edn (Blackwell, 2006). 4 P van der Veer and H Lehmann, ‘Introduction’ in P van der Veer and H Lehmann (eds), Nation and Religion: Perspectives on Europe and Asia (Princeton, Princeton University Press, 1999) 1. 5 See, eg ibid 4; PW Barker, Religious Nationalism in Modern Europe: If God be for Us (Oxon, Routledge, 2009). 6 See SV Monsma and JC Soper, The Challenge of Pluralism: Church and State in Five Democracies, 2nd edn (Lanham, Rowman & Littlefield Publishers, 2009) 131 (discussing the largely symbolic constitutional recognition of religion); see section IIIA of this chapter (discussing Greece’s hearty constitutional recognition of religion). 7 Anderson (n 3) 6. 8 ibid. 9 ibid 7.
National Identity, Religion and State 505
are imagined’.10 One way of imagining the nation and creating the necessary mutual bond is through a common religion, which can create a sense of comradery between the members of the nation and serve as the basis of national identity. Ernest Gellner suggests that the concept of the nation, although contingent and elusive, includes two important elements, the cultural and the voluntaristic.11 According to Gellner, the cultural element suggests that ‘two men are of the same nation if and only if they share the same culture, where culture in turn means a system of ideas and signs and associations and ways of behaving and communicating’.12 The voluntaristic element suggests that ‘Two men are of the same nation if and only if they recognise each other as belonging to the same nation’.13 The voluntaristic element is crucial since [a] mere category of persons (say, occupants of a given territory, or speakers of a given language, for example) becomes a nation if and when the members of the category firmly recognise certain mutual rights and duties to each other in virtue of their shared membership of it. It is their recognition of each other as fellows of this kind which turns them into a nation, and not the other shared attributes, whatever they might be, which separate that category from non-members.14
Thus, common religion can serve to unite its members into a nation, but it will only do so if the co-religionists recognise each other not merely as co-religionists but also as members of a common nation that share mutual rights and duties as members of the nation. As will be further discussed below, in cases where other factors may hinder the creation of a unified national identity, nation-states may look to religion as a means of uniting their potential members and garnering their support, and may even give religion official status and authority in the state in order to do so. B. The Gatekeeping Function Nations are founded and preserved not only on the basis of the unity of their members, but also by preserving the difference between those who belong to the nation and those who do not. In this respect national communities resemble religious communities. One way through which religious communities control their membership is through restrictive rules of marriage. It has even been suggested that the function of religious family law for the religious community is akin to the function of citizenship laws for the sovereign state. In both cases the laws serve to demarcate the boundaries of the community and decide who belongs to the community and who is excluded.15 State laws that give a monopoly to religious communities over family law, and recognise only religious marriages between community members, may serve the need of religious communities to preserve and demarcate their boundaries and thus may have an important gatekeeping function for the religious communities. At the same time, as will be demonstrated below, in nation-states that wish to demarcate the boundaries of the ibid 6. Gellner (n 3) 6–7. 12 ibid 6. 13 ibid 7. 14 ibid. 15 A Shachar, Multicultural Jurisdictions: Cultural Differences and Women’s Rights (Cambridge, Cambridge University Press, 2001) 45–47. 10 11
506 Gila Stopler nation along religious lines, the maintenance of a religious monopoly over family law, at least with respect to the majority religion, has an important gatekeeping function for the national community as well. However, one must not forget that this manner of preserving community boundaries, both religious and national, may result in violations of the rights of individuals within and outside the community. These violations may include the violation of women’s right to equality through the imposition of patriarchal and discriminatory religious laws; the violation of the right to family life through restrictions on inter-religious marriages; and the violation of the right to freedom of conscience and to freedom from religion by the compulsory imposition of religious laws.16 III. RELIGION–STATE RELATIONS AND NATIONAL IDENTITY – A COMPARATIVE PERSPECTIVE
A. Greece The close ties between Greek national identity and the Greek Orthodox religion are officially articulated in the Greek Constitution through the status that it grants to the Eastern Orthodox Religion. Article 3 of the Constitution stipulates that: The prevailing religion in Greece is that of the Eastern Orthodox Church of Christ
and Article 16 states that the development of national and religious consciousness
through education is a fundamental state objective.17 These provisions are the basis for the daily prayers in Greek schools and for state financing of the teaching of Orthodoxy in these schools.18 The interrelationship between the Greek Orthodox Church and Greek national identity goes back to the establishment of the modern Greek state in the 1820s.19 Prior to the establishment of modern Greece, the Ottoman Empire ruled over the region through its millet system which divided the population into groups according to their religion. The Orthodox Church used its relative freedom under the millet system in order to teach its members the Greek language and customs and played an important role in the Greek revolution.20 During and after the revolution a struggle was waged over the direction in which modern Greek national identity should be developed, between those who favoured the cultivation of a Westernised secular liberal identity and those who favoured promoting a Byzantinist nationalist identity of a Greek Orthodox nation.21 While Greek political leaders decided to fashion the state along European lines in order to appeal to central powers such as Britain and France, they also realised that Church support would legit See nn 59–65 below and accompanying text. 1975 Syntagma [Constitution] Arts 3, 16(2) (Greece). 18 E Karagiannis, ‘Secularism in Context: The Relations between the Greek State and the Church of Greece in Crisis’ (2009) 60 European Journal of Sociology 133, 146. 19 CP Danopoulos, ‘Church–State Relations, National Identity, and Security in Post-Cold War Greece’ (2004) 30 Journal of the Hellenic Diaspora 7, 13. 20 ibid 16–17. 21 ibid 14. 16 17
National Identity, Religion and State 507
imise their nation building efforts.22 Consequently, Greece’s dual commitment to Western democracy and to an Eastern Church, which continues to this very day, is reflected in Greek law and society.23 This duality makes Greece, which has never directly experienced the reformation or the enlightenment, and which is highly homogenous religiously, with around 97 per cent of the population Orthodox Christians, quite different from Western European countries.24 Thus, the state established Orthodox Christianity as its official religion in order to utilise religion’s unifying function, while at the same time subjecting the Church to state control through measures such as appointing the members of the Church’s governing body (the Synod) and subjecting all synodal decisions to government approval.25 Until recently the state enabled the Church to maintain its gatekeeper function by allowing it to retain exclusive control over marriages and divorces, but this changed in 1982 when a socialist government introduced civil marriage after much conflict and debate.26 B. Malaysia Malaysia is a state in which the Muslim religion and the Malay national identity are closely intertwined. Article 3 of the Malaysian Constitution defines Malaysia as an Islamic Federation. It states that: Islam is the religion of the Federation; but other religions may be practiced in peace and harmony in any part of the Federation.27
Around 40 per cent of Malaysia’s population is non-Muslim.28 While the country maintains a façade of interracial harmony and religious pluralism, clear preference is given in the Constitution and in federal law to the Malay ethnic group, who are Muslims.29 The Constitution defines ‘Malay’ as a person, who professes the religion of Islam, habitually speaks the Malay language, conforms to Malay custom and was born in Malaysia or Singapore, or is the issue of such a person.30 Because Islam is such an important com ponent of the Malay national identity the state exercises tight control over Islam and the ruling coalition is promoting a relatively moderate form of Islam (Islam Hadhari or civilizational Islam) in order to prevent the spread of more radical forms of Islam.31 The Malaysian Constitution guarantees individual religious freedom and the right of every religious group to manage its own religious affairs.32 However, the Constitution ibid 17. L Molokotos-Liederman, ‘The Greek ID Cards Conflict: A Case Study on Religion and National Identity against the Challenges of Increasing EU Integration and Pluralism’ 1 www.faithineurope.org.uk/idcards.pdf; Karagiannis (n 18) 147; Danopoulos (n 19). 24 Karagiannis (n 18) 147. 25 ibid 149. 26 Danopoulos (n 19) 18; Molokotos-Liederman (n 23) 2. 27 Federal Constitution of Malaysia 27 August 1957, Art 3. 28 FA Noor, ‘From Pondok to Parliament: The Role Played by the Religious Schools of Malaysia in the Development of the Pan-Malaysian Islamic Party (PAS)’ in FA Noor et al (eds), Madrasa in Asia: Political Activism and Transnational Linkages (Amsterdam, Federal Constitution of Malaysia Amsterdam University Press, 2008) 191, 192. 29 R Hirschl, Constitutional Theocracy (Harvard, Harvard University Press, 2010) 128. 30 Federal Constitution of Malaysia 27 August 1957, Art 160(2). 31 Hirschl (n 29) 128–29. 32 Federal Constitution of Malaysia 27 August 1957, Art 11(1)–(3). 22 23
508 Gila Stopler includes special provisions for Islam, which both give preference to Muslims and at the same time restrict their behaviour. For example, the Constitution allows state and federal law to restrict the propagation of any religious doctrine or belief among persons professing the religion of Islam.33
Consequently, it is forbidden to propagate non-Muslim religious doctrines to Muslims, and those wishing to propagate Muslim religious doctrines and beliefs to Muslims must obtain permission from state religious departments.34 The control over the propagation of Muslim religious doctrines granted to the government in the Constitution has enabled the government, among other things, to clamp down on dissident Muslim organizations and shut down their schools.35 In addition to establishing Islam as a state religion and attempting to unite the Malayan nation around the relatively moderate Islam Hadhari, Malaysia uses Islamic personal law as a gatekeeper for the Malayan nation. Thus, personal law in Malaysia clearly distinguishes between Muslims and non-Muslims for the purpose of marriage. While Muslims can only marry Muslims and must do so according to Islamic law, nonMuslims can marry in a civil marriage, irrespective of their religion.36 Furthermore, if a non-Muslim wishes to marry a Muslim they must convert to Islam before the marriage.37 In this manner Malaysian law is able to minimise intermarriage between Malays and non-Malays and maintain the distinction between citizens belonging to the Malay nation and citizens who do not, which stands at the core of Malaysia’s national identity. C. Turkey Unlike Malaysia Turkey is a secular country. When the Republic of Turkey was proclaimed in 1923, religion was banished from the public sphere in order to pave the way for modernization.38 Hundreds of religious Muslim schools (medreses) that were seen as incompatible with modern academic requirements were closed, and the state established compulsory schools that followed a national curriculum devoid of any religious instruction.39 Secularization was central to the Kemalist40 modernization project, and various ibid Art 11(4). AF Abdul Hamid, ‘Islamic Education in Malaysia’ (RSIS Mono No 18, 2010) 25 www.rsis.edu.sg/ publications/monographs/monograph18.pdf. 35 Eg in 1994 the Malaysian Government banned the Darul Arkam Movement and closed its schools. See ibid 60–63. 36 See the Malaysia Government’s official portal, ‘myGovernment’, ‘Marriage’: www.malaysia.gov.my/en/ relevant%20topics/society%20and%20life/citizen/family/marriage/pages/marriagemain.aspx. 37 See The Malaysia Government’s official portal, ‘myGovernment’, ‘Marriage Procedures between Muslims and Non-Muslims’: www.malaysia.gov.my/en/relevant%20topics/society%20and%20life/citizen/family/marriage/ proceduremarriagemuslimandnonmuslim/pages/marriagebetweenmuslimandnonmuslim.aspx. 38 B Agai, ‘Islam and Education in Secular Turkey: State Policies and the Emergence of the Fethullah Gulen Group’ in R Hefner and MQ Zaman (eds), Schooling Islam the Culture and Politics of Modern Education (Princeton, Princeton University Press, 2007) 149, 150. Nevertheless, the 1924 Constitution stated that Islam was the religion of the Turkish State and while this Article was removed in 1928, the principle of laicism (secularism) was inserted into the Constitution only in 1937. L Köker, ‘Religion, Education and the Turkish Constitution: A Critical Assessment’, Turkish Review, 14 October, 2010, 36–45. 39 Agai (n 38) 150. 40 Mustafa Kemal Atatürk was the founder of the Republic of Turkey. 33
34
National Identity, Religion and State 509
Muslim practices such as the pilgrimage to Mecca (hajj) were banned by law until 1947. Nevertheless, Islam played an important role in the Kemalist understanding of the Turkish nation due to Islam’s importance as the ‘unspoken bond’ that created the Turkish nation from a multitude of separate ethnic groups including Anatolians, Kurds, Caucasians, Albanians, Bosnians Tartars, etc.41 Thus, while on the one hand the early Kemalist state repressed Islam, on the other hand it promoted its own interpretation of Islam in order to legitimate its secular nationalism.42 The introduction of democracy in 1946 along with the realization that the ban on public expressions of religion is leading many to seek it via channels over which the state has no control, has led to the gradual reintroduction of religion into the public sphere and into the state system of education.43 Consequently, the notion of laicism, which initially meant a complete ban on Islam, was transformed to mean the control of religious expression by the state, and the following years saw a gradual increase in state-controlled Islamic education.44 In the 1980s the role of Islam in Turkish society has further strengthened. Islam was portrayed as a national trait of the Turks and as a source of social and moral stability, and obligatory religious courses were introduced in state schools.45 As a result, perhaps paradoxically, although Turkey is defined in its 1982 Constitution as a secular state,46 state control over Islamic education and its compulsory introduction into state schools are enshrined in the Constitution, which stipulates that education and instruction in religion and ethics shall be conducted under state supervision and control
and [i]nstruction in religious culture and moral education shall be compulsory in the curricula of primary and secondary schools.47
Furthermore, the secular state through its Department of Religious Affairs (DIR) controls 70,000 mosques and thousands of Qur’anic courses, and supervises private forms of religious activities.48 Thus, the state, whose control over religion is so tight that it even distributes the Friday sermons to the mosques around the country, is the most important religious player in Turkey.49 The Turkish Constitution expressly instructs the DIR to use Islam in order to foster national unity when it stipulates that [t]he Department of Religious Affairs . . . shall exercise its duties . . . aiming at national solidarity and integrity.50
Accordingly, the DIR promotes a relatively progressive form of Islam, which has been called ‘Turkish-Islamic-Synthesis’, and which is aimed at undermining Islamic influences Agai (n 38) 151. UC Sakallioglu, ‘Parameters and Strategies of Islam–State Interaction in Republican Turkey’ (1996) 28 International Journal of Middle Eastern Studies 231, 236; U Azak, Islam and Secularism in Turkey: Kemalism, Religion and the Nation State (London, IB Tauris, 2010) 12. 43 Agai (n 38) 152; see also D C¸akmak, ‘Pro-Islamic Public Education in Turkey: The Imam-Hatip Schools’ (2009) 45 Middle Eastern Studies 825, 829. 44 Agai (n 38) 152. 45 ibid 152–53; C¸akmak (n 43) 833. 46 Constitution of the Republic of Turkey 7 November 1982, Art 2. 47 ibid Art 24. 48 Agai (n 38) 153–54. 49 Azak (n 42) 12. 50 Constitution of the Republic of Turkey (n 46) Art 136. 41 42
510 Gila Stopler outside of state control and assisting in the project of national homogenization.51 While Islam performs an important unity function for Turkey’s national identity, it is not used as a gatekeeper to preserve the boundaries between Muslims and non-Muslims, and civil marriage is the only legally valid form of marriage in Turkey.52 The redundancy of religion’s gatekeeping function in Turkey is undoubtedly related to the fact that 98 per cent of Turks are Muslims. IV. RELIGION–STATE RELATIONS AND NATIONAL IDENTITY IN ISRAEL
A. Religion–State Relations in Israel Israel is defined in its Basic Laws as a Jewish and Democratic state. This definition is relatively new, and was adopted along with the two Basic Laws on human rights – Basic Law: Human Dignity and Liberty, and Basic Law: Freedom of Occupation. The origins of this definition can be traced to the Israeli Declaration of Independence, which states that Israel is to be a ‘Jewish state’, but at the same time that it will ensure complete equality of social and political rights to all its inhabitants irrespective of religion, race or sex; it will guarantee freedom of religion, conscience, language, education and culture.53
It is important to note that there is an ongoing and as of yet unsettled debate with regard to the exact meaning of the definition of Israel as a Jewish state in the Basic Laws. While some consider that the definition of Israel as a Jewish state constitutes an establishment of the Jewish religion in the state through the Basic Laws, others argue that the term ‘Jewish state’ should be understood merely as designating the character of Israel as the home of the Jewish people, where Jews realise their right to self-determination.54 The persistence of this debate is in itself proof of the extremely fuzzy lines between Jewish national identity and religion in Israel. The definition of Israel as a ‘Jewish state’ was enacted in the Basic Laws only during the 1990s. However, the Orthodox Jewish religion, which is represented in Israel by the Zionist Orthodox community and by the ultra-Orthodox community, has been established in the state from its inception in 1948 through laws granting legal status to Orthodox Jewish religious authorities in several areas. The most important area is that of personal status law, which was left to the exclusive control of the religious authorities.55 In addition, there are several other areas in which the Orthodox Jewish religion is Agai (n 38) pp 152–53, 156. Constitution of the Republic of Turkey (n 46) Art 174.4. 53 Declaration of the Establishment of the State of Israel, 1 LSI 7 (1948). 54 See A Levontin, ‘“Jewish and Democratic” – Personal Reflections’ in JE David (ed), The State of Israel: Between Judaism and Democracy (Jerusalem, The Israel Democracy Institute, 2000) 251 (taking a position against interpreting the term Jewish in the Basic Laws to include the Jewish religion). See CF M Alon, ‘Constitution by Legislation: The Values of a Jewish and Democratic State in Light of Basic Law: Human Dignity and Personal Freedom’ (1993)17 Tel-Aviv University Law Review 659, 668–70 (in Hebrew), republished in English in A Maoz (ed), Israel as a Jewish and Democratic State (Jewish Law Association Studies, 2011) 21 (taking a position supporting the inclusion of the Jewish religion in the term ‘Jewish’); R Gavison, Can Israel be Both Jewish and Democratic? or: Israel between Jewishness and Democracy (1 January 2011): www.gavison.com/ c935-book-can-israel-be-both-jewish-and-democratic-or-israel-between-jewishness-and-democracy. 55 Rabbinical Courts Jurisdiction (Marriage and Divorce) Law, 5713-1953, s 1-2. 51 52
National Identity, Religion and State 511
given preferential status by the state, either through statutes or through administrative decisions, which confer to it state power as well as money. Thus, the state has established a Chief Rabbinate and has given full control over it to Orthodox Judaism.56 The Chief Rabbinate is a powerful state organ which enjoys large budgets and which controls the religious services given by the state to the Jewish population.57 Some of the state and municipal institutions established and financed by the state and subject to the religious authority of the Chief Rabbinate are the Rabbinical Courts that deal with matters of marriage and divorce of Jews in Israel, the regional religious councils which deal with the supply of religious services – such as burial (public cemeteries in Israel are overwhelmingly religious), synagogues, kashrut (the requirement that food be kosher), etc – to Jews on a regional basis, and the conversion courts which deal with conversion to Judaism.58 B. Religious Personal Law’s Gatekeeper Function The most important aspect of the partial establishment of Orthodox Judaism is that all Jews in Israel are subject to Jewish religious personal laws. Members of other recognised religious communities, such as Muslims and various Christian denominations, are also subject to the personal religious laws of their particular religions.59 The current system of personal laws is essentially a continuation of the Ottoman millet system, which was based on the principle of community self-rule, giving each religious community full control over the personal status of its members, regardless of whether they wish to abide by the religious rules and even regardless of whether they consider themselves members of the community.60 This state of affairs has been maintained by the British Mandate and later by the State of Israel, which until this very day does not have a procedure for civil marriage. By recognizing only religious marriages conducted by a religious tribunal of a recognised religious community Israeli law in effect prevents people of different religions from marrying in Israel.61 One of the main reasons for the decision of the Israeli legislature to adopt this statutory scheme was its wish to prevent mixed marriages that would dilute the Jewish community, obscure the boundaries between the Jewish majority and the Arab (Muslim and Christian) minority and threaten Jewish national identity.62 When introducing the Rabbinical Courts Jurisdiction (Marriage and Divorce) Law in 195363 the Deputy Minister for Religious Affairs explained that one of the purposes of granting Chief Rabbinate of Israel Law, 5740-1980. H Lifshitz and G Sapir, ‘Jewish Religious Services in Israel – a Normative and Administrative Assessment in Preparation for Reform’ (2006) 23 Bar Ilan Law Studies 147–56. 58 Jewish Religious Services Law (Consolidated Version), 5731-1971. 59 The authority of the various religious communities was established through legislation from the period of the British Mandate that was later incorporated into Israeli law. See King’s Order in Council, 1922, sign 51(1). The detailed authority of the Jewish Rabbinical Courts is set out in the Rabbinical Courts Jurisdiction Law (n 55). The detailed authority for the Muslim Religious Courts can still be found in the King’s Order in Council, 1922, sign 52. 60 R Halperin-Kaddari, Women in Israel – A State of Their Own (Philadelphia, Penn Press, 2004) 227–28. 61 King’s Order in Council, 1922, sign 51; Rabbinical Courts Jurisdiction Law (n 55). 62 S Fogiel-Bijaoui, ‘Why Won’t There be Civil Marriage Any Time Soon in Israel?’ (2003) 6 Nashim: Journal of Jewish Women’s Studies & Gender Issues 28–34. 63 n 55 above. 56 57
512 Gila Stopler legal recognition exclusively to religious marriages was to exclude the possibility of mixed marriages that might result in the conversion of Jews to other faiths.64 Similarly, when it became known that the Muslim Sharia Courts in Israel were willing to marry Muslim men to Jewish women, the Ministry of Religious Affairs has instructed the Sharia Courts to refrain from conducting such marriages.65 Has the use of Orthodox Jewish personal law as a gatekeeper for the Jewish nation been successful? On the one hand the number of interreligious marriages and especially of interreligious marriages between Jews and Arab Muslims or Christians in Israel is very low. Thus, unlike Jews outside Israel whose interreligious marriage rates are often high, the rate of interreligious marriages of Jews in Israel is less than five per cent.66 It is likely that the lack of civil marriages in Israel has contributed to the dearth of inter religious marriages. Nevertheless, it is important to note that Israeli law does recognise civil marriages that have been held abroad, and although more than 10 per cent of Israeli couples marry civilly abroad, the number of interreligious marriages remains small.67 The fact that more than 10 per cent of the couples shun religious marriage and marry abroad is in itself proof that the use of Orthodox Jewish personal law as the gatekeeper for the Jewish nation has been only partially successful. A major reason for the Orthodox religious law’s limited success as a gatekeeper is the human rights violations that occur as a result of its imposition.68 One obvious violation is the violation of the right to freedom of conscience and belief of all those who do not wish to marry in a Jewish Orthodox religious ceremony.69 Many Jews are not religious and do not want to marry in a religious ceremony. In addition, within Judaism there are different streams, including for example Reform and Conservative Judaism, whose members wish to marry in ceremonies conducted by their own Rabbis, according to their traditions. Because Israeli law grants the power to marry Jewish couples exclusively to Orthodox and ultra-Orthodox Rabbis anyone who wants to marry in a Reform or Conservative ceremony and have her marriage recognised by the state is denied that right. Furthermore, while allowing only religious marriages restricts the rights of both men and women, it has a particularly severe impact on women’s equality rights. This is because the Jewish Orthodox personal laws as applied by Israeli Rabbinical Courts are highly patriarchal and give clear preference to men over women in most matters pertaining to the marriage relationship, especially the resolution of the marriage.70 Moreover, 64 Z Triger, ‘There is a State for Love: Marriage and Divorce between Jews in Israel’ in O ben Naftali and H Nave (eds), Trials of Love (Tel-Aviv, Ramot, 2005) 173–226 (in Hebrew). 65 P Shiffman, ‘Civil or Sacred: Marriage and Divorce Alternatives in Israel – A Necessary and Feasible Change’ (ACRI, 2001): www.acri.org.il/he/wp-content/uploads/2011/07/kadat-o-kadin.pdf, n 16 and accom panying text (in Hebrew). It is interesting to note that although according to Islam, the Sharia Courts can marry Muslim men to non-Muslim women, they cannot marry Muslim women to non-Muslim men. This is presumably because women will eventually convert to men’s religion, and, thus, become non-Muslim. 66 Z Triger, ‘Love and Prejudice: On the Paradox of the Phenomenon of Interfaith Marriages in Israel’ in D Barak-Erez et al (eds), Readings in Law, Gender, and Feminism (New York, Oceana Publications Inc, 2007) 733, 753–54 (in Hebrew). 67 See Israel Central Bureau of Statistics (CBS) data and compare the numbers of all marriages registered in Israel with the number of Israelis married abroad. See CBS websites: www.cbs.gov.il/www/population/ marrige_divorce/luach1.pdf and www.cbs.gov.il/www/population/marrige_divorce/marriage_all.pdf. 68 This claim holds true not just for Israel but for any country that imposes religious obligations on a population which contains a considerable number of people who are unwilling to accept such impositions. This segment of the population will try to find ways to by-pass these impositions. 69 See, eg F Raday, ‘On Equality’ in F Raday et al (eds), Women’s Status in Israeli Law and Society (Jerusalem, Schocken, 1995) 241 (in Hebrew). 70 Halperin-Kaddari, Women in Israel (2004) (n 60) 233–39.
National Identity, Religion and State 513
the subordination of women within the marriage has far-reaching effects on their ability to achieve equality in all areas of life.71 While partial attempts have been made to ameliorate this situation by means such as the establishment of civil family courts, which have parallel jurisdiction to religious courts in matters that do not pertain directly to marriage and divorce, these attempts do not weaken the hold that Orthodox Judaism has on marriage and divorce themselves.72 The use of Orthodox Jewish personal law as a gatekeeper for the Jewish nation leads many Jews to reject the option of marrying in Israel and to criticise the state for the human rights violations that the Orthodox monopoly creates. However, an even more serious problem for the state in its attempt to use Orthodox Judaism as gatekeeper for the Jewish nation is the fact that the strictness of Orthodox Jewish precepts prevents many people whom the state has an interest in integrating into the Jewish nation from being able to marry in Israel. During the 1990s the state has brought hundreds of thousands of Jews and people of Jewish descent from the former Soviet Union to Israel under the Israeli Law of Return,73 and has taken numerous steps since then in order to integrate them into the Jewish community for the explicit purpose of strengthening the Jewish majority in Israel and the Jewish nation. Contrary to the intentions of the state, the control of Orthodox Judaism over marriages of Jews in Israel is preventing a vast majority of these new immigrants from marrying other Jews, and even each other, and from fully integrating into the community.74 Thus, from the perspective of the state the use of Orthodox Judaism as a gatekeeper for the Jewish nation has backfired. It is worth noting that the Law of Return itself could arguably be seen as an example of the use of religion as a gatekeeper for the nation, since it defines its beneficiaries in relation to their Jewishness and grants only Jews the right to immigrate to Israel. However, unlike personal status law, which was entrusted to the Orthodox religious establishment and given a strictly Orthodox interpretation, the Law of Return, which makes only a partial use of the religious definition of who is a Jew, was never entrusted to the Orthodox establishment or given a strictly Orthodox interpretation. It is precisely the discrepancy between the definition of a Jew under the Law of Return and the definition of a Jew under Orthodox religious law that has been the cause of the inability of so many new immigrants, who were brought to Israel by the state, to marry within its borders.75 C. Religious Establishment and the Unity Function The Zionist movement that struggled for the establishment of the State of Israel was a national secular movement, led by Eastern European secular Jews who were staunchly opposed to the traditional religious leadership of Eastern European Jewish communities and aspired to create in Israel a modern national and secular Jewish state.76 Nevertheless, ibid 239–40. ibid 228–29, 233–35. 73 Law of Return, 5710-1950 (hereinafter: Law of Return). 74 Y Sheleg, ‘Not Halakhically Jewish: The Dilemma of non-Jewish Immigrants in Israel’ (Policy Paper 51, Jerusalem, The Israel Democracy Institute, 2004) 43–44 (in Hebrew). 75 A Yakobson, ‘Joining the Jewish People: Non-Jewish Immigrants from the Former USSR, Israeli Identity and Jewish Peoplehood’ (2010) 43 Israel Law Review 218, 219–20, 235–36. 76 Shapira, ‘Secular Politicians’ (1998) (n 2) 663–65. 71 72
514 Gila Stopler the leaders of the Zionist movement, who later became the leaders of the new state, decided to partially establish Orthodox Judaism in the state. One historical explanation to this decision revolves around the Zionist leaders’ need to get the support of Orthodox religious factions in the struggle to establish the state. In order to secure their support the secular Zionist leaders have come to an agreement with the Orthodox factions which is known as the ‘Status Quo’ and which outlined in general terms what later became the contours of the partial establishment of the Orthodox Jewish religion in the new state.77 A second explanation for the partial establishment of Orthodox Judaism was the secular Zionist regime’s need to enhance the legitimacy of the Jewish nation-state in the eyes of its Jewish citizens and of diaspora Jews, who were thought of as potential future citizens, by maintaining a connection with the Jewish past.78 Thus, secular national leaders, and especially the head of the Zionist movement and Israel’s first Prime Minister, David Ben-Gurion, decided to use the Orthodox Jewish religion as a legitimating and unifying force. They believed that in due time, after the establishment of the state, the religious element of Jewishness, which was initially used to buttress the emerging national element of Jewishness, would gradually disappear and be replaced by a fully-fledged Jewish nationalism.79 This assumption was closely related to the assumption that Orthodox Judaism was in itself in a gradual process of disintegration as an inevitable result of modernity, and that consequently no harm could be done by acceding to the demands of Orthodox religious leaders.80 Contrary to Ben-Gurion’s expectations, not only has Orthodox Judaism not dis appeared, but with the help of generous state funding, of its control over the religious establishment, and of its increasing political power, Jewish Orthodoxy and in particular its radical ultra-Orthodox branch, has experienced an unprecedented revival and expansion in Israel.81 However, far from promoting Jewish national unity, the existence of an extensive religious establishment in Israel and the continued exclusive Orthodox and ultra-Orthodox control over it, have been main causes for a deepening divide within Israeli Jewish society, which has come to be known as the religious–secular divide.82 In recent years the increasing radicalization of those controlling the religious establishment, and especially of the ultra-Orthodox, has put unprecedented pressure on Jewish national unity within Israel as well as on the relations between Israel and more liberal diaspora Jews, particularly those from the Reform and Conservative streams of Judaism. Due to lack of space I will focus only on one manifestation of this problem, which exposes the depth of the divide, not only between ‘religious’ Jews (ie Orthodox and ultra-Orthodox) and ‘secular’ Jews (who also include Reform, Conservative and some modern-Orthodox), but also between the Jewish nation-state and its own religious establishment.83 77 D Barak-Erez, ‘Law and Religion under the Status Quo Model: Between Past Compromises and Constant Change’ (2009) 30 Cardozo Law Review 2495, 2496–97. 78 Shapira (n 2). 79 Z Zameret, ‘Yes to the Jewish State, No to a Clericalist State: The Mapai Leadership and its Attitude to Religion and Religious Jews’ in M Bar-on and Z Zameret (eds), On Both Sides of the Bridge: Religion and State in the Early Years of Israel (Jerusalem, Yad Ben Zvi, 2002) 175, 177, 197–98 (in Hebrew). 80 ibid 198. 81 G Gorenberg, The Unmaking of Israel (HarperCollins, 2011) 165–91. 82 See, eg H Eden et al, Being Citizens in Israel: A Jewish and Democratic State (Tel-Aviv, Ma’alot, 2001) 299–316 (in Hebrew). This is an official high school citizenship textbook, which dedicates almost 20 pages to describing this divide and its causes, as a major internal divide within Jewish society. 83 In discussing the Jewish nation-state, I am referring to Israel’s civil state institutions, such as the government, the Knesset (the Israeli Parliament) and the Supreme Court.
National Identity, Religion and State 515
As already mentioned in the previous section, during the 1990s Israel has brought hundreds of thousands of Jews and people of Jewish descent from the former Soviet Union for the express purpose of strengthening the Jewish nation by increasing the Jewish majority. Because of the strong relationship between the Jewish religion and Jewish national identity the state has from the onset viewed it as an important national interest to help those immigrants who wish to do so to convert to Judaism. Since the Jewish Orthodox conversion process in Israel is protracted and complicated the state has attempted to facilitate the process by establishing Special Conversion Courts, which although staffed by Orthodox Rabbis and following Orthodox practice, were at the same time expected to be committed to the national mission of assisting the new immigrants who wish to convert. However, these courts turned out to be of very little use to those wishing to convert. Faithful to strict interpretations of Orthodox conversion requirements these courts have been very slow in approving conversions and in particular have been rejecting many candidates for conversion on the basis of their alleged failure to commit to observing Jewish religious commandments.84 While one could argue that religious courts should be free to apply their own understanding of religious law, it is important to note two things. First, this strict interpretation of Orthodox Jewish religion is not the exclusive interpretation and the official stance of the courts as published by the government is that a declaration of intent to observe Jewish religious commandments is sufficient.85 Second, and perhaps more importantly for our purposes, the failure of these state appointed religious courts to follow government policy, and to assist immigrants who wish to convert into Judaism and facilitate their inclusion in the Jewish nation, is a stark example of the paradoxical nature of the Israeli attempt to use the establishment of religion as a means for advancing national unity. Furthermore, the paradox does not stop here. Despite the fact that the Special Conversion Courts are the official state organs in charge of conversion, the Rabbinical Courts, which have exclusive jurisdiction on matters of marriage and divorce, and are controlled by ultra-Orthodox Rabbis, have several years ago decided that the conversions conducted by these courts were religiously invalid and have pronounced them retroactively void.86 While this does not affect the validity of the conversions in the eyes of the civil authorities of the state, it does have crucial implications in all realms governed by religious law and especially with respect to the status of the converts’ marriages and for the status of their children. Thus, the Rabbinical Court’s declaration that these conversions are retroactively void leads to the retroactive annulment of the converts’ marriages to Jewish spouses (since under religious law Jews cannot marry non-Jews) and, in cases where the converts are women, to the annulment of their children’s status as Jews (since the children have not been born to a Jewish mother). Recently, the Israeli Supreme Court published its decision 84 See, eg R Mendel, ‘Through the Agony of Conversion. I don’t want to be Jewish Anymore’, Ynet, 20 March 2012: www.ynet.co.il/articles/0,7340,l-3863646,00.html (in Hebrew) and the statistics brought there, as well as ITIM (Organization for Resources and Advocacy for Jewish Life), 2012 Conversion Report page, itim. org.il/_uploads/dbsattachedfiles/doch_giyur_2012.pdf (in Hebrew). 85 The state’s position on the issue can be found on the Israeli Government’s website – which provides information for potential converts. See ‘Israel Citizenship Guide’: www.gov.il/firstgov/topnav/situations/ spopulationsguides/ipnationality/ipgiur (in Hebrew). On the various religious opinions in this debate, see A Adrei, ‘Are We Not Responsible for Them? – More on the Conversion Debate’ (2000) 24 Akdamot 178 (in Hebrew). 86 HCJ 5079/08 Jane Doe v Rabbinical Court Judge, Rabbi Sherman 2-3 (25 April 2012), Nevo Legal Database (by subscription) (in Hebrew).
516 Gila Stopler in the case of two women whose conversion was retroactively declared void by the Rabbinical Courts.87 After the women’s appeal to the Supreme Court the Rabbinical Courts have reconsidered the women’s individual cases and have decided that their conversions were not void. The Supreme Court relied on this change in the Rabbinical Courts’ position in order to reject the petitioners’ request to issue an order declaring that the Rabbinical Courts have no authority to question a convert’s conversion.88 The Supreme Court’s refusal to issue such a declaration is unfortunate, because it leaves Rabbinical Courts with the discretion to retroactively declare conversions void in the future, thereby violating the basic rights of the converts and undermining the state’s attempt to facilitate conversions. V. CONCLUSION
As the Israeli example clearly demonstrates, using the establishment of religion as a means of facilitating national unity is a dangerous gamble, and the deeper the establishment, and the more the state relies on religion as a gatekeeper and/or as a unifying factor, the greater is the risk. Ben-Gurion has famously confided to the renowned Jewish philosopher Yeshayahu Leibowitz, who was a vociferous advocate of the separation of religion and state, that he (Ben-Gurion) established religion in the state so that the state could control religion.89 Having come from a strong collectivist Eastern European background, and not having been exposed to enlightenment liberalism, Ben-Gurion and other Zionist leaders of that time could not have anticipated the effects of the unhealthy combination of thick religious establishment and liberal ideals which is the marker of Israel’s religion and state relations for the past decades. In a bizarre twist on the liberal ideal of religious freedom the Israeli state, including its Supreme Court, has ended up respecting the freedom of religion of the ultra-Orthodox and Orthodox judges staffing its state religious tribunals much more than it respects the religious freedom of most of its citizens, and has consequently lost control over its own religious establishment. Putting aside for a moment the serious rights violations that a thick establishment of religion in the state can cause, and which deserve separate treatment, the comparative cases discussed in this chapter, and especially the cases of Malaysia and Turkey, reveal an awareness on the part of both these states, of an important truth that Ben-Gurion identified but which the State of Israel failed to implement, which is that tight state control over a powerful national religion is a prerequisite if a state intends to use it to promote its national interests. Perhaps paradoxically, such state control, illiberal though it may be, can be more conducive to promoting a moderate form of national religion than an allegedly liberal defence of the religious freedom of those who control the religious establishment.
ibid. ibid. 89 Y Leibowitz, ‘Separation of Religion and State – Summary and Prospects’ (1960) Beterem tpeople.co.il/ leibowitz/leibarticles.asp?id=8 (in Hebrew). 87 88
34 The Dilemmas of Identity in a Jewish and Democratic State: A Comparative Constitutionalist Perspective on Bakshi and Sapir, Gans, and Stopler SUSANNA MANCINI AND MICHEL ROSENFELD
I. INTRODUCTION
A
LL CONSTITUTIONS, NO matter how abstract or universal they may appear in their formulation and scope, must necessarily possess an identity1 as do nation-states that can only achieve unity and coherence as ‘imagined commun ities’.2 Moreover, constitutional identity and national identity are not identical though they are related and bound, at least to some extent, to be in conflict with one another.3 Although it is by no means exceptional for a nation-state to experience tensions and even contradictions within both its national and constitutional identities, Israel stands out in as much as its cleavages seem to run much deeper and therefore they loom as much more intractable. Not only does Israel confront a series of obvious profound tensions between its ‘Jewish’ and its ‘democratic’ commitments, but it is also deeply internally divided over its identity as ‘Jewish’. At one end of the spectrum are secular Israelis, who understand ‘Jewish’ as meaning a common culture shared by all Jews; at the other, the ultra-Orthodox who equate ‘Jewish’ with the strictest interpretation of religious dogma encountered within Judaism. Furthermore, in between these positions are several others such as those espoused respectively by the modern-Orthodox, more liberal denominations, such as Conservative and Reform Judaism (which lack official recognition by the state), and traditional approaches that regard ‘Jewish’ as some blend between culture and religion approached primarily from the perspective of tradition. 1 See M Rosenfeld, The Identity of the Constitutional Subject: Selfhood, Citizenship, Culture, and Community (London, Routledge, 2010). 2 B Anderson, Imagined Communities: Reflections on the Origin and Spread of Nationalism (London, Verso Books, 1991). 3 See Rosenfeld (n 1). An example of a conflict between the two identities would be a case where national identity strongly incorporates the polity’s majority religion whereas the constitution affords vigorous protection to minority religions.
518 Susanna Mancini and Michel Rosenfeld What further complicates matters is the unique and unprecedented way in which Israel has been building, and to a large extent continues to build, its identity as a nation entitled to the geographic location where it is presently situated due to the special history of the Jewish people. In this connection Zionism has been the major propelling force behind Israel as an imagined community as Chaim Gans’ chapter amply illustrates.4 Although as Gans emphasizes, there are many versions of Zionism,5 they all concur on a narrative whereby the Jewish nation built on some fusion between the Jewish people, the Jewish religion and Jewish culture had developed the core of its identity prior to establishing its current links – which started in the late nineteenth century – with the land it claims as its own and therefore as the legitimate space for it to fulfill its destiny as a fully-fledged modern nation-state. This is in sharp contrast with the building of nationhood in other nation-states. For example, in France nation-building was in full force during the 1789 Revolution, but this was in the context of a fully-developed state erected by the absolute monarchy, and in relation to a population with millenary ties with the territory within which the French nation was to seek to fulfill its destiny.6 Even in the United States, where nation-building has taken several generations and been animated by several waves of immigration, it is after entering American territory, and not before, that new immigrants became engaged in contributing to American nation-building. Another vexing difficulty confronting Israel’s endeavours to reconcile its Jewish and democratic nature relates to a seeming blurring between the country’s Jewish religion/ culture as the actual majority culture and religion within Israel and Zionist based or inspired arguments for special protection of Jewish culture and religion at the expense of minority cultures and religions within Israel. Indeed, the latter arguments often give the impression that, contrary to fact, Jewish culture and religion are presented as vulnerable within Israel as if they were minority ones. The above-mentioned difficulty could easily be solved if that were the whole story. But this is not the case to the extent that several of Israel’s neighbours are hostile to its existence and way of life. In short, within its borders Israel’s Jewish culture is majoritarian, but within the hostile Middle Eastern region in which Israel is embedded, Jewish culture is a minority one that is widely regarded with hostility. Does this justify Israel treating Jewish culture internally as if it were a minority one? The unique combination of factors discussed above certainly weighs on Israel’s ability to build a successful fully-shared national identity and a workable constitutional identity. Tellingly, in this latter respect, Israel has been unable thus far to adopt a written constitution in spite of many initiatives in that direction throughout its history. Nevertheless, through the adoption of Basic Laws and through judicial decisions, Israel has instituted a constitutional path that has yielded elements that may be used in the construction of a plausible and coherent constitutional identity. The three chapters respectively by Bakshi and Sapir, Gans, and Stopler are fully aware of the complexities surrounding Israel’s endeavours to build a broadly acceptable national and constitutional identity, and each of them, in its own way, provides critiques and suggestions, that both help clarify matters and suggest how Israel may progress beyond its current See ch 31 in this volume. ibid part II. 6 U Preuss, ‘Constitutional Powermaking for the New Polity: Some Deliberations on the Relations between Constituent Power and the Constitution’ in M Rosenfeld (ed), Constitutionalism, Identity, Difference and Legitimacy, Theoretical Perspective (Durham, Duke University Press, 1994) 143. 4 5
Dilemmas of Identity: Jewish and Democratic 519
quandaries. We basically agree with Stopler’s analysis and conclusions and will therefore only briefly comment on her essay. On the other hand, we disagree for very different reasons, with both Gans and Bakshi and Sapir. As we will discuss more fully, with respect to the latter, we do not think there can be any legitimate constitutional justification for wholesale exclusion of spousal reunification between Israeli Arabs and their Palestinian spouses living in the Territories. With respect to Gans, in contrast, our disagreement is much more nuanced. Nevertheless, we will argue that his egalitarian Zionism is ultimately unsatisfactory as it is not as egalitarian as he presents it to be. II. THE CONTRADICTIONS OF ISRAEL’S INTEGRATION OF STATE AND RELIGION
Modern constitutionalism certainly does not require separation between religion and the state as the American and French Constitutions prescribe. As Stopler indicates, Greece provides a vivid example of a robust state religion and of a close partnership between state and religion without exceeding the bounds of constitutionalism.7 Moreover, even certain secular countries, such as Germany, allow for significant intertwining between state and religion as exemplified by state sponsored religious education in public schools and state collection of taxes for distribution to various religious communities.8 The key from the standpoint of constitutionalism, however, is that the freedom of conscience, of religion, and from religion be afforded adequate protection. And that, as Stopler convincingly argues, requires ultimate state control over the state religion. Although Israel affords full recognition and collective rights to minority religions through its continuation of the millet system established by the Ottoman rulers over Palestine, it amply fails the test of modern constitutionalism in relation to its majority Jewish population.9 This is because, as Stopler systematically demonstrates, Israel has in essence yielded effective control of the state (Jewish) religion to the Orthodox and in some cases the ultra-Orthodox over whom it ultimately wields little, if any, control. This in effect leaves out all other versions of Judaism, including in some cases the modern-Orthodox, as well as all secular Jews. Accordingly, state deference to the Orthodox produces clear violations of freedom of, and from, religion as well as of freedom of conscience. To cite but one example, by giving the Orthodox Rabbinate exclusive jurisdiction over the marriage of Jews,10 and by not providing for any civil marriage, Israel deprives those Jews who wish to marry under the rituals of Conservative or Reform Judaism of freedom of religion; those secular Jews who object to a religious marriage of freedom from religion and freedom of conscience; and those Jews who wish to marry a non-Jew of basic liberty, privacy, and equality rights routinely vindicated under modern constitutionalism.11 Ch 33 in this volume, part IIIA. E Eberle, Church and State in Western Society: Establishing Church, Cooperation, and Separation (Burlington, Ashgate, 2011). 9 This is not meant to imply success or failure with respect to Israel’s Christian and Moslem religious minorities. Those issues remain beyond the scope of the present endeavour. 10 Strictly speaking, it may be possible for an Israeli Jew under the millet system to be married by state recognized Christian or Muslim authorities, but that would constitute a very rare circumstance that might require religious conversion or other onerous religious commitments on the part of the would be Jewish spouse. 11 cf Loving v Virginia 388 US 1 (1967) (prohibition against racial intermarriage held to violate due process and equal protection rights under the US Constitution). 7 8
520 Susanna Mancini and Michel Rosenfeld To the extent that the state gives the last word to the Orthodox on matters touching all Israeli Jews, Israel confronts the kind of constitutional problem and clash with constitutionalism that the much criticized 2012 Egyptian Constitution does. On the one hand, Articles 43 and 45 of that Constitution protect respectively freedom of belief and freedom of thought and opinion whereas the Constitution’s Preamble specifies that that [e]quality and equal opportunities are established for all citizens, men and women, without discrimination.
On the other hand, however, Article 219 of the new Egyptian Constitution enshrines [t]he principles of Islamic Sharia
as framed in accordance with credible sources accepted in Sunni doctrines.
Thus, it appears that both in the Israeli and the recent Egyptian cases, important constitutional rights, such as equality between men and women are ultimately bound to yield a religious view not shared by a significant percentage of the members of the country’s majority religion.12 Although the Israeli state defers too much to the Orthodox, it does not completely yield control to the latter on all issues with crucial religious implications. One problematic case discussed by Stopler is that concerning the Law of Return, 5710-1950.13 The Law of Return plays a key role in Israel’s Zionist agenda as it favours Jewish immigration and access to citizenship with the aim of boosting and perpetuating the Jewish majority in Israel. The Law of Return, however, adopts a much broader criterion to determine who is a Jew than does Orthodox Judaism. Thus, for example, a person whose father is Jewish, but whose mother is not, would qualify as a Jew under the Law of Return, but not under the criteria embraced by the Orthodox. Paradoxically, the partial state control over religion as it pertains to who is a Jew for purposes of the Law of Return does not mitigate the constitutional deficiencies prompted by lack of state control over religion in other areas, but it instead aggravates them. Going back to our example, the Israeli state spends resources to convince a foreigner with a Jewish father but not a Jewish mother to leave his country of origin, to settle in Israel and a short time later to obtain Israeli citizenship. After all that is done, the newly minted citizen who fully identifies with Israel’s Jewish majority decides to marry a fellow Israeli Jew. Soon thereafter, the new citizen is bound to find out that he has no right to marriage in his new country because for the Orthodox Rabbinate with a monopoly on marriage for Jews, he is not a Jew. Needless to add, this kind of disjunction can only exacerbate the frustrations in the quest for a viable constitutional identity and even that for a sustainable national identity. It is clear, as Stopler’s analysis unambiguously indicates that the current relationship between the Jewish religion and the state fails the modern constitutionalism test. It is also true, as many Israelis point out, including secular ones, that things are not as bad as they seem, as Jews who cannot meet Orthodox criteria for marriage or refuse to submit 12 Stopler emphasizes how giving the Orthodox a monopoly over Jewish marriage and divorce in Israel in effect deprives Jewish Israeli women of equality with men. See n 7, text accompanying nn 70–72. 13 ibid nn 73–75.
Dilemmas of Identity: Jewish and Democratic 521
to Orthodox rituals can marry abroad and have the Israeli state automatically recognize their marriage as valid within Israel.14 Nevertheless, even if one concedes, for the sake of argument, that recognition of foreign marriages provides some legal, political and social relief against the official monopoly given by the state to the Orthodox, this does not in any way alter the utter unacceptability of the monopoly in question from the standpoint of modern constitutionalism. To achieve a viable constitutional order and to construct a sustainable constitutional identity Israel will have to move beyond the current status of the relationship between the Jewish religion and the state. At the very least, Israel would have to provide for civil marriage and civil divorce. Beyond that, if the millet system is to be perpetuated, other denominations within Judaism, such as the Conservative and the Reform, as well as all Christian and Muslim denominations and all non-Abrahamic religions with a de minimis number of members15 should be given the same rights over the personal status and relationships of their members as do the Orthodox today. It would also be possible to move in other directions and emulate some of the countries with a state religion con stitutionally controlled by the state such as those described by Stopler.16 Moreover, it would even be conceivable for Israel to move towards a secular model either with significant entanglement between religion and the state as is the case in Germany or a more separationist one as in France or the United States. Ultimately, the problem does not lie with a lack of plausible or suitable constitutional models, but with intractable political conflict and deep splits among incompatible conceptions of national identity. Some have argued that Israel does not need a constitution because Jews already have the Torah. Also, some among the ultra-Orthodox do not believe in the legitimacy of the State of Israel as they are convinced that only God can restore the Jews to the Biblical land He once led them to in ancient times. Consistent with this view, Jewish nationhood, if at all separable from Jewish peoplehood, would have to be imagined against the Israeli state to the extent that the latter is considered to have emerged against God’s will. Others in the Orthodox community would like to enshrine Jewish law, the Halakha, in the fabric of Israeli constitutionalism while at the other end of the spectrum – but still within the confines of Zionism – some secular Jews would like to bestow a constitutional dimension on Jewish culture but relegate the Jewish religion to the private sphere. III. THE RELEGATION OF MINORITIES IN THE DEMOCRATIC PERIPHERY
Bakshi and Sapir discuss at length the link between the human right of all peoples to self-determination and the right of the Jewish people to statehood.17 They also discuss different liberal–national theories, in order to sustain that to preserve a people’s unique culture, an homogeneous (as opposed to a multinational) state is the optimal political form.18 This discussion is, however, of little help in sustaining their conclusions that 14 Y Merin, ‘The Right to Family Life and Civil Marriage under International Law and its Implementation in the State of Israel’ (2005) 28 Boston College International and Comparative Law Review 79. 15 It is obvious that it might be impractical to empower a religion with only a handful of members living in the country. A certain minimum of adherents, such as 1,000 or maybe even 5,000, and a certain minimum of internal organization may be set by law as a minimum threshold. 16 Ch 33 part III. 17 Ch 32 in this volume, part II. 18 ibid.
522 Susanna Mancini and Michel Rosenfeld limitations on the fundamental rights of Palestinians are necessary in, and compatible with, a Jewish and democratic state. In the first place, the same arguments that Bakshi and Sapir use to sustain Jewish statehood can be used to sustain the right of Palestinians to self-determination and to cultural preservation within the State of Israel. The right to self-determination has two dimensions: an external one, which enables groups to build their own state, and an internal one, that states must satisfy to comply with their international law obligations. The internal dimension of self-determination places an obligation on states to respect the cultural rights of minorities and to fully include them in the democratic process.19 Most importantly, however, the point is not whether the Jewish people are entitled to statehood in order to maintain its culture and identity. One can fully support Israeli statehood and its ‘Jewish’ character without accepting the idea that any nation-state can disregard the legitimate cultural and identity- related rights of sub-national groups, or, worse, establish a system of differentiated citizenship. Western political theories have traditionally neglected the claims of minorities. As Eric Hobsbawm explains, until the first part of the twentieth century, the ‘great nations’ were seen as the carriers of historical development, while smaller, less developed nations could only progress by abandoning their national character and assimilating with one of the ‘great nations’.20 Contemporary democracies, however, can no longer rely upon a high level of consensus on matters such as identity, religion and language as a prerequisite for their functioning. Contemporary constitutionalism postulates a close connection between freedom and equality. Virtually all consolidated democracies protect groups through a variety of legal tools. In Europe, one can roughly distinguish between ‘multinational’ states, such as Switzerland and Belgium, which reject the notion of minority and promote equality among all national groups, and states which recognize the exist ence of a majority/minority relationship and provide for the protection of minority rights through special clauses,21 which are enshrined in the Constitution and are part of the latter’s fundamental principles.22 Minority protection was listed as one of the pre requisites that candidate countries had to satisfy prior to their access to the European Union,23 and during the enlargement process the European Commission monitored very strictly the implementation of minority protection in Central and Eastern European countries. As a result, clauses protecting the rights of minorities are now a common feature of European constitutionalism. France, however, has remained immune to this trend. Despite the existence on its territory of a vast number of ethnic and linguistic
19 See S Mancini, ‘Secession and Self-Determination’ in M Rosenfeld and A Sajo (eds), Oxford Handbook of Comparative Constitutional Law (Oxford, Oxford University Press, 2012) 481–500. 20 E Hobsbawm, Nation and Nationalism Since 1780: Programme, Myth and Reality (Cambridge, Cambridge University Press, 1990). 21 See, eg Costituzione [Constitution] (Italy), Art 6 (‘The Republic protects linguistic minorities through special laws’). 22 See S Mancini and B de Witte, ‘Language Rights as Cultural Rights. A European Perspective’ in F Francioni and M Scheinin (eds), Cultural Human Rights (Leiden-Boston, Martinus Nijhoff Publishers, 2008), 247–84. 23 European Council in Copenhagen, 21–22 June 1993, Conclusions of the Presidency, DOC/93/3, s 7Aiii (‘Membership requires that the candidate country has achieved stability of institutions guaranteeing demo cracy, the rule of law, human rights and respect for and protection of minorities, the existence of a functioning market economy as well as the capacity to cope with competitive pressure and market forces within the Union. Membership presupposes the candidate’s ability to take on the obligations of membership including adherence to the aims of political, economic and monetary union’).
Dilemmas of Identity: Jewish and Democratic 523
minorities, the French legal system does not confer any special status to minority groups.24 Nevertheless consistent with all other European countries, France does protect sub-national groups from discrimination in that it confers all individuals regardless of group membership equal citizenship rights. In other words, irrespective of what constitutional model a state chooses to adopt in order to manage ethnic pluralism, it cannot depart from formal equality for all its citizens. Multinational states place all groups on an equal footing, which might result in the prevalence of collective interests over individual rights. This is typically the situation in Belgium and, to a lesser extent, in Switzerland, where equality among groups applies at the federal level, while linguistic territoriality is the rule within sub-national units. As a consequence, a French-speaking Belgian will be prevented from exercising his or her linguistic rights within Flanders, because the collective cultural rights of the Flemish population to preserve its language and tradition prevails. This is, however, a symmetric situation, because the same pattern applies in Wallonia to the detriment of Flemish speakers. Moreover, at the federal level, French and Flemish-speaking Belgians enjoy identical citizenship status. In states which recognize the existence of a majority/minority relationship, such as Italy, minorities are not placed on an equal footing with the majority as collectivities, individually, however, their members enjoy full citizenship rights. In Italy, Article 3 of the Constitution confers ‘equal social dignity’ and ‘equality before the law’ to ‘all citizens’, without distinction of, inter alia, race, language and religion.25 Moreover, in order to protect and promote the culture of sub-national groups, and help them overcome the disadvantage that being in a minority position almost inevitably poses, the Italian legal system grants them special collective rights, that range from cultural and educational rights to territorially-based political autonomy. Finally, in democratic states which recognize the existence of a dominant majority, members of minority groups are always granted the right to exit from their community and find a place within the majority group.26 In Israel, in contrast, there are no such rights, as individuals are automatically and permanently assigned to a group upon birth. The public sphere is made up of closed communities, with no open space between them. Those who do not (wish to) belong to an officially recognized group fluctuate in a sort of constitutional limbo, deprived of basic rights, such as the right to marry. No one is simply an ‘Israeli’ citizen: ‘Israeli’ must be accompanied by inclusion in one of the officially recognized groups. Bakshi and Sapir argue for a solution that is inconsistent, we believe, with all models of managing constitutional coexistence between different ethnic groups. They argue against a multinational state, in the name of the right of the Jewish people to enjoy their culture as the dominant group, which is per se a reasonable position, but they are not prepared to grant full equal citizenship rights to the members of the second largest ethnic group in the country. On the other hand, they seem to be willing to grant certain collective rights to this group, but not as an equal national group, but rather as a cluster of religious communities. This enables Israel to pursue its millet based system and thus 24 Significantly, in 1999, the Constitutional Council ruled that the European Charter for Regional or Minority Languages (5 November 1992) CETS 148, is not consistent with the French Constitution because it confers ‘specific rights to those speaking regional or minority languages within the territories in which such languages are spoken’ (Decision 99-412 DC, 15 June 1999). 25 Costituzione [Constitution] (Italy), Art 3 (‘All citizens have equal social dignity and are equal before the law, without distinction of sex, race, language, religion, political opinion, personal and social conditions’). 26 See J Raz, ‘Multiculturalism’ (1998) 3 Ratio Juris 193–205.
524 Susanna Mancini and Michel Rosenfeld preserve the homogeneity of its Jewish majority.27 The result of this unfortunate alchemy of collective and individual dimensions, however, is that Palestinians in Israel are not equal either as a sub-national group or as individual citizens. They are indeed separate and different. In this respect, the arguments by Justice Aharon Barak in his dissenting opinion in the Adalah v Minister of Interior case, are right on target.28 The Basic Law: Human Dignity and Liberty should certainly be interpreted as also including the right to establish a family, and this right encompasses the right of the Israeli citizen to receive his/her spouse in Israel and to carry on their joint life. Hence, the Citizenship and Entry into Israel Law (Temporary Provision), 5763-2003 (hereinafter: Temporary Provision Law), infringes on the right to equality of Arab Israeli citizens, notably on their constitutional right to establish a family. The Temporary Provision Law has a disparate impact, since in practice it is directed almost exclusively toward Palestinian-Arabs, and thus it discriminates against them, in comparison to Jewish citizens, in relation to their ability to realize their right to a family. The justification upon which the Temporary Provision Law relies for its discriminatory infringement of the rights of Palestinian-Arab citizens is security. Said differently, Palestinian spouses of Israeli citizens of Palestinian ethnicity ostensibly pose a security risk solely because of their ethnic background, irrespective of their personal conduct and of their connection to terrorist groups. Race-based profiling practices have been used (and justified)29 by various countries in the past to the detriment of different groups. After 11 September 2001, a clear tendency has emerged worldwide to use terrorist profiling based on national or ethnic origin and religion in the context of immigration controls, to target immigrants of Arab/Muslim background.30 Such practices are radically incompatible with basic human rights standards, and with the understanding of demo cracy as a system which treats each individual with equal respect and concern. They have repeatedly been condemned by international bodies;31 in the case at hand, Justice 27 See the insightful critique by M Karayanni, ‘Two Concepts of Group Rights for the Palestinian Arab Minority Under Israel’s Constitutional Definition as a “Jewish and Democratic State”’ (2012) 10 International Journal of Constitutional Law 304, who identifies two distinct concepts of group rights in respect to the Palestinian-Arabs: a ‘thin’ concept which pertains to them as a single national group, according to which the democratic values of the State of Israel will accommodate them with only a small measure of group rights, and only to the extent that this accommodation will not jeopardize the hegemonic status of collective Jewish interests, and a ‘thick’ concept that pertains to the Palestinian-Arabs as a cluster of religious communities, which strengthens the religious identity of this community. 28 HCJ 7052/03 Adalah v Minister of Interior 61(2) PD 202 [2006] (in Hebrew). 29 Korematsu v United States 323 US 214 (1944). 30 Eg the United States has implemented policies and practices in the field of immigration, designed to counter terrorism that single out certain groups of immigrants based on their country of origin or nationality and, indirectly through the choice of the targeted countries, their ethnicity and religion. Under the so-called ‘Voluntary Interview Program’, male immigrants, who had entered the United States after January 2000 and were not suspected of any criminal activity, were selected for ‘interviews’ only because of their age and country of origin, that is a country ‘in which intelligence indicated that there was an Al-Qaida terrorist presence or activity’. Although a list of such countries was never provided, about 8,000 men eventually interviewed were Arab and/or Muslim. See Migration Policy Institute, America’s Challenge: Domestic Security, Civil Liberties, and National Unity after September 11 (2003) 41. 31 See, eg UN Human Rights Council, Implementation of General Assembly Resolution 60/251 of 15 March 2006 entitled ‘Human Rights Council: Report of the Special Rapporteur on the Promotion and Protection of Human Rights and Fundamental Freedoms while Countering Terrorism’ (29 January 2007) UN Doc A/HRC/4/26, para 83 (‘Terrorist-profiling practices that are based on “race” are incompatible with human rights. Profiling based on ethnicity, national origin and/or religion involves differential treatment of comparable groups of people. Such differential treatment is only compatible with the principle of non-discrimination if it is a proportionate
Dilemmas of Identity: Jewish and Democratic 525
Barak’s dissenting opinion unmistakably shows how the harm race-based profiling does to human rights outweighs all security considerations. In her separate dissenting opinion, Justice Ayala Procaccia points to the demographic preoccupations that lurk behind the security argument.32 Bakshi and Sapir argue that a nation-state enables the majority to implement an immigration policy that coincides with the interest in preserving the national culture. Immigration policies vary from country to country. Some countries have particularly restrictive policies in this field. It is, however, not compatible with basic human rights standards for a state to apply a race-based classification in the field of immigration, targeting the race of its own citizens. The only potentially justifiable exceptions to family reunification are those added by the third version of the Temporary Provision Law, which broadened the prohibition on reunification of families and applied it also to petitioners whose country of origin was Iran, Lebanon, Syria, or Iraq, because these are countries hostile to Israel. But the West Bank is not comparable to a hostile country: it is occupied territory and, as the Israeli Supreme Court has unambiguously stated, the Palestinians in the West Bank have human rights which must ‘stand at the center’ of the occupiers’ considerations.33 The state model that emerges from Bakshi and Sapir’s analysis of the Adalah v Minister of Interior case is not consistent with any model implemented in a fully-fledged democratic state. In 2011, the Council of Europe Commission for Democracy through Law (the so called ‘Venice Commission’) expressed a number of concerns regarding the new Hungarian Constitution. In particular, the Commission pointed out that the latter seemed ‘to be premised on a distinction between the Hungarian nation and (other) nationalities living in Hungary. The Hungarian nation, in turn, also includes Hungarians living in other states’.34 The Commission was also troubled by the language of the Preamble, which ‘has been written in the name of “we the members of the Hungarian nation”, intimating that members of the “nationalities living with us” are not part of the people behind the enactment of the Constitution’. The latter, according to the Commission, should be seen as the result of the democratic will-formation of the country’s citizens as a whole, and not only of the dominant ethnic group. Therefore, the language used could/should have been more inclusive (such as, for example ‘We, citizens of Hungary . . .’).35
means of countering terrorism. Profiling practices based on ethnicity, national origin and/or religion regularly fail to meet this demanding proportionality requirement: not only are they unsuitable and ineffective means of identifying potential terrorists, but they also entail considerable negative consequences that may render these measures counterproductive in the fight against terrorism’). 32 Ch 32, text accompanying n 60. 33 HCJ 2056/04 Beit Sourik Village Council v Government of Israel 58(5) PD 807 [2004] (in Hebrew) para 34 (‘The Hague Convention authorizes the military commander to act in two central areas: one – ensuring the legitimate security interest of the holder of the territory, and the other – providing for the needs of the local population in the territory held in belligerent occupation . . . The first need is military and the second is civilian–humanitarian . . . The concern for human rights stands at the center of the humanitarian considerations which the military commander must take into account’). 34 According to the Preamble, ‘we promise to preserve the intellectual and spiritual unity of our nation torn apart in the storms of the last century’. This statement implies obvious historical references and should be read in conjunction with the Magyarország Alaptörvénye [Fundamental Law of Hungary], 1 January 2012, Art D, establishing Hungary’s ‘responsibility for the fate of Hungarians living beyond its borders’. 35 European Commission for Democracy through Law (Venice Commission), Opinion on the New Constitution of Hungary, Opinion no 621/201, paras 39–40.
526 Susanna Mancini and Michel Rosenfeld The exceptional situation of Israel, which we fully recognize, should not serve as a justification for the entirely one-sided reading of human rights provisions and requirements that Bakshi and Sapir provide. Dignity, equality and non-discrimination are fundamental human rights that pertain to all human beings, and must be guaranteed by all states, irrespective of how they manage ethnic pluralism in their constitutional system. The Jewish majority’s undisputable right and legitimate desire to preserve and promote its culture is not irreconcilable with equal respect and consideration for all. So long as Palestinian-Arabs will constitute a minority in a Jewish state, a balance must be drawn, in order not to exclude them from the enjoyment of fundamental rights and to deprive them of the status of full Israeli citizens. IV. HOW EGALITARIAN IS ‘EGALITARIAN’ ZIONISM?
There is a debate in Israel among proponents of different strands of Zionism and among all the latter and post-Zionists. As stressed by Gans, post-Zionists wish to strip Israel of its Zionist vestiges and endow it with the kind of individualistic constitutional culture and identity that is prevalent in France and the United States. One virtue of post-Zionism is that it seems particularly well suited to remove all badges of second-class citizenship that, as noted in part III above, currently afflict Israel’s Arab citizens. Gans argues, however, that post-Zionism is neither necessary nor desirable, and that Zionism need not be inequitable so long as the ‘egalitarian’ version of it that he has put forth is chosen over its rival ‘proprietary’ and ‘hierarchical’ versions.36 Whereas we believe that Zionism provides a plausible basis for constructing a viable constitutional identity for Israel, none of the arguments in favour of any of the three versions discussed by Gans strikes us as persuasive. Moreover, as we shall attempt to demonstrate below, Gans’s version of Zionism is certainly more egalitarian than are its rivals, but not egalitarian enough to satisfy the evolving demands of contemporary constitutionalism. We agree with Gans that ‘proprietary’ Zionism, which justifies Jewish entitlement to Israel on the basis of the Bible,37 amply fails any plausible criteria compatible with modern constitutionalism. We also agree with Gans that ‘hierarchical’ Zionism which is based on an ahistorical claim that every people has a right to self-determination and that therefore the Jewish people has such a right within the State of Israel, while less objectionable than ‘proprietary’ Zionism, is nevertheless constitutionally deficient as it would relegate Arabs within Israel to second-class citizenship.38 Gans goes on to claim that his ‘egalitarian’ Zionism by submitting Arab claims for a homeland on the same footing as their Jewish counterpart cures the egalitarian deficiencies of the two rival versions of Zionism and ought therefore to pass constitutional muster.39 Gans bases his own version of Zionism on three arguments advanced by the other two versions – namely, ‘(a) the historical link of the Jews to the Land of Israel; (b) the right to national self-determination; and (c) the Jewish history of persecution’40 – but he claims that by giving these a different interpretation he overcomes the problems that saddle the Ch 31, part III. ibid part IIA. 38 ibid part IIB. 39 ibid part IIC. 40 ibid. 36 37
Dilemmas of Identity: Jewish and Democratic 527
rival versions’ gloss on these arguments.41 Before proceeding any further, however, we should point out that none of these arguments, under any plausible interpretation, provides a sufficient constitutional basis for the implantation of Zionism as the core foundation of Israeli national and constitutional identity. Indeed, the historical link of the Jews to the Land of Israel, in the sense understood by Gans that Israel has always been central in the self-understanding and aspiration of the Jewish people, seems no more sufficient to entitle the Jews collectively to that geographic parcel than would a Basque claim to a territory with which that people strongly identifies but which encompasses lands that now belong in part to France and in part to Spain. Concerning the argument based on selfdetermination, moreover, even assuming that every people and every nation has a right to self-determination,42 why would the Jewish people have a compelling claim to rule Israel, a land they ceased ruling over and left for the most part two millennia ago? The pre- existence of a ‘nation’, and the existence of a relationship between the latter and a given territory are necessary conditions for territorial self-determination; they do, however, pose a first problem, in that various different groups might have equally legitimate claims on the same territory. Moreover, even if in principle all nations should be granted the right to their own state, in practice, to satisfy the aspirations of given nationalities through statehood, necessarily implies the frustration of other nationalities.43 Finally, regarding persecution, given a history of countless centuries of endless persecution culminating in the Holocaust, there is a strong case based on justice, fairness, and human decency that call for Jewish people to have their own homeland. But why should that be in what is today Israel? Is not there a much better case under restitutive and compensatory justice for a homeland for the Jews being located in a ceded portion of Germany’s territory? Or else, on land ceded voluntarily by a country, such as Australia or Canada, with huge virtually uninhabited spaces? Notwithstanding the above critiques of the three versions of Zionism discussed by Gans, there is an alternative version of Zionism that seems amenable to sound constitutional justification. That alternative version would draw on Israel’s existence as a nation-state for the past 65 years, as a member of the United Nations recognized by most other countries in the world, which happens to have a citizenry that is in its large majority Jewish. Whereas it is true that Jews have always had Israel close to their heart since the beginning of the diaspora, that they occupied significant portions of Palestine prior to independence in 1948, and that their yearning for self-determination as a nation may have been particularly acute given the horrendous history of persecution to which they were subjected, these do not add up to any valid constitutionally grounded claim for a Jewish state. On the other hand, the alternative version under discussion, which we may label as ‘State Democratic Majority’ Zionism, is constitutionally sound as should become plain if one keeps in mind that there are two dominant constitutional approaches that are compatible with the tenets of contemporary constitutionalism, as already discussed in part III. The first of these is the one centered on individualism and the second approach, in contrast, does give central importance to collective rights, whereas the individualist approach is incompatible with any plausible conception of Zionism and would therefore call for a post-Zionist constitutional identity and order, its rival collectivist ibid. But see the discussion above in part III. 43 E Gellner, Nations and Nationalism (Ithaca, Cornell University Press, 1983). 41 42
528 Susanna Mancini and Michel Rosenfeld approach would be compatible with the version of Zionism that we briefly outlined above. In Israel Jewish culture could inform the country’s constitutional identity and could frame a collectivist constitutional approach that could satisfy the precepts of constitutionalism much like the Belgian, Canadian and Swiss Constitutions do.44 Consistent with this, the three arguments that Gans advances as justifying the three versions of Zionism would have no impact on Israel’s constitutional identity, but they could certainly figure, at least in some interpretations, in the making and evolution of Israel’s national identity. The collectivist constitutional approach gives rise to a special obligation to vindicate the collective and individual rights of minorities, and particularly of significant and historically grounded minorities. Moreover, failure to abide by the obligation in question would deprive any project based on majority collective goals of constitutional legitimacy. Israel has Arab citizens that represent 20 per cent of the country’s population and that for the most part have lived within Israel’s current boundaries for many years before the country’s independence. In view of Gans’s claims for his ‘egalitarian’ Zionism, we will now briefly focus on whether its deployment would lead to the minimum necessary individual and collective needs of Israel’s Arab citizens to comply with the basic precepts of constitutionalism. Gans starts from the unimpeachable premise that ‘self-government and cultural preservation rights . . . should be granted equally to all homeland groups’.45 He then proceeds correctly to link equality to proportionality. If the state is obligated to provide the minimum feeding needs of indigent children then a family with six children ought to get double in state subsidies than a family with three children. But from these unassailable premises, Gans draws the wrong conclusion as he in essence asserts that because Jews are much more numerous than Arabs in Israel, it is the latter’s cultural and self- government rights that should prevail. What seems to lead Gans astray, is that whereas numbers matter both collectively and individually regarding equality, they do not matter in the same way. Spain should not force Catalans to use Spanish in their dealing with local authorities in Cataluña because the vast majority of citizens in Spain are native Spanish speakers. Forcing all citizens to deal with public authorities in French may be justified in France because the country has an exclusively individualistic constitutional approach. But once group rights are recognized, the group, and not the individual, becomes the subject of equality. This means that to the extent that Zionism commits Israel to a collectivist constitutional approach, Israel’s Arab minority ought to be entitled to both self-government and cultural collective rights. Now it is true that proportionality and numbers matter even in the context of group related equality. If Arabs represent only one-fifth of Israel’s population then they should receive proportionately less from the state to promote their cultures than Jews should, but that would not in any way detract from the Arab minority’s collective right to promote their own culture with state assistance. Gans, however, argues for individual equality for Israeli Arabs, but seems to leave no room for Arab group rights within Israel. And that represents a serious depravation of equality in the context of a collective constitutional approach. 44 Israel could even supplement a cultural collectivist constitutional approach with the constitutional incorporation of Judaism as its state religion by emulating the Greek example or that of any other country that grants the state ultimate control over religion. 45 Ch 31, part IIC.
Dilemmas of Identity: Jewish and Democratic 529
V. CONCLUSION
All three essays examined above highlight the constitutional, political and ideological obstacles confronting Israel in the context of constructing a sustainable constitutional identity that may mesh successfully with a constitutionally viable narrative that would promote a national identity incorporating essentials of Jewish culture and history, and even conceivably religion. These essays, moreover, suggest directly or indirectly how Israel may preserve a Zionist vision and agenda while endeavouring to satisfy the fundamental tenets of contemporary constitutionalism. Stopler underscores how Israel’s current handling of the relationship between religion and the state stands as an impediment to a successful constitutional handling of the matter, but at the same time she sheds light on how the country might overcome this difficulty through greater state control over religion. Bakshi and Sapir tackle a thorny human rights issue that closely split the Israeli Supreme Court, but in our opinion came to the wrong solution because they shuttle between the individualist and the collectivist constitutional approaches without remaining fully consistent to either. Gans makes the case that Zionism is consistent with contemporary constitutionalism, but his own ‘egalitarian’ Zionism comes up somewhat short in that it leaves Israel’s Arab minority with the same individual rights as the country’s Jewish majority, but not with collective rights comparable to those granted to minorities in contemporary constitutional democracies, such as Italy and most European countries. In the end, Israel seems to have multiple paths to viable combined and differentiated constitutional and national identities. From our vantage point at least, the most intractable question remains whether Israel will be able to muster the political will necessary to translate its constitutional potential into a fully-fledged constitution.
Index adjudication, see constitutional adjudication anti-formalism, 162–3 implications, 163 Basic Law: Freedom of Occupation (BLO), 2, 4, 51, 55 constitutional rights, 73–83, 396–7 development of constitutional law, 159, 177 limitation clause: proper purpose, 78 ‘notwithstanding’ mechanism, 20, 73–83 proportionality test, 21, 192–3 see also human rights; limitation clause Basic Law: Human Dignity and Liberty, 2–3, 177 dignity, 3, 279–81 interpretation of right to human dignity, 269, 277, 283–4, 315 justification of recognition, 275–7 legislator’s intention, 273–5 partial recognition, 277 recognition of unenumerated rights, 270–3 social and economic rights, 278–9, 315 societal values as human dignity, 277–8 legislator’s intention, 273–5 liberty, 3, 269 movement from and to Israel, 3 personal autonomy, 283 privacy, 3, 269 property, 3 protection from humiliation, 271, 277, 282–3 regulation of national security matters, 434 scope of protected rights, 102, 267–8, 297–9 security threats, 455 unenumerated constitutional rights, 269–70 freedom of expression, 269 freedom of religion, 269 interpretation of right to human dignity, 269, 315 justifying the recognition, 275–7 recognition of, 270 right to a clean environment, 270 right to a minimal standard of living, 269 right to an adequate standard of living, 270 right to education, 270 right to equality, 269 right to establish a family, 524 social and economic rights, 278–9, 315 societal values as human dignity, 277–8 see also human rights; limitation clause Basic Law: The Government, 443–4 limitations of Government’s powers, 433–4 national security, 432–4, 455 regulation of emergency situations, 432 Basic Laws, 2 basis for Israeli Constitution, 41–3
common attitude to, 42 ‘Constitutional Revolution’, 3, 100–1 constitutional scope, 102 ease of amendment, 20 impact on Knesset legislation, 47 institutional aspects of constitutional law, 2 interpretation by the courts, 48–50, 315 judicial review and, 48–50, 100–1 legal status, 41 limitation clauses, 160–1 national security, 431, 432–4 proportionality test, 161 redesigning, 68 role, 42, 43–7 see also Basic Law: Freedom of Occupation; Basic Law: Human Dignity and Liberty; Canada, 78–9, 86 constitutional deviations, 88–90 constitutional rights, 383 fiscal objections to social welfare rights, 342 judicial appointments, 97, 98 judicial review, 96, 123 legitimacy of state action, 258 ‘notwithstanding’ clause, 13–14, 66, 74, 75–6 proportionality, 190, 197, 199, 209, 213, 217–18, 220 social welfare rights, 342 use of foreign precedents, 155–6, 161 Central Elections Committee, 245 citizenship: Belgium, 523 conditions for: France, 484 United States, 484 eligibility for social welfare rights, 356 Europe, 358–9 Israel, 356–7, 362–6 South Africa, 357–8 family reunification, 494–5 France, 484, 525 Law of Return, 520 minority rights, 492 nationality distinguished, 483 political nationalism, 489 social citizenship, 356–9 Zionism, 520, 522, 526 civil and political rights: social and economic rights distinguished, 343 concentrated judicial review, 121, 127–8, 131–2 concept, 123–4 diffuse model distinguished, 128 institutional variations, 130–1 see also judicial review
532 Index consensus, 53–4, 59 constitutional adjudication, 94–9, 139–40 adjudication models: Common Law model, 96–9 Continental Europe, 95–6, 160–1 democratic accountability and, 95–6, 97–9 prevalence of foreign law, 151–2 United States, 94–5 courts, 140–2 European law, 160–1 hybridity, 145–7 judicial accountability, 91–4 judicial review, 142–5 political accountability, 140–2 use of foreign law, 151–2, 160–72 constitutional law, 41–3, 53, 56 constitutional adjudication, 94–9 democratisation, 57 judicial accountability, 91–4 judicial review, 43, 48, 57–8, 105–8 justification for, 112–20 legitimacy and, 108–12 majestic Constitutionalism, 73–90 ‘notwithstanding’ mechanism, 73–90 substantive constitutional law, 44, 48–50 United States, 67 use of foreign law, 156–63 constitutional models, 50-7 adjudication: Common Law model, 96–9 Continental Europe, 95–6, 160–1 democratic accountability and, 95–6, 97–9 prevalence of foreign law, 151–2 United States, 94–5 creating public dialogue, 9, 12–14 current Israeli model, 20–22, 51–2 defending basic values, 9–12, 15–16 democracy-reinforcing constitutionalism, 186–8 preservative constitutionalism, 186 serving as a silencing mechanism, 9, 14–15 transformative constitutionalism, 173–86 weaknesses, 53–4 see also constitutional adjudication; constitutional law; judicial review constitutional ‘moments’: fixed revolutionary, 305 non-revolutionary, 305 perpetual revolutionary, 305–6 ‘Constitutional Revolution’, 3, 47, 92, 120, 137–8, 178, 286, 403, 451 consensus, 172, 298 constitutional adjudication, 100–2 constitutional law, 156–63 expansion of judicial review, 116 majestic Constitution, 76–7 political accountability of the judiciary, 102 property law, impact on, 414 supremacy of Basic Laws, 206, 295 use of foreign law, 151–2 see also Basic Law: Freedom of Occupation; Basic Law: Human Dignity and Liberty; Basic Laws
constitutional rights: application to the judiciary, 380, 384–5, 420 Basic Laws, 397 Canada, 383 common law and, 384–5 constitutional ‘moments’: fixed revolutionary, 305 non-revolutionary, 305 perpetual revolutionary, 305–6 direct application model, 380, 381–3, 420 power of individuals as the state, 381 privatisation, 382 European Convention on Human Rights, 386 indirect application model, 380, 385–7, 419, 421 definition, 385 European Convention on Human Rights, 386 impact on interpersonal relationships, 399 Israel, 397–8 individual and, 379 interpersonal relationships and, 399 Israel, 397–8 legal variables, 425–6 models: application to the judiciary, 380, 384–5 direct application model, 380, 381–3 indirect application model, 380, 385–7, 419, 421 non-application model, 380, 383, 420, 424 non-application model, 380, 383, 420 objective values, as, 388 power of individuals as the state, 381 private law and, 379–80, 419–26 interpreting private law, 389–92 privatisation, 382 public law: non-application model, 380, 383, 420, 424 scope, 386 South Africa, 380 state and, 379 third parties, 379 United States, 384 constitutional supremacy, 206 constitutions: adoption, 1–2 aims, 28 compromises, 61–3 ‘Constitution of Conversation’, 67 role of judges, 68 ‘Constitution of Settlement’: United States, 68 constitutional supremacy, 206 content and form, 1, 2 creating public dialogue, 9, 12–14 defending basic values, 9–12 determination of legal rules, 43 entrenchment, 206 form and content: consensus, 2 imposed constitutions, 60 Israeli Constitution: Basic Laws, 41–3 definition as a ‘Jewish’ state, 487–9, 503 demographic goals, 498–501
majestic Constitution model, 81–3 national identity and religion, 503 Orthodox control, 511–13, 517, 519 proportionality doctrine, 192–4, 209 relationship between religion and the state, 503–5, 510–11, 519–21 rights of minorities, 522–3 role of a constitution, 41 Supreme Court and jurisdictional capacities, 124–5 judicial reinforcement, 206–7 ‘notwithstanding’ clause, 20–2 overriding the constitution, 22 rationales for: creating public dialogue, 9, 12–14 defending basic values, 9–12 serving as a silencing mechanism, 9, 14–15 serving as a silencing mechanism, 9, 14–15 United States, 68 see also Israeli Constitution contract law: application of human rights law, 416 contractual patterns, 415 nature of parties to a contract, 416–17 private law, 415–17 types of alliance: contract work contracts, 415 insurance contracts, 415 present contracts, 415 rental contracts, 415 sales contracts, 415 types of contract: relational contracts, 415 standard contracts, 415 creation of public dialogue, 9, 12–14 appropriate constitutional model, 18 –20 function of a constitution, as a, 12 weaknesses of the model, 12–13 creative/purposive interpretation: growing opposition, 25 judicial review, 26 Declaration of Independence, 1, 54, 430 Constituent Assembly, 1–2 definition of Israel as a ‘Jewish state’, 487, 510, 515 egalitarian Zionism, 479 elections for legislative authority hierarchical Zionism, 477 process of enacting a Constitution, 1–2 proprietary Zionism, 475 Provisional Council of State, 1–2 democracy-reinforcing constitutionalism, 186–8 demographic goals of the State of Israel: family unification, 498–501 legal reasoning, 498–9 legitimacy, 498 marriage, 511–13 diffuse judicial review, 121, 125–7, 132–6 concentrated model distinguished, 128 concept, 123–4 see also judicial review
Index 533 discrimination, 404, 469 access to medication and health services, 311–13, 319, 322, 328, 338, 374 administrative equality, 409–10 age discrimination, 342 citizenship rights, 362 constitutional equality, 409 Drittwirkung doctrine, 184 Israeli Arabs, 237, 496 Palestinian Arabs, 524 racial discrimination, 480–1 residency rights, 362 right to equality, 275 right to a family, 496 security threats: pre-emptive measures, 446–50 social rights entitlement, 343, 350, 359–61 Israeli Supreme Court (ISC) intervention, 187 Zionism, 476, 477 – 8, 526 see also family unification; human dignity dominant instrumentalist theory, 29 Drittwirkung doctrine (impact on third parties), 174, 379, 385 Germany, 182 Lüth case, 182 rise in Israel, 183–4 South Africa, 182 United States, 183 elitist dominance, 28, 30–31, 136 European conception of rights, 161 judiciary, 27, 133, 146 emergency constitution, see national security European Convention on Human Rights, 65, 98, 191, 199, 359 constitutional rights, 386 indefinite detention, 442, 468–9 right to a fair trial, 468 European Court of Human Rights, 164, 212, 345 closed material procedure (UK), 465 deportation issues, 468–9 eligibility for social welfare rights, 358–9 proportionality, 199–201 social citizenship, 358–9 European Court of Justice, 164, 190, 195 European Union, 164, 171, 191 consensus, 61 constitutional history, 61 proportionality of law, 195 protection of human rights, 64–5 minority rights, 522 family unification, 480 academic discourse, 499–500 background, 494 demographic goals of the State of Israel, 498–501 legal reasoning, 498–9 legitimacy, 498 judicial regulation, 496–8 legal reasoning, 494–8 legislative regulation, 494–5
534 Index family unification (cont): rights of minorities, 522–3, 525 right to a family life, 496 right to equality, 496 see also discrimination foreign constitutional law, 210–13 applications in ISC, 152 criticism in Israel, 165–72 bad comparisons, 167–9 cherry-picking by judges, 167 undermining original intent, 166 undermining sovereignty, 166 definition, 152 global constitutionalism, 212–13 negative impact: false sense of consensus, 172 interrelations between countries, 169 interrelations between courts, 169 ratching up effect, 169–70 transjudicialism, 170–2 prevalence in Israeli constitutional adjudication, 152–5 reasons, 156–65 prevalence in other jurisdictions, 155–6 Canada, 155–6 South Africa, 156 proportionality, 210 reasons for prevalence in Israel: ‘Constitutional Revolution’, 159–60 geopolitical isolation, 164, 211 historical background, 156–60 lack of historical case law, 163–4 non-formalist nature of constitutional law, 162–3 non-textual nature of constitution, 156–60 professional and academic relationships, 164–5 reliance on interpretation, 211 France, 61, 96, 299, 474, 506, 518, 528 constitutional judicial review, 123 cultural identity, 526–7 cultural neutrality and, 484 rights of minorities, 521– 3 freedom of speech, 241 offence to feelings and, 248–54 Gaza Disengagement Plan, 249–50, 343–4 Germany, 10, 521 constitutional court, 96 constitutional national security, 437 constitutional rights model, 386, 422 - 4 Drittwirkung doctrine, 182 ethno-cultural nationalism, 474 human dignity, 176 ‘imposed’ constitutions, 60 integration of state and religion, 519 judicial influence on Israeli law, 165 judicial review, 123 diffuse judicial review, 131 legitimacy of state action, 259 proportionality doctrine, 185, 190, 209, 220 illegitimate reasons, 257 three-pronged formula, 195, 213
social citizenship, 356 transformative constitutionalism, 173, 174, 175–6, 207–8 positive rights, 179 rhetoric of values, 176–7 global constitutionalism, 205–7 definition, 205 impact, 205 Greece, 506–7 ‘Harari’ decision, 2, 100, 106, 157, 177, 267, 295 health, see right to health Health Services Basket (HSB), 311–12, 320 determination of the basket, 324, 336 discretion of the basket committee, 324 human dignity, 279–84 expanding the scope, 297–9 Germany, 176, 182 Israel, 102, 178, 268, 275–7 Basic Law, 267–8 central pillar, as a, 269–70 humiliation and, 271, 277, 282–3 interpretation, 181, 270–3 legislator’s intention, 274 protecting societal values, 277–8 protection of autonomy, 282–3 scope, 268 social and economic rights, 278–9 South Africa, 177, 182 see also Basic Law: Human Dignity and Liberty; human rights human rights: administrative equality, 409–10 constitutional equality distinguished, 410 Basic laws and protection of, 47–8 clashes with private law: administrative equality, 409–10 constitutional equality, 409 constitutional norms, principles and rules, 407 constitutional relevance, 408 contract law, 415–17 hierarchy of rights, 408 nature of entities, 412–13 negative rights, 409 positive rights, 409 property law, 414–15 sub-constitutional value, 408 constitutional equality, 409 administrative equality distinguished, 410 individual rights and the courts, 37–8 Jewish right to a nation-state, 489–94, 521–6 judicial review and protection of, 140 minority rights, 30 private law, in, 401–18 unresolved clashes, 407 protection, 30 derogation in states of emergency, 436 right to culture, 489–93, 521–2 right to refrain from persecution, 493–4 right to self-determination: compliance with international law obligations, 522
right to build a state, 522 rights of minorities, 522–3 rights of minorities, 522–3 scope and weight, 30 waivable nature of rights, 410–12 see also right to health; social welfare rights Hungary: rights of minorities, 525 Israeli Constitution: Basic Laws, 41–3 definition as a ‘Jewish’ state, 487–9, 503 demographic goals, 498–501 family unification, 498–501 legal reasoning, 499–501 legitimacy, 498 marriage, 511–13 eligibility for social welfare, 356–7, 362–6 establishment, 1 Jewish nation state as a human right: justifications, 489 - 94 right to culture, 489–92 right to refrain from persecution, 493–4 judicial review, 41, 91, 139–47 arguments against, 27–9 arguments for, 29–30, 32 background, 106–8 constitutional adjudication models, 94– 9, 140–2 double hybrid model 128–9 ‘Constitutional Revolution’, 100–2, 131 diffuse judicial review, 121, 123–8, 132–7 hybridity, 128–9, 145–7 impact of Basic Law: Human Dignity and Liberty, 101 influence of common law system, 99–100 institutional considerations, 43–6, 130 instrumentalist / meritocratic justification, 30, 32 judicial accountability, 91–4, 140–2 justification, 91, 105–20, 142–5 legal reasoning, 115–20 legitimacy, 102–3, 108–14 opposition to, 25, 106–8 right to a hearing, 33–40 rules and discrepancies with Basic Laws, 48 social welfare cases, 361–2 Supreme Court and jurisdictional capacities, 124–5 United Mizrahi Bank v Migdal Cooperative Village, 101–2 majestic Constitution model, 81–3 grandeur and dignity, 79 national identity and religion, 503 Orthodox control, 517 marriage, 511–13, 519 proportionality doctrine, 209 overarching principle, 192–3, 194 tool in rights-protection issues, 192, 194 relationship between religion and the state, 503–5, 510–11, 519–21 gatekeeper function of religious law, 511–13
Index 535 marriage, 519 unity function of religious establishment, 513–16 rights of minorities, 522–3 role of a constitution, 41 judicial review, relationship with, 41 Supreme Court and jurisdictional capacities, 124–5 Israeli Supreme Court (ISC): criticism of growing powers, 25 doctrines of administrative law, 2 elitist dominance, 28, 31 holistic theory, 286–8 three-track democracy, 285 interpretation of substantive constitutional law, 44, 67 Basic Laws, 54–7 invalidating statutes: principle of equality, 54–5 judicial activism, 2, 25–7 national security issues, 452 - 3 judicial moderation and, 44 judicial review, 361–2 burden of proof, 362 jurisdictional capacities: High Court of Appeals, 124 High Court of Justice, 124–5 limiting governmental power, 2 protection of human rights, 2, 30–1 social welfare rights, 361–2 three-track democracy, 288–95 assumptions 290–2 legitimacy, 289 see also judicial activism; judicial review Japan, 60 Jewish Majority in Israel, 498 Jewish nation-state, see Israeli Constitution judges, see judiciary judicial accountability, 91–2 constitutional mechanisms, 92–4 appointment processes, 93 constitutional text, 92–3 judicial tenure, 93 judicial activism, 2, 25–7 anti-formalism, 162–3 court intervention in legislative decisions, 26 creative interpretation, 26 increasing, 162–3 legitimacy of military activities, 452 -3 judicial review, 41, 91, 107, 121–2, 124, 139–47 arguments against, 27, 29 anti-democratic, 27 elitist dominance, 28 arguments for, 29, 32 superior knowledge of judges, 29–30, 32 background to Israeli adoption of judicial review, 106–8 common law systems, 96–8 appointment of judges, 97 judicial accountability, 97 power of courts to review legislation, 98
536 Index judicial review (cont): concentrated judicial review, 121, 127–8, 131–2 concept, 123–4 diffuse model distinguished, 128 institutional variations, 130–1 legal reasoning, 136–7 constitutional adjudication models, 94, 140–1 Common Law model, 96–9 accountability, 141–2 Continental Europe, 95–6 double hybrid model, 128–9 United States, 94–5 constitutional methodology, 112–14 Constitutional Revolution, 100–2, 131 creative/purposive interpretation, 25–6 diffuse judicial review, 121, 125–7, 132–6 concentrated model distinguished, 128 concept, 123–4 legal reasoning, 136–7 hybridity of judicial review in Israel, 128–9, 145–7 impact of Basic Law: Human Dignity and Liberty, 101 implications, 39 influence of common law system, 99–100 institutional considerations, 43–6 concentration of power of judicial review in Israeli Supreme Court (ISC), 130 establishment of a constitutional court, 130 instrumentalist / meritocratic justification, 30, 32 judicial accountability, 91–4, 140–2 justification, 91, 105–20, 142–5 legal reasoning, 115–20 legitimacy of judicial review, 108–9 legitimacy, 102–3, 108–12 concept, 108 constitutional methodology, 112–14 counter-majoritarian difficulty, 108 opposition, 25, 106–8 primary legislation, of, 106–7 protection of human rights, 140 right to political participation, 27 right to a hearing, 33–7 justification for, 37–40 rules and discrepancies with Basic Laws, 48 social welfare cases, 361–2 burden of proof, 362 statutes, of, 27–33 Supreme Court and jurisdictional capacities, 124–5 United Mizrahi Bank v Migdal Cooperative Village, 101–2 see also constitutional adjudication judiciary: independence, 29 judges: election, 10 disadvantages, 11–12 full-life tenure, 69 interpretation of constitutional law, 67 limiting terms of service, 11–12, 69 superior knowledge, 28, 29–30, 32 transjudicialism, 170–2
legislative, relationship with, 43 protection of rights, 29 role, 44, 65 see also Israeli Supreme Court; judicial accountability, judicial activism Justice and Security Bill (UK), 463, 465, 466–7 Knesset, 2–3 accountability, 141 appointment of judges, 69 Basic Laws, 22, 41–2, 43, 44, 47–8, 51–2, 54–5, 77, 99–102, 294–5 emergency constitution and, 432–4 Constitution, Law and Justice Committee, 24, 131 duty to uphold Basic Laws, 44 emergency constitution, 434–5 Basic Laws and, 432–4 establishment, 2 family reunification, 494–8 human dignity, 267, 273–5 human rights protection, 174 constitutional rights, 299–301 protection of individual rights, 38 impact of basic laws, 47 judicial review, 106–8, 114, 119, 130–1 justiciability, 113–14 legitimacy and, 108–12 ‘notwithstanding’ clause, 21–2, 73–4, 78–82 constitutional amendment and, 83–9 role in national security, 448–50 activist approach, 450–3 detention of suspects, 448–9 evolving involvement, 448–50 legitimising pre-emptive measures, 446–50 limiting police powers, 448 minimalist approach, 446–8 regulating responses to security threats, 453–4 structure, 68–9 supremacy over courts, 20, 66–7 see also Israeli Constitution Labour Courts, 330–1 discrimination, 184 health insurance, 319–21 legal feminism, 31 legal systems, see constitutions legislative: accountability, 141 appointment of judges, 69 family reunification, 494–8 human dignity, 267, 273–5 human rights protection, 38, 174, 299–301 judicial review, 106–14, 119, 130–1 judiciary, relationship with, 43 promoting interests of the majority, 29–30 role in national security, 446–54 structure, 68–9 supremacy over courts, 20, 66–7 see also Israeli Constitution; Knesset
Index 537
legitimacy: constitutional methodology, 105, 120 conservative phase, 118–20 pragmatic phase, 112–14 radical phase, 114–18 judicial review: legal legitimacy, 108 political legitimacy, 109 sociological legitimacy, 108–9 legal reasoning, 112–20 military activities, 453–3 state action, 257–60 Canada, 258 Germany, 259 South Africa, 258–9 Limitation Clause, 192, 358, 382, 402–3, 496, 498 Basic Laws, 77, 119, 160–1 constitutional proportionality, 225, 384, 387 eligibility conditions for social welfare rights, 363–5 private law, 392 unenumerated constitutional rights, 301 majestic Constitution model: ‘notwithstanding’ mechanism, 81–83 reasons for adoption, 81 discourse and education, 82–83 normality in the face of the international community, 82 stability and consensus, 81–2 Malaysia, 507–8 national identity: constitutional identity distinguished, 517 function of religion, 504 gatekeeping function, 503 unity function, 503, 504–5 gatekeeping function of religion, 503 exclusionary, 505–6 Greece, 506–7 Malaysia, 507–8 Turkey, 508 - 10 unity function of religion, 503 inclusivity, 504–5 National Labour Court, 184 see also labour courts national security, 440–1 Basic Laws, 431, 432–4 comparative law, 435–40 Germany, 437 historical examples, 435 South Africa, 437 Spain, 438 United States, 438 constitutional foundations, 430 Basic Laws, 431 judicial precedents, 431 legislation, 430–1 constitutional regulation, 429–30 emergency constitution, 432, 434–5 emergency powers, 448 international law and, 440–1
judicial review, 439–40 legislative role, 454 constitutional law, 455–6 deliberation, 456–7 policy considerations, 456–60 preventing effective address of threats, 460 regulation of the fight on terror, 458 scrutinising government activity, 457–8 symbolic aspects, 458–60 limitations of Government’s powers, 433–4 power to declare emergency situation, 441 ordinary laws, 441 sunset provisions, 441–3 ‘temporary’ laws, 441 regulation of emergency situations, 432 Basic Law: Human Dignity and Liberty, 434 response to security threats, 445–6 determining legitimacy, 446–50 martial law, 446–7 minimalist approach, 446–8 regulation of, 453 - 5 role of the legislature, 445–50, 454 right to counsel, 449 see also human rights; Israeli Constitution negative effectiveness test, 233 proportionality, 233 ‘notwithstanding’ mechanism: amending powers, 74 Basic Law: Freedom of Occupation (BLO), 73–4 Canada, 13–14, 74 empowered legislature, 75–6 constitutional amendment, 83–5, 87–8 constitutional deviation, 85–7, 88–9 constitutional crisis, 73–4 constitutional deviation, 73, 85–7 temporary nature, 88–9 dialogical potential, 66 Israel, 20–2 Basic Law: Freedom of Occupation (BLO), 73–4 majestic Constitution model, 76–81 reason for introduction, 74 majestic Constitution model, 76–81 background, 76–81 addition of mechanism to the Basic Law: Freedom of Occupation (BLO), 77–9 amendments to the Basic Law: Freedom of Occupation (BLO), 80–81 ‘Constitutional Revolution’, 76 creation of Basic Laws, 77 right to import kosher meat, 73–4 offence to feelings, 241–54 reasons for not recognising, 248–54 parliamentary sovereignty, 2, 84, 97 Britain, 64–5 judicial review versus, 142–4 Peel Partition Plan 1937, 476 People’s Council, 1 pluralism of Israeli society, 66 political discourse, 16, 24
538 Index positive rights, 179–82 enforcement in Israel, 181–2 Germany, 179–80 rejection of concept by United States, 180 South Africa, 180 post-war constitutionalism, 1, 161 Post-Zionism: concept of nationality: ‘Israeli’ versus ‘Jewish’, 483 concern for equality, 474 self-determination and equality, 481–2 concern for freedom, 474 self-determination and freedom, 482–5 cultural neutrality of the state, 484–5 freedom of association, 474 Jewish right to self-determination in Israel: rejection of, 473–4 religious neutrality of the state, 484 self-determination and equality, 481–2 rights for Israeli Arabs, 483 self-determination and freedom, 482–5 Zionism distinguished, 473 preservative constitutionalism, 174, 186–7 see also constitutions private law: application in human rights, 401–3 background, 403–5 Canadian non-application model, 404 development, 403–5 Israeli ‘indirect application model’, 405 judicial implementation, 402 justification, 402 South African ‘indirect application model’, 405 US model of ‘application to the judiciary’, 404–5 constitutional rights and, 379–400 developing private law, 393–7 developing common law, 393, 394 negative rights, 395–6 new legislation, 393–4 positive rights, 395–6 proportionality, 394–5 human rights in, 401-18 unresolved clashes, 407–17 autonomy of the free will, 407 inherency of human rights, 407 indirect application model, 380, 385–7, 419 definition, 385 European Convention on Human Rights, 386 impact on interpersonal relationships, 399 Israel, 397–8 interpretation of private law, 389–92 good faith, 391 objective aspects, 390 public policy, 390–1 validity, 392 value concepts, 390 judicial implementation, 402 Kestenbaum case, 405–6, 417–18 negative aspects, 392–3 proportionality, 419 see also contract law; human rights; property law
property law: concept of property, 414 nature of the property: interchangeable property, 414 personal property, 414 private law, 414–15 relationship between parties, 414–15 proportional representation, 68 proportionality doctrine, 161, 189–90, 202–3, 213–14, 220, 255–7 ancillary powers doctrine, 217 appropriate purpose test, 228–9, 238 consistency with values of the state, 228 subjective purpose, 228–9 balancing, 239 advantages, 240–1 categorical approach versus, 240 exclusionary reasons, 240 freedom of speech versus offence to feelings, 241, 248–54 methodology, 255 principled balancing, 202 Canada, 190, 197, 209, 213, 217, 220 common law, 394–5 concept, 190, 203, 214 constitutional proportionality, 225–7 infringement ‘to an extent no greater than is required’, 225, 235–7 deficiencies, 239–42 detention, 232–3 European Convention of Human Rights: tool in rights-protection issues, 191 European Court of Justice, 190, 220–1 formulae, 195 European Court of Human Rights, 190, 197 means versus aims test, 196 formula, 190, 194–6, 203 doctrine distinguished, 194, 214 three-pronged formula, 195 freedom of speech: offence to feelings, 239–42 Germany, 185, 190, 209, 220 three-pronged formula, 195, 213 guiding considerations, 236–7 Israel, 209 overarching principle, 192–3, 194 tool in rights-protection issues, 192, 194 least restrictive means test, 234–5 legal reasoning, 190, 201–2 legitimacy, 257–60 assumption of legitimate objective, 260 means to ends analysis, 260–1 limitation clauses, 192–3, 218–19 morality, 215 nature of, 190 overarching principle, 190–1, 202 tool in rights-protection issues, 191, 202–3, 216 necessity test, 184, 226, 234–5, 238 comparing alternative measures, 234 disadvantages, 234 offence to feelings, 242–7 private law, 394–5
proceduralising proportionality analysis, 261 proportionality stricto sensu, 184, 194–5, 201, 226–7, 235–7, 238 value-based decisions, 226, 228 rationality, 184 reasoning device, as a, 190, 201–2 relationship between tests, 237 scrutiny tests, 194 social welfare cases, 344–6, 373 constitutionality of eligibility conditions, 362–3 South Africa, 209 subtests of, 226–7, 231–7 suitability test, 226, 231–4 disadvantages, 233 third parties, 261–3 three-pronged formula, 195–6, 213 development of doctrine, 197–201 ‘to an extent no greater than is required’, 225, 230–7 suitability test, 231–4 transformative constitutions, 184–6, 208–9 transnational comparisons, 221–2 Canada, 190, 197, 209, 213, 217, 220 Germany, 185, 190, 195, 209, 213, 220 South Africa, 209 United Kingdom, 190, 198 Zimbabwe, 190, 198 United Kingdom, 190, 198 Zimbabwe, 190, 198 Rabbinical Courts: marriage, 512–13 religion and the nation-state: function of religion, 504 gatekeeping function, 503 unity function, 503, 504–5 gatekeeping function of religion, 503 exclusionary, 505–6 Greece, 506–7 Israel, 510–16 Malaysia, 507–8 Turkey, 508–10 unity function of religion, 503 inclusivity, 504–5 right of property, 240 see also property law right to a hearing, 33–7 adjudication process and, 37–40 complaint against infringement of a right, 34 courts and, 38 duty to provide an explanation, 33, 35, 37 duty to reconsider, 33, 35, 37 infringement of prima facie rights, 35 justification for infringement of a right, 33–4 justification for judicial review, 37 opportunity to raise a grievance, 33, 35, 37–38 procedural complaint, 34 resolution of disputes: existence of a prima facie right, 33 weight of the right, 33 right to free speech, 157–9 see also offence to feelings
Index 539 right to health, 311-15, 329–31 access to drugs and services, 319–26 insurers’ claims, 319–21 Labour Courts, 319–21, 330 bypassing ‘bare minimum’ approach: duty to exercise discretion in certain cases, 320 duty to provide unconditional emergency care, 320 dynamic interpretation of the HSB, 320 judicial review of decisions to exclude services, 320 co-payments, 326 constitutional status, 322–3 discrimination, 311–13, 319, 322, 328, 338, 374 equality, 318 evidence-based medicine, 337 financial access, 318 Israeli (cochlear implant) case, 311–13 Israeli (world cup) case, 312–13 Israeli health system, 316 access to drugs and services, 319–26 financing, 318–19 Health Services Basket (HSB), 311–12, 320 Israeli Supreme Court: intervention against government, 321–6 judicial role, 337 legal justification, 320–1 litigation, 321–6 public / private programmes, 326–9 role, 329–31 scope, 323–4 rolling procedures, 42, 46–7 principle of equality, 46–7 separation of powers, 214 proportionality and, 219 social welfare rights and, 335–7 silencing mechanism, 9, 14–15, 22–3 appropriate constitutional model, 17–20 social and economic rights: civil and political rights and, 315, 343, 345, 348 interpretation of ‘human dignity’, 315 see also human rights; right to health; social welfare rights social citizenship, 356 Europe, 358–9 Israel, 356–7 South Africa, 357–8 social welfare rights, 333–5, 346–8 budgetary implications, 339, 342, 348 co-payments: judicial review, 345 constitutionality of social rights, 361–2 discrimination, 343, 359–61 education, 338–9 effective enforcement, 374–5 eligibility conditions, 350–1 constitutionality, 353–4, 362–6 human rights theory, 352–3 social policy, 351–2 verification of entitlement distinguished, 354–5
540 Index social welfare rights (cont): emergence of consensus, 369 fiscal objections, 341–6, 346, 348 Canada, 342 South Africa, 341, 342, 373 judicial enforcement, 334, 337, 340, 374–5 competence, 337–9 emergence of consensus, 369 Israeli Supreme Court, 344–6, 361–2 judicial review, 361–2 burden of proof, 362 proportionality, 373 qualifying conditions: eligibility and verification methods, 354–5 social citizenship 356 - 9 verifying entitlement, 355 remedies: dialogic remedies, 371–2 executive and plaintiff dialogues, 372 judicial, legislative and executive dialogues, 372 separation of powers in, 335–7, 370–1 Ireland, 335 United Kingdom, 335 United States, 335 social citizenship, 356 European Court of Human Rights, 358–9 Israel, 356–7 South Africa, 357–8 see also human rights; right to health; social and economic rights South Africa: constitutional national security, 437 constitutional rights, 380 Drittwirkung doctrine, 182 eligibility for social welfare rights, 357–8 fiscal objections to social welfare rights, 341, 342, 373 human dignity, 177 legitimacy of state action, 258–9 proportionality doctrine, 209 Social Assistance Act 2004: eligibility requirements, 357–8 social citizenship, 357–8 social welfare rights, 341, 342, 373 sovereignty: negative impact of use of foreign law in constitutional adjudication, 166 parliamentary sovereignty, 2, 84, 97 Britain, 64–5 judicial review versus, 142–4 Spain: constitutional national security, 438 Special Conversion Courts, 515 state’s duty to act, 398 scope of duty, 398–9 substantive constitutional law, see constitutional law Supreme Court, see Israeli Supreme Court (ISC) terrorism, see national security three-track democracy, 288, 302 assumptions, 290–2 collective will as the basic norm, 290
dynamic legal systems, 290–1 incompetence of Parliament, 292 universal versus founding formulae and values, 291–2 Israel as a, 292–5 Basic Law: The Knesset, 292–3 doctrine of unwritten fundamental principles, 293–4 implied eternity clause, 292 origin and nature of Basic Laws, 294–5 legitimacy, 289 UCRs (unenumerated constitutional rights) doctrine, 295–7, 302 transformative constitutionalism, 207–9 concept, 173, 175–6 Constitutional Revolution, 178 constitutional values, 207, 208 doctrines, 176 rhetoric of values, 176–9 Drittwirkung doctrine, 182–4 Germany, 174–5, 176, 179–80, 182, 208 impact on third parties, 182–4 judicial activism, 173, 207 positive rights, 179–82, 207–8 enforcement in Israel, 181–2 rejection of concept by United States, 180 private relations and, 207, 208 proportionality, 173, 184–6, 207, 209 South Africa, 174, 176, 180, 182, 208 value orientated discourse, 207 see also constitutions transjudicialism, 170–2 Turkey, 508–10 unenumerated constitutional rights (UCRs), 269–70 expanding the scope of ‘human dignity’, 297–9 freedom of expression, 269 freedom of religion, 269 individual autonomy, 298–9 interpretation of right to human dignity, 269, 315 justifying the recognition, 275–7 recognition of, 270 judges authority, 285 right to a clean environment, 270 right to a minimal standard of living, 269 right to an adequate standard of living, 270 right to education, 270 right to equality, 269 social and economic rights, 278–9 societal values as human dignity, 277–8 see also human rights unlawful detention, 462–4, 467 - 8 United Kingdom: asset-freezing, 464–5 closed material procedure (CMP), 465–7 judicial response to terrorist threats, 462–4, 467–9 legislative response to terrorist threats, 462, 463–7 pre-charge detention, 462–4 proportionality doctrine, 190, 198 special advocates, 465–6 terrorism, 461–2 asset-freezing, 464–5
closed material procedure (CMP), 465–7 House of Lords, 464 judicial response, 462–4, 467–9 legislative response, 462, 463–7 pre-charge detention, 462–4 see also constitutions; judicial review United Nations: counter-terrorism measures, 464–5 Resolution 194 of the General Assembly, 1 United States: citizenship: requirement for a knowledge of English, 484 compromises, 62–3 ‘Constitution of Settlement’ 68 constitutional history, 60–1 constitutional law, 67 constitutional national security, 438 constitutional rights model, 380, 421–2 Drittwirkung doctrine, 183 Marbury v Madison, 94, 113, 145, 160, 168 originalist movement, 167 rejection of concept of positive rights, 180 proportionality, 214–16 public values and private law, 208 Yeshiva students, 254, 299 equality, 46, 119
Index 541 Zimbabwe, 190, 198 Zionism, 473–5 egalitarian Zionism, 474, 526–8 interpretation of the right to self-determination, 480 primacy of the Land of Israel, 479 ethno-cultural principle, 473 hierarchical Zionism, 474, 526 Jewish hegemony, 478 primacy of the Jewish people, 479 regime of inequality, 478 interpretations: egalitarian, 474, 479–81 hierarchical, 474, 477–8 proprietary, 474, 475–6 proprietary Zionism, 474, 526 collective rights of Arabs, 475, 527 collectivist moral ontology, 475 individual rights of Arabs, 475, 528 primacy of Jewish People, 479 proprietary justice, 475 rights over territory of Israel, 475 religion and ethno-culturalism, 473 right to ethno-national self-determination, 473 struggle for a secular state, 513–14 see also Post-Zionism