II. What is semi-liberal constitutionalism?

My first task is to offer a definition of semi-liberal constitutionalism. It is tempting to think of liberal constitutionalism and illiberal regimes in dichotomous terms – while a liberal constitutional regime is one that is fully committed to the protection of liberal individual rights, an illiberal regime is one that is fully committed to buttressing the power of the state and its rulers and harnessing it to further their ideology, their party or their own personal gain, with complete disregard to the individual rights of the citizenry.^{Footnote 13} However, many countries in the world do not fall into these dichotomous ideal types. As Tushnet observes, even within liberal constitutionalism there are big variations which he attributes to three possible sources: reasonable disagreement on accepted liberal concepts, such as the scope of freedom of speech; historical contingency, such as the example given by Tushnet, of the limitations on the right to vote in 20th century Switzerland; and genuine ‘shortfalls’ in the realisation of liberal commitments, which he describes as ‘reasonably enduring ‘‘failings’’ of a generally liberal constitutionalism to realize some of liberal constitutionalism’s normative commitments’.^{Footnote 14} He observes that shortfalls may be self-stabilising and enduring in a way that ‘ought to lead us to identify the system as neither liberal nor illiberal nor authoritarian’.^{Footnote 15} This suggests that a system which is considered liberal, but has enduring and stable shortfalls comes close to what I would term a semi-liberal constitutional system.

Tushnet uses the example of the selective right to vote in Switzerland (where women were denied the vote until 1971) as an example of a liberal regime with a historical contingency.^{Footnote 16} In my view Swiss constitutionalism until 1971 cannot be described as liberal constitutionalism (even with a historical contingency). When a country withholds the right to vote from half of its population it can hardly be considered liberal. At the same time, since Switzerland was committed to granting its citizens many other liberal rights it is also difficult to consider its regime at that time as illiberal. I would argue that a more accurate definition of such a regime is to define it as a semi-liberal regime. As I will explain, one important advantage in conceptualising such regimes as semi-liberal is that such conceptualisation alerts us to the fact that human rights conflicts in such regimes should not be analysed within the regular liberal paradigm because such an analysis would skew the result. Instead, their analysis should be adjusted to reflect the semi-liberal characteristics of the particular setting and their effects on the rights of the parties involved. Thus, for example, in Switzerland a constitutional amendment giving women the right to vote was put to national referendums and denied repeatedly by a majority of the voters until finally approved in 1971.^{Footnote 17} In a liberal setting where the right to vote is universal the results of such national referendums would be entitled to a high degree of legitimacy, at least prima facie. However, in a semi-liberal context where women are barred from participation in the electoral process such referendums are inherently flawed and their results cannot legitimate any forms of restriction on women. Acknowledging the semi-liberal character of pre-1971 Swiss constitutionalism exposes the power differentials inherent to it and helps alert us to its illegitimate nature, while its designation as liberal masks it.

In several articles written in the mid-1990s Graham Walker offers what he terms the new mixed constitution as the appropriate constitutional structure for post-communist countries. Walker posits that these countries, as well as others, reject what he views as two alternative absolutisms, the illiberal politics of ethnicity, religion, or moral solidarity on the one hand, and the liberal politics of individual autonomy rights on the other. He therefore suggests that such countries should draft and adopt mixed constitutions.^{Footnote 18} The ideal mixed constitution according to Walker is a principled mixture that is neither wholly liberal nor wholly illiberal, and that is wisely constructed so as to prevent the collapse of the polity into either its liberal or non-liberal poles.^{Footnote 19}

Walker argues that the mixed constitution is morally superior not only to its wholly illiberal and tyrannical counterpart but to its fully liberal counterpart as well.^{Footnote 20} He accuses the liberal constitution, with its commitment to individual autonomy and to state neutrality, of trivialising pluralism, and of displacing substantive commitments to religious, ethnic and moral communities and replacing them with a uniform commitment to commercialised individualism, rather than tolerating them.^{Footnote 21} Walker gives the complete banishment of religion from the public sphere as the prime example of the flaws in liberal constitutionalism.^{Footnote 22} Contrary to Walker’s claim, which seems to be based on the US experience, it is important to note that the complete banishment of religion from the public sphere is the exception in liberal states rather than the rule. Many liberal states have much closer ties to religion than Walker seems to concede, including liberal states with established religions and with state-supported churches.^{Footnote 23} Thus, it is questionable whether the criterion of state neutrality can serve to distinguish between a liberal constitution and a mixed constitution. With respect to the distinction between a liberal and a semi-liberal constitution I will argue that the true fault line between these two types of constitutionalism is not their neutrality towards religion, but their interest in protecting human rights and their ability to do so.^{Footnote 24}

Two examples of mixed constitutions which Walker discusses rather generally are Israel and the Native American Nations.^{Footnote 25} In his discussion of Israel Walker puts much emphasis on Israel’s lack of religious/ethnic neutrality and on its commitment to its Jewish nature. However, Walker is much less specific regarding the ways in which Israel’s commitment to its Jewish nature violates individual rights. Walker’s critique of the liberal constitution, as well as his preference for the mixed constitution, stems from his support for the prerogatives of the community over the rights of individuals.^{Footnote 26} He believes that the mixed constitution is an ideal model for what Breslin terms a communitarian constitution.^{Footnote 27} Walker’s definition of the mixed constitution and his analysis of it are more focused on its advantages for the community than on its adverse effects on individuals. Conversely, my definition of semi-liberal constitutionalism and my analysis of it are more concerned with the semi-liberal constitution’s impact on individual rights.^{Footnote 28} Consequently, while Walker and I both see Israel as an important example of a mixed/semi-liberal constitution, our definitions of these types of constitutionalism and their analyses are quite different.^{Footnote 29} In my discussion below I will first define what I believe are the essential characteristics of semi-liberal constitutionalism and then delve more deeply into Israel’s semi-liberal constitutional structure and its effects on individual rights.

In my definition a semi-liberal constitutional system is a system that meets two conditions: first, it exhibits a simultaneous dual normative commitment to liberal rights and principles and to other values or interests that result in enduring and significant restrictions on some of these rights; second, this dual normative commitment is constitutive and is expressed in the basic elements of the system.

I will call the first condition – that a semi-liberal constitutional system exhibits a dual normative commitment to liberalism and to non-liberal values – the dual normativity condition. One of the fundamental ideas of liberal theory is that all individuals are free and equal and must be treated as such.^{Footnote 30} A second and related fundamental principle of political liberalism is the priority of the right over the good, which, according to Rawls, means that institutions that form a part of the basic structure of a constitutional democratic regime may not impose ideas of the good on unwilling free and equal persons.^{Footnote 31} While liberal constitutionalism must be committed to these two principles, in a semi-liberal constitutional system these principles are adhered to only partially and only by some state institutions and in some areas of law. At the same time, a semi-liberal constitutional system will be committed to one or more values or interests, such as a religious, communitarian or socialist dogma, or pressing economic or security needs, in a manner that intentionally rejects the liberal commitments to persons as free and equal and the priority of the right over the good in the areas in which these values and needs take precedence.

The second condition, that the dual normative commitment to liberal rights and non-liberal values is constitutive and expressed in the basic elements of the system, I will call the basic elements condition. The basic elements of the system are the elements that are part of the structure of government (in each of its three branches) and that hold coercive authority over individuals. The evaluation of these basic elements is conducted not only on the basis of their express legal authority but also on the basis of their concrete functioning. Thus, if some institutions that form part of the basic elements of the system are committed to liberal values while others are committed to non-liberal values, the system will be a semi-liberal constitutional system.^{Footnote 32}

As an example I suggest looking at the constitutional system in Israel. By international standards such as the Freedom House Report Israel is considered a free liberal democratic country with an aggregate score of 78 out of a hundred and a freedom rating of 2.5 out of 7 (where 1 is the most free).^{Footnote 33} However, my claim is that defining Israel as a liberal democracy is misleading. A closer look at the Israeli constitutional system reveals that there are three significant aspects in which this system is marked with illiberality: the first deeply illiberal aspect is the continued occupation of the territories occupied in the 1967 war, although since Israel has not claimed sovereignty over most of this territory it may still passably be argued that this aspect should not be considered part of Israel’s constitutional system; the second aspect has to do with some Israeli laws, such as those pertaining to land appropriation, that have different effects along ethnic lines and disparately impact the rights of the Arab citizens of Israel;^{Footnote 34} the third aspect of illiberality is expressed in some features of the relations between religion and the state in Israel. These three aspects differ from each other in the extent and mode of their illiberality and each deserves a separate treatment, but for the purposes of this article I want to focus only on the third aspect – the relations between religion and the state – and explain what I mean by semi-liberal constitutionalism using this aspect as an example. In particular I want to focus on the fact that in Israel religious courts have an exclusive jurisdiction in matters of marriage and divorce and there is no possibility of civil marriage. While there are other facets of religion–state relations in Israel that deviate from the common liberal model, this is the one facet that to my mind is a clear mark of Israel’s semi-liberal constitutionalism in the area of religion–state relations.

Whereas most liberal democratic states do not define themselves in reference to their ethnic or religious character, 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 Basic Law: Freedom of Occupation.^{Footnote 35} However, the origins of this definition can be traced to the Israeli declaration of Establishment. This declaration 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’.^{Footnote 36} Long before the enactment of the Basic Laws, the Jewish religion in its Orthodox version has been partially established in the state through laws granting legal status to Orthodox Jewish religious authorities in several areas, the most important of which being that of personal laws.^{Footnote 37} All Jews in Israel are subject to Orthodox Jewish religious personal laws. At the same time, 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.^{Footnote 38} This state of affairs has first been instituted at the period of the Ottoman rule over the Eretz-Israel/ Palestine region and has been maintained by the British mandate and later by the state of Israel. As will be discussed in more detail below, the imposition of the religious personal laws of the various religious communities on all residents, and the lack of an alternative civil marriage, constitute a violation of the right to freedom of conscience and belief, as well as a violation of the rights of women who are subject to the discriminatory patriarchal religious laws of the various religious communities.^{Footnote 39} I will claim that due to the deeply illiberal nature of Israel’s personal laws, Israel’s constitutionalism cannot be considered a fully liberal one, and I will show how the illiberal nature of these laws interacts with the liberal aspects of the constitutional system and how this interaction affects the rights of all those subject to the Israeli legal system.

The centrepieces of the liberal facets of Israel’s constitutional system are Basic Law: Human Dignity & Liberty and Basic Law: Freedom of Occupation, which afford constitutional protection to several fundamental rights.^{Footnote 40} These basic laws on human rights do not explicitly protect the right to freedom of religion and conscience, or the right to equality or to free speech. However, the Israeli Supreme Court has ruled that all these rights are protected by Basic Law: Human Dignity and Liberty, in so much as they are all an essential part of the right to dignity, which is expressly enshrined in the Basic Law.^{Footnote 41} In most contexts the Supreme Court has given these rights, as well as other rights enshrined in the Basic Laws, a liberal interpretation based on the commitment of the Israeli legal system to individual liberty and to substantive equality.^{Footnote 42} A similar liberal commitment is also evident in various human rights laws such as the Equal Employment Opportunities Law, the Prohibition of Discrimination in Products, Services, and Entry into Places of Entertainment and Public Places Law, the Freedom of Information Law etc.^{Footnote 43} Thus, it would be fair to say that in general Israel is committed to maintaining a form of liberal constitutionalism that upholds liberal rights in all areas in which this commitment is not explicitly set aside in order to pursue other constitutive values.

However, as is evident in the area of religion–state relations, alongside its liberal commitment Israel is simultaneously committed to a form of non-liberal theocratic constitutionalism which is set to promote non-liberal religious ideals in particular areas of law. The legal framework determining religion–state relations in Israel is not enshrined in any of its basic laws; rather it was determined by regular legislation enacted prior to the Basic Laws. Because the Basic Laws have constitutional status, regular legislation is generally subject to the Basic Laws. However, Basic Law: Human Dignity and Liberty contains a provision stating that its enactment shall not affect the validity of any laws enacted prior to it.^{Footnote 44} The result is that although the Supreme Court has interpreted Basic Law: Human Dignity and Liberty as affording constitutional status to fundamental rights such as equality, freedom of religion and freedom of speech, this status cannot be used to challenge the validity of the pre-existing legislation that gives state power to religion, regardless of the violation of basic rights that it entails. The most important legislation for our purposes are the various pieces of legislation that subject all matters of marriage and divorce in Israel to the religious courts of the various religious communities, that are staffed by religious judges who adjudicate cases according to the community’s religious law.^{Footnote 45} The content of said religious law is determined by each religious judge on the basis of his own understanding of the applicable law.^{Footnote 46} The largest and most powerful courts are the Jewish religious courts, the Rabbinical Courts, which are mostly staffed by Ultra-orthodox judges whose interpretation of Jewish religious law is most conservative and highly discriminatory against women. All Jews in Israel, whether religious or not, and regardless of their own preferences, are subjected to the exclusive jurisdiction of the Rabbinical Courts on matters of marriage and divorce. ^{Footnote 47} Evidently, the legal arrangements just described are illiberal and constitute a violation of the rights to equality and to religious liberty. Nevertheless, they form a basic element of the Israeli constitutional system that is fully theocratic and immune to any liberal challenge. My claim is that a constitutional system which combines basic constitutional elements such as the ones just described, that are theocratic (or otherwise non-liberal), with basic constitutional elements that are liberal should be considered a semi-liberal constitutional system.

One could wonder why not consider the state of affairs just described merely as constituting a shortfall in a liberal constitutional system, in line with Tushnet’s suggestion, as opposed to viewing it as a state of affairs that in and of itself renders the Israeli constitutional system semi-liberal. Where should we draw the line between shortfalls in a liberal constitutional system and a semi-liberal constitutional system? While there certainly can be hard cases, I think that there are three important factors to take into account when evaluating each system: the intent behind the system, the structure of the system, and the extent of the deviation from liberal principles. In all these parameters the legal arrangements pertaining to marriage and divorce in Israel point to the deliberate creation of a semi-liberal constitutional system and not merely to a liberal constitutional system with a shortfall. The intent of the Israeli parliament when enacting the legislation establishing the various religious courts was clearly to remove this area of law from the purview of Israeli civil law, entrust it wholly in the hands of religious judges and take it out of the scope of liberal rights review. This is evident in the structure of the laws of marriage and divorce in Israel. For example, nowhere in the Rabbinical Courts Act or in any other civil law is there any mention of the substantive content of the applicable religious law, and it is entrusted wholly in the hands of respective religious judges, who are authorised to determine what the applicable religious law is and apply it as they see fit.^{Footnote 48} The Israeli Supreme Court has held that it has no jurisdiction to review the decisions of state-authorised religious courts as long as they operate within the boundaries of their given authority.^{Footnote 49} The Supreme Court has further held that it will not intervene in decisions of state-authorised religious courts even in cases of a clearly mistaken application of the relevant religious law as long as these courts apply religious law within their given authority.^{Footnote 50} Thus, the Supreme Court sees itself as structurally barred from dealing with those matters of marriage and divorce, which the civil legislature entrusted at the hands of religious courts applying non-liberal religious law. As we will see below, the Supreme Court does intervene in decisions of religious courts in matters that are subject to civil law, but the very distinction between these two types of matters points to the structural barrier between the liberal and non-liberal facets of the constitutional system. Finally, given the above, the extent of the deviation from liberal principles is clear. The situation as it currently stands is that the Israeli constitutional system has carved out an important area of law which significantly affects the lives of all individuals – the laws of marriage and divorce – and insulated its core from liberal oversight. Thus, I would argue, on the basis of this deviation alone, and without engaging the other deviations mentioned above, Israel cannot be considered to have a liberal constitutional system but a semi-liberal one. It should be noted that even in the area of religion–state relations the laws of marriage and divorce are not the only deviation from liberal constitutionalism, although they are the most serious one. Arguably some of the other deviations in this area, such as the various legal restrictions pertaining to Jewish religious dietary laws may be considered merely shortfalls and would not in and of themselves justify defining Israel’s system as a semi-liberal constitutional system in the way the laws of marriage and divorce do.

This distinction also serves to highlight an important difference between Walker’s definition of the mixed constitution and my definition of semi-liberal constitutionalism.^{Footnote 51} For Walker Israel’s lack of Neutrality with respect to the Jewish religion is a major component of its mixed constitution. Under my definition the lack of neutrality in and of itself does not make Israeli constitutionalism semi-liberal just as the established church in Britain does not make British constitutionalism semi-liberal. Certain forms of religious establishment can be reconciled with liberal constitutionalism and considered either as reasonable disagreements, as historical contingencies or as shortfalls under Tushnet’s definition.^{Footnote 52} What makes Israeli constitutionalism semi-liberal in the area of religion–state relations is the imposition of religious discriminatory personal laws on all citizens in disregard of their rights to religious freedom and to equality. Such intentional and pervasive violations of rights that form part of the basic structure of the system and that arise from a commitment to a non-liberal value system by some state institutions and in some areas of law render the system semi-liberal.

Arguably, another constitutional system that is semi-liberal due to its personal laws is the Indian system, and especially due to its extensive deference towards minority personal laws. On one hand, the Indian constitution defines India as a secular democratic state that guarantees full equality to women.^{Footnote 53} However, on the other hand, out of deference to communal values the constitution has left the religious personal laws of the various religious communities in India in place, only calling on the state to endeavour to secure a uniform civil code for all citizens.^{Footnote 54} Subsequently Hindu personal law was codified by the state, granting Hindu women rights which they did not enjoy before, but minority personal laws, including Muslim personal law were left intact. Furthermore, although the Indian Constitution states that any laws that are inconsistent with the constitution are void, the Indian Supreme Court has held that religious personal laws should remain in force until replaced by alternative laws.^{Footnote 55} Consequently, at the same time that the Indian Constitution guarantees women full equality, some religious personal laws such as the Muslim personal law that are considered part of the law of the land in India uphold discriminatory practices such as polygamy and the triple talaq.^{Footnote 56} This has significant adverse effects on minority women in India and even the very limited attempts by the Indian Supreme Court to intervene in favour of women’s rights, such as the famous Shah Bano case, have been thwarted by the Indian legislature.^{Footnote 57} Recently the Indian Supreme Court has issued a very narrow ruling holding that the practice of triple talaq is unconstitutional, but many other discriminatory practices attributed to the core of religious personal laws are still in place.^{Footnote 58} I would argue that the pervasive nature of the violations of women’s rights that are caused by the basic structure of the Indian constitutional system, which simultaneously guarantees women the right to equality but shields religious personal laws that violate this right from almost all scrutiny, renders the Indian constitutional system a semi-liberal one.^{Footnote 59}

III. Engaging with semi-liberal normativity

Very few scholars have tried to engage with semi-liberal normativity or have even acknowledged that liberal theories do not fit well in settings that are not fully liberal. One such scholar is Will Kymlicka who in his more recent discussions of the internationalisation of multicultural theory has observed that the theory of liberal multiculturalism as it currently stands is easily applicable only to states with a consolidated liberal democratic system.^{Footnote 60} However, multicultural conflicts arise not only in consolidated liberal democracies but also in settings that do not meet this definition, either because they are still in the process of consolidating a liberal democratic system or because they deliberately hold conflicting commitments to liberal rights and to illiberal values and interests simultaneously. In such cases some of the underlying assumptions of liberal multiculturalism do not hold true, leading to a failure in its application. One example Kymlicka gives for a concept that underlies liberal multiculturalism but does not hold in countries that are not consolidated liberal democracies is what he terms ‘liberal expectancy’. Liberal expectancy is the expectation that illiberal minorities will liberalise over time as a result of their exposure to the liberal society around them, even if they are granted multicultural protections that insulate them to some extent from external influence. This expectation is predicated on the strength of the liberal environment surrounding the minority and is essential for the continued stability of a state that grants multicultural accommodations to illiberal minorities. However, when the larger polity is itself not fully liberal, Kymlicka concedes that liberal expectancy is likely to fail.^{Footnote 61} Similarly, the application of liberal multiculturalism is destined to fail where the protection of liberal human rights is not fully entrenched, since in such settings the granting of multicultural accommodations to illiberal minorities may facilitate the creation of what Kymlicka terms ‘Islands of tyranny’ where the illiberal minority governs and in which the rights of citizens are jeopardised.^{Footnote 62} Kymlicka’s conclusion is that currently there is no applicable theory for the application of liberal multiculturalism in states that are not consolidated liberal democracies and that such a theory would require a great deal of additional research and have to take into account the divergent historical and geopolitical contexts in each relevant case.^{Footnote 63} The case of Israel discussed here, with its semi-liberal constitutional structure and with its grant of multicultural accommodations to the illiberal ultra-Orthodox minority, begins to tackle this task.

In Kymlicka’s discussion, which is mostly centred on post-communist countries, Kymlicka seems to assume that these countries are all in transition towards becoming in due course consolidated liberal democracies. Other scholars such as Smooha are more inclined to view some of these countries, such as Estonia and Latvia, as evolving into stable ethnic democracies.^{Footnote 64} Walker urges post-communist countries to adopt mixed constitutions, arguing that the mixed constitution is both more politically suitable for these countries and normatively superior to liberal constitutionalism. In the terms used in this article, while Kymlicka hopes that post-communist countries will eventually adopt fully liberal constitutional systems, Smooha expects them, and Walker urges them to adopt semi-liberal constitutional systems that combine their commitment to liberal rights with a firm commitment to their ethnic character which will be expressed in basic elements of their constitutional systems. The process of democratic backsliding which we are witnessing across the globe in recent years, in which even countries that have already become fully liberal, such as Hungary and Poland, have been backsliding and losing their liberal democratic characteristics seems to support Smooha’s and Walker’s prediction and highlights the need to engage more comprehensively with settings that are less than fully liberal.^{Footnote 65}

The aforementioned suggests that identifying and defining semi-liberal constitutionalism as a separate category is not merely of a theoretical interest. Many countries in the world choose to stop short of adopting a fully liberal regime and attempt to combine in the basic elements of their constitutional system both liberal and non-liberal elements. As will be shown below the use of liberal normative reasoning within such a semi-liberal constitutional setting, by courts or other decision makers, must be done with the outmost care. When such reasoning fails to take into account the special characteristics of the semi-liberal constitutional system within which it is applied it may backfire and end up harming liberal individual rights rather than promoting them. It should be recalled that courts and decision makers in such semi-liberal settings often have a strong commitment to the promotion of liberal rights except where it was explicitly decided that other constitutive values take precedence. It is therefore important to examine whether the straightforward application of liberal theory in such cases will adequately promote liberal rights, or whether an application which takes into account the semi-liberal nature of the constitutional system is required. The predominance, or even exclusivity, of liberal normative analysis within global constitutionalism, as well as the bias courts and other decision makers have towards viewing their own constitutional system as a fully liberal system that fits well with the global constitutional model, lead them to analyse and justify their actions and decisions on the basis of allegedly applicable liberal principles, and ignore the semi-liberal context within which these decisions are taken.

A major problem with the straightforward application of liberal reasoning in semi-liberal systems is that one consequence of the semi-liberal setting is that it may make some individuals more vulnerable than they would have been in a liberal system, while granting others more power to harm the rights of their fellow citizens. As the following analysis will show, ignoring these power differentials and applying strictly liberal rules can end up harming individual liberal rights rather than promoting them, while at the same time cloaking this unwanted result in alleged liberal justifications. The following analysis will be done specifically with respect to the case of Israel, and within it only with respect to the semi-liberal nature of religion–state relations in Israel. Nevertheless, a similar analysis, mutatis mutandis, can and should be carried out in other semi-liberal contexts, both in Israel and in other countries, in order to facilitate our understanding of the normative workings of semi-liberal constitutionalism.

Importantly, the analysis below will be done from the point of view of the more vulnerable groups, who in the particular context of religion–state relations in Israel are often women. In this sense this analysis is an exercise in Feminist Standpoint Theory that takes seriously Sandra Harding’s claim that: ‘Starting off research from women’s lives will generate less partial and distorted accounts not only of women’s lives but also of men’s lives and of the whole social order.’^{Footnote 66} In a semi-liberal constitutional system even more so than in a liberal constitutional system power differentials are deeply embedded into the system and standpoint analysis becomes even more crucial to understanding the effects of the system on differently situated individuals. In addition, the analysis builds on the feminist critique of liberalism’s inadequate perception of power and extends it to the semi-liberal setting where it is even more pertinent.^{Footnote 67} The feminist critique of power is based on the writings of the most influential theorist of power, Michel Foucault, who is said to have revolutionised the Western perception of power. Foucault posits that a major problem with Western perceptions of power is both the representation of power in a juridical form and its attribution to the sovereign.^{Footnote 68} ‘The language of power is law, not magic, religion or anything else.’^{Footnote 69} It is focused on the rights of the individual vis-à-vis the sovereign and neglects all other forms of power that threaten individual rights.^{Footnote 70} Thus, liberal individual rights are designed to protect individuals from the direct coercive power of the state, but fail to protect them from other forms of power. The distinction between the public and the private is an important manifestation of this problem within liberal thinking. According to Foucault, a proper analysis of power must extend beyond the limits of the state both ‘because the state, for all the omnipotence of its apparatuses, is far from being able to occupy the whole field of actual power relations, and further because the state can only operate on the basis of other, already existing power relations’.^{Footnote 71} Thus, Foucault posits that power is everywhere; it is exercised from ‘within the social body rather than from above it’, and it is deeply, albeit covertly, invasive.^{Footnote 72} Power is a productive social network defined by ‘a more-or-less organised, hierarchical, co-ordinated cluster of relations’.^{Footnote 73} Furthermore, power cannot be recognised simply by its appearance as a negative force. Power is accepted because it is a hierarchical, productive force that ‘induces pleasure, forms knowledge, and produces discourse’.^{Footnote 74} As Phelan explains, ‘[p]ower operates through discourses that define and legitimate its operation’.^{Footnote 75} This understanding of power reveals that religion and culture are major sites of power, both in the private and in the public spheres. According to Foucault power relies on the appearance of freedom and it covertly guides conduct and suggests possible outcomes because the person upon whom it is exercised is envisioned as a person whose actions are freely chosen.^{Footnote 76} In this respect, religion and culture are quintessential sites of power because they are highly determinative of individual actions and are regarded as positively informing rather than obstructing free choice.

In the context of the semi-liberal constitutionalism analysed here this view of power has several implications. First, it alerts us to the fact that religion and culture are major sites of power even within a liberal framework, where they are restricted to the private sphere. As we will see, the liberal tendency to associate religion and culture with the private, power-free, sphere carries into the semi-liberal system even though within such a system religion is often at least partly endowed with state power. Second, it highlights that within a semi-liberal constitutional system which promotes liberal precepts but at the same time uses state power to enshrine religious and cultural values religion and culture become even stronger because they are perceived by many as legitimately and positively informing individual actions and choices in both the private and the public sphere rather than as coercive. Third, as Nancy Fraser points out in reference to the liberal framework that attributes coercive power only to direct state coercion: ‘the proliferation of discourse governed by this liberal framework may itself function … to mask the actual character of power and thus to conceal domination’.^{Footnote 77} This insight is particularly true in a semi-liberal setting in which the liberal discourse serves to conceal not only private sphere domination, but the domination caused by indirect state power as well.

IV. Three examples from Israel

In this part of the article I will discuss three examples from Israel that demonstrate the special challenges of applying liberal rights reasoning within a semi-liberal constitutional system. Before delving into the cases three clarifications are in order. First, the claim that Israel is a semi-liberal constitutional system has already been asserted and demonstrated in Part II. That part laid out the two conditions of a semi-liberal constitutional system – the dual normativity condition and the basic elements condition – and analysed Israel’s constitutional and legal structure, highlighting how it meets the two conditions. The purpose of the following case analyses is to demonstrate the additional claim that applying liberal rights reasoning within a semi-liberal constitutional system can result in problematic decisions since the liberal reasoning ignores the power differentials inherent to the semi-liberal context. Second, I am not claiming that the semi-liberal nature of Israel’s legal system affects all legal disputes. Rather, it affects only those disputes in which the semi-liberal nature of the system changes the power balance between the parties. If the court applies liberal rules while ignoring the existing power imbalance then the resulting decisions cannot be justified even on their own terms. Thus, for example, while the semi-liberal nature of the Israeli legal system will not affect an ordinary contractual dispute between two litigants, it often affects contractual disputes between divorcing spouses because of the power imbalance between men and women ingrained in the religious personal laws that govern all divorces in Israel.^{Footnote 78} Third and last, a similar demand to take into account the semi-liberal nature of the system would be applicable, mutatis mutandis, to the use of liberal concepts to analyse and justify court and policy decisions in other semi-liberal contexts, such as that of personal laws in India and their effects on the rights of women.

The following discussion will include three cases decided by the Israeli Supreme Court in matters pertaining to religion–state relations. In the first and the last examples discussed I will demonstrate how the liberal analysis conducted by the court ignores the semi-liberal nature of the system and leads to problematic results in terms of human rights protection.^{Footnote 79} Conversely, the second example will be of a case in which the Court took into account the semi-liberal nature of the system and strove to restrict its adverse effects in order to protect the rights of vulnerable women to the largest possible extent.^{Footnote 80} The first example I will discuss is of a Supreme Court case in which the Court preferred protecting the religious freedom of the officiating Orthodox Rabbi – a state official – over the rights of the marrying couple.^{Footnote 81} As explained, in Israel the only recognised marriages are religious marriages conducted by religious officials. The Jewish religious establishment is controlled by Orthodox and ultra-Orthodox religious officials and accordingly the stream of Judaism practised and enforced by the Jewish religious establishment in Israel is Orthodox Judaism. All Jews, regardless of their religious beliefs, whether they belong to a different stream of Judaism or are secular or even atheist, must abide by the strictures of Orthodox Judaism as determined by these officials in order to obtain services from the state such as marriages or divorces.

While couples do not have free choice as to the nature of the marriage, and must be married religiously by Orthodox rabbis, the petitioners in the Guardians of the Tradition case hoped that they would have a choice as to the venue, since in Israel marriages can take place anywhere, including at the beach or in a public park. The petitioners who were Conservative Jews (a different, more modern, stream than Orthodox Judaism) wanted to hold the ceremony in a restaurant that does not hold an Orthodox Kosher certificate for the food that it serves, but instead maintains Conservative supervision over the food.

The Orthodox Rabbi refused to marry the couple in a restaurant that served food that did not have an Orthodox Kosher certificate. The state’s religious establishment backed him, insisting that the Rabbi has both the right and the authority to insist that the restaurant have an Orthodox Kosher certificate. The couple petitioned the Supreme Court asking it to require the state’s religious establishment to provide a Rabbi that would marry them in the restaurant that they chose, but the Court denied the petition. First, the Court examined whether there was a categorical religious (Halachic) prohibition in Orthodox Judaism on conducting a marriage ceremony in a place that does not hold an Orthodox Kosher certificate. The court concluded that there are conflicting religious opinions on the matter and that it was therefore questionable whether the Rabbi was authorised to refuse to conduct the ceremony on this ground. Nevertheless the Court went on to reject the petition using an allegedly liberal reasoning according to which the couple must respect the freedom of conscience and religion of the officiating Rabbi and may not force him to conduct a ceremony which contradicts his personal conscience and his religious beliefs.

My claim is that in the semi-liberal context in which this decision was given, the suggested reasoning is mistaken and does not serve the purpose of advancing liberal rights, as the Court seems to believe, but quite to the contrary it serves to enhance the illiberal (theocratic) nature of the state. The court strikes an alleged balance between the freedom of the Rabbi to refuse to perform the ceremony in a venue in which he believes he is not allowed to perform the marriage, with the freedom of the couple to choose a restaurant compatible with their religious persuasion, and determines that the Rabbi’s freedom should gain the upper hand. This may be a plausible, liberal, resolution of the conflict in a liberal system where the couple has a civil alternative for the marriage or can be married within their own religious denomination.^{Footnote 82} However, the Supreme Court disregards the fact that the Rabbi whose liberal religious freedom it is protecting is a state official, part of a religious establishment, who, along with his peers, holds the exclusive authority to perform marriages, within a system that is committed ideologically to a religious and profoundly illiberal violation of the rights of the citizens subject to it. The only reason why this couple is marrying before this specific Rabbi is that the couple’s freedom of religion was fundamentally violated by the state, requiring them to undergo a religious ceremony which is contrary to their conscience and belief in order to marry.

Furthermore, the Rabbi himself is not an innocent bystander. Just like the establishment that he is part of, he too is ideologically committed to imposing his own beliefs on the couple by marrying them according to his own version of a Jewish religious wedding even though this version goes against their own religious beliefs. The religious establishment in Israel and its officials adamantly oppose the institution of civil marriages in Israel despite the wide public support for such a move.^{Footnote 83} Under such circumstances protecting the religious freedom of the Rabbi at the expense of the already severely thwarted freedom of the petitioning couple does not promote liberal rights and cannot be justified in liberal terms. Had the court taken proper account of the semi-liberal nature of Israeli constitutionalism its analysis should have been quite different. Since there is no categorical religious prohibition on conducting an Orthodox religious marriage in a venue that serves food without an Orthodox Kosher certificate the Court should have accepted the petition, and required the Orthodox religious establishment to provide the couple with a Rabbi that is willing to perform the wedding in their chosen venue. Such a decision would have properly reflected the commitment of the Supreme Court to protecting the citizens’ liberal rights to the largest extent possible, while still respecting the role of the religious establishment in conducting marriages as established by state law. Instead, the Court analysed the case as a horizontal clash of rights within which no power differentials exist between the parties and where the fundamental private right to freedom of religion of the Rabbi must therefore be given the upper hand. This case is a good demonstration of Nancy Fraser’s claim that the liberal discourse serves ‘to mask the actual character of power and thus to conceal domination’.^{Footnote 84}

The second example I will discuss is one in which the Court has shown much more awareness to the particular rights violations created by the semi-liberal system and to the need to restrict their scope as much as possible. The context of the Doe case was a petition to the Israeli Supreme Court against the Muslim Sharia Court that refused to appoint a woman as arbitrator in a family law dispute despite the wife’s request to have a woman as arbitrator on her behalf.^{Footnote 85} The Sharia court reasoned that influential schools of thought in Islamic law such as the Maliki school of thought to which the court adheres forbid the appointment of a woman arbitrator. The Supreme Court rejected this argument, ruling that when deciding what arbitrator to appoint on behalf of the wife the Sharia court may not reject women as arbitrators merely on account of their sex even if such a rejection was religiously justified. The Supreme Court reasoned that while the civil legislature gave the Sharia courts the authority to apply Muslim religious law to matters of personal laws of Muslims in Israel, the Sharia Courts are nevertheless subject to equality legislation in all matters that are not clearly exempted from equality strictures by the authorising civil law. The Israeli Women’s Equal Rights Act, while declaring a general commitment to women’s full equality in all legal matters allows specific exceptions in matters pertaining to the substantive application of religious laws of marriage and divorce, and to the appointment of religious judges.^{Footnote 86} The Supreme Court held that the appointment of a female arbitrator on behalf of the woman pertains neither to the substantive application of the religious laws of marriage and divorce, nor to the appointment of religious judges, and is therefore subject to the equality mandate and must be allowed by the Sharia Court.^{Footnote 87} This case is an example of the Supreme Court’s willingness to restrict religious law’s adverse impact on women’s right to equality in order to advance liberal egalitarian constitutional principles, wherever this is possible under the semi-liberal system within which it operates. In this case the Supreme Court has struck the proper balance between the liberal and non-liberal components of Israel’s semi-liberal constitutionalism. Unlike religious bodies in liberal states which are private and which are therefore largely immune to egalitarian constitutional norms, most religious courts in Israel operate as state organs and are therefore subject to constitutional supervision wherever prescribed by law. As the Doe court explained with respect to religious state organs:

Thus, in the Guardians of the Tradition example the Supreme Court, seemingly oblivious to the power differentials between the petitioners and the Rabbi, their different positions in the state, and the coercive nature of the Rabbi’s authority, chose to protect the religious freedom of the Rabbi over that of petitioners. Conversely, in the Doe case the Court, clearly aware of the disadvantaged position of women due to the imposition of patriarchal religious personal law on all citizens, restricted the reach of the religious court’s authority and enforced the liberal principle of equality. Nevertheless, it is important to understand that despite its good intentions, the Supreme Court’s ability to ameliorate the discrimination against women, and the violation of freedom of religion and of conscience inherent in the operation of the religious establishment in Israel – Jewish and non-Jewish – is highly restricted because of the authority that these institutions have been granted by the state to enforce their illiberal religious ideology in areas under their control. For this reason it is highly important for the Supreme Court to take into account the impact of the semi-liberal constitutional structure on liberal rights not only in cases that directly involve these religious institutions but also in cases that indirectly affect their powers such as in the following example. Unfortunately, as this last example will show, the Supreme Court is not sufficiently aware of this.

While the Guardians of the Tradition and the Doe examples involved the direct action of state organs belonging to the religious establishment, the last example – the Rubinstein case – is more indirect but its consequences are substantial.^{Footnote 89} The case demonstrates how within a semi-liberal constitutional system the grant of multicultural accommodations on allegedly liberal grounds to a religious minority that controls the religious establishment exacerbates the violation of the rights of others rather than advancing rights as multicultural theory would expect. The background to the case is the exemption by law of ultra-Orthodox high school male students from studying the core curriculum that is mandated by the Ministry of Education. This law allows the ultra-Orthodox high schools (Yeshivot) to teach only religious studies while still getting state funding.^{Footnote 90} Although the exempting law was enacted in 2008, it served to legalise a situation which has been ongoing for many years prior to its enactment, in which ultra-Orthodox schools for boys did not teach them any secular subjects. Ultra-Orthodox boys were, and still are, taught only religious studies through most of their school years, and are not even taught core non-religious subjects such as maths, English and citizenship education. This situation, which is unique to the ultra-Orthodox community, was justified by the legislature as a necessary multicultural accommodation needed in order to enable the ultra-Orthodox community to preserve its unique, highly traditional, way of life. The petition to the Israeli Supreme Court in the Rubinstein case claimed that the exempting law violated the constitutional right to human dignity and liberty because it infringed on ultra-Orthodox boys’ right to education. The Court rejected the petition. While the Court held that the community’s right to culture does not necessarily grant it the right to prevent their boys from studying the core curriculum, it nevertheless decided that if the state chose to give the ultra-Orthodox community such a right, there was no reason for the Court to intervene.^{Footnote 91}

In my view the Rubinstein Court was mistaken in rejecting the petition and in analysing the case as though the context in which it is taking place is that of a fully liberal constitutional system rather than of a semi-liberal one. As already explained and demonstrated in the previous examples a characteristic of a semi-liberal system is that its illiberal aspects result in the increased vulnerability of some of the citizens (due to the authority of illiberal institutions over them) while strengthening the power of other citizens (because of their control over those institutions and over other citizens through them). Israel’s extensive religious establishment is controlled mostly by the political and religious leaders of the ultra-Orthodox community, and staffed mostly with ultra-Orthodox men that have been educated exclusively within the ultra-Orthodox education system. Accordingly, Israel’s Jewish religious courts, the Rabbinical Courts, are mostly staffed by ultra-Orthodox judges (all of whom are men) whose rulings are based on ultra-conservative interpretations of the Jewish religion that are discriminatory towards women. By exempting ultra-Orthodox boys from studying the core curriculum and allowing their education to be dedicated solely to ultra-conservative religious studies, with no civic or secular education, the state and the Supreme Court are ensuring, perhaps inadvertently, that the extensive discrimination against women practised in the Rabbinical Courts will continue and even exacerbate. Because under Israeli law all Jewish women are subject to the Rabbinical Courts in matters of marriage and divorce, the alleged multicultural accommodation that is given to the ultra-Orthodox community has widespread negative ramifications for the rights of all Jewish women in Israel, well beyond the community itself. These ramifications, which would not occur in a fully liberal system – in which citizens are not subjected to mandatory religious law in matters of marriage and divorce – were not even considered by the Israeli Supreme Court in Rubinstein. Instead, the court applied a regular liberal multicultural analysis that assumes that the ultra-Orthodox community is a private entity and that the results of the law will only affect the community internally. The Rubinstein Court failed to notice the indirect, but very real, consequences resulting from the semi-liberal nature of the state which gives ultra-Orthodox men appointed as state religious judges the power to make life-altering decisions that impact the rights of all Jewish women. To use Kymlicka’s vocabulary, the consequence of allowing the ultra-Orthodox community to confine the education of its boys to religious studies and later having them appointed as Rabbinical Court judges (or as officials in the religious establishment in general) increases the tendency of these courts and of the religious establishment in general to function as ‘Islands of Tyranny’.^{Footnote 92}

A possible objection to the preceding argument may be that while it is unfortunate for the liberal commitments of the state that discrimination against women will deepen due to the exemption of ultra-Orthodox boys’ schools from the core curriculum, such discrimination is nevertheless an inevitable and justified result of the commitment of the state to the Jewish religion in matters of marriage and divorce. However, this objection is mistaken. While Israel’s semi-liberal system is committed by law to maintaining the exclusivity of religious laws of marriage and divorce, this commitment does not in any way require the state to enforce the most conservative and discriminatory ultra-Orthodox interpretation of Jewish religious law on its citizens. Quite to the contrary, most organs of the civic/secular state in Israel are quite vexed by the rigid and discriminatory nature of the rulings of Rabbinical Courts, and most interpretations of Jewish religious law in this area are far more flexible and egalitarian than those applied by rabbinical court ultra-Orthodox judges.^{Footnote 93} In both the Rubinstein and the Guardians of the Tradition examples the Court’s liberal tendency to view the cases as involving only private interests (those of the Rabbi and of petitioners in the first case and those of the ultra-Orthodox community in the current case) and ignore questions of direct and indirect power differentials stemming from the semi-liberal nature of the system result in an excessive violation of rights which cannot be justified by the semi-liberal setting. Working towards a relative liberalisation of the judgments of Rabbinical Courts as well as of the other functions carried out by the religious establishment, whether through expanding the secular and civic education of ultra-Orthodox boys or through other means is perfectly compatible, and I would argue even required, by the Israeli semi-liberal constitutional system.