Constitutionalizing secularism, alternative secularisms or liberal-democratic constitutionalism? A critical reading of some Turkish, ECtHR and Indian Supreme Court cases on ‘secularism’

Disclaimer/Complaints regulations If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.


Introduction
Recently, the heated debates on secularism in social sciences, politics and, increasingly, in political theory and political philosophy have also infected legal theory, constitutional law and comparative constitutionalism.Generally speaking, it may indeed be 'too late to ban the word "secular"' or to remove 'secularism' from our cultural vocabulary.The argument, however, that 'too many controversies have already been stated in these terms' 1 seems unconvincing to me.For the purposes of normative theory generally, and particularly for those of constitutional law and jurisprudence, the focus of this article, I have proposed to replace all normative concepts of 'secularism' by 'priority for liberal-democracy' 2 because I am convinced that we are better able to economize our moral disagreements or to resolve the substantive constitutional, legal, jurisprudential and institutional issues and controversies by avoiding to restate them in terms of 'secularism' or 'alternative secularisms'.
With regard to the 'constitutional status of secularism' we can discern three distinct positions all sharing the argument that 'secularism (...) in most liberal democracies (…) is not explicitly recognized in the constitutional text or jurisprudence' 3 and that it has 'no clear standing among constitutional values' either: it is 'not clear which established constitutional category secularism fits into or what is the underlying value behind it'. 4 The first position, defended most vividly and outspokenly by András Sajó -a Judge at the European Court of Human Rights (ECtHR) and Professor at the Central European University in Budapest -tries to overcome the absence of 'secularism' in most liberal-democratic constitutions 5 and to streamline their messiness by developing a 'more robust theory of constitutional secularism' to remedy the fact that 'most democracies are without a strong normative theory or practice of constitutional secularism', to present a 'clear agenda' and a 'coordinated action plan' in order to 'defend' vulnerable constitutions against the threats of ('strong') religions.'Constitutionalism relies on the use of the human faculty of reason and on popular sovereignty.The first consideration translates into the duty of public reason giving in law and denies the acceptability of divine reasons; the second precludes any source of law but the secular.A robust notion of secularism, animated by these considerations, is capable of patrolling the borders of the public square'. 6he second position proposes to fully contextualize secularism 7 also in normative theory and constitutional law by developing theories of alternative secularism(s): 'inclusive', 'passive', 'moderate', 'evolutionary', 'weak', 'tolerant', 'liberal', 'benevolent' or 'ameliorative' secularism, laïcité plurielle, positive, de gestion, bien entendue in opposition to 'exclusive', 'assertive', 'aggressive', 'strong', 'intolerant', 'statist', or 'malevolent' secularism (Bhargava, Modood, Jacobsohn, Shankland-Hurd, An-Na'im, Willaime, and Bali or in politics by the Turkish AK Party).
The third, rather radical position presented and defended below is to criticize 'secularism' in all its varieties as a viable constitutional principle or, in other words, to drop secularism from our constitutional language and replace it by liberal-democratic constitutionalism. 8ere are my main reasons: First, secularism is not only, obviously, a very complex, polysemic and -as all our basic concepts -an essentially contested concept but also a 'fuzzy', 9 chameleonic, highly misleading or 'cacophonous' concept.If we are able to discuss the substantive issues of state-religion relations with less fuzzy concepts we should do so instead of translating all and everything into the language of 'secularism'.Second, and fortunately, constitutions and constitutional jurisprudence provide for such concepts both in terms of rights or first-order principles and in terms of 'underlying values' or second-order principles.The absence of 'secularism' in most liberal-democratic constitutions demonstrates this clearly.Third, the really important substantive issue is not whether states and politics are 'modern' or 'secular', whatever that may mean, but rather whether they are liberal-democratic or, in other words, live up to the demands of minimal, minimal-liberal, and minimal-democratic morality and what this requires in terms of constitutional and institutional arrangements and politics/policies.In the fashionable language of systems theory: in the perspective of constitutionalism the important leading distinction (Leitdifferenz) is not secular versus religious but 'liberal versus non-liberal' and 'democratic versus non-democratic'.This allows one to research and detect incompatibilities between both 'secular' and 'religious' regimes, practices and ideologies with liberal and/or democratic ones.Fourth, the principle of constitutional secularism does not allow one to investigate, but rather it conceptually eliminates deep and serious tensions between 'secularism' and liberal-democratic constitutionalism (e.g. in Turkey).Fifth, 'constitutional secularism' tends to hide from view structural tensions between liberal constitutionalism (Rechtsstaat, the rule of law) and democratic constitutionalism and it undermines reasonable balances amongst the two in general, particularly in cases of militant democracies such as Turkey and India. 10efore outlining how I will proceed it seems appropriate to summarize the main characteristics of my own approach that combines moral and political philosophy, political and legal theory and constitutional law and is explained at length in other texts.
First, I defend a theory of differentiated morality. 11Its hard core is minimalist morality, trying to prevent 'malfare' by guaranteeing basic rights to security (rights to life, liberty, bodily integrity, protection against violence; minimal due process rights; minimal respect and collective toleration) and subsistence.Safeguarding moral and legal minimalism distinguishes minimally 'decent states' from plainly immoral ones but not all decent states have been or need to be liberal or democratic states.Liberal-democratic morality is much more demanding but still minimalist, requiring equal civic and political rights (equal legal rights and equality before and in the law; free and equal active and passive voting rights, freedom of political communication, modern (negative) non-discrimination rights (equal respect) and individual toleration (freedom of conscience, freedom from religion)).It can and also should be defended wherever circumstances allow, but only if this does not infringe minimalist morality.Minimalist or 'thin' liberal-democratic morality is different from a much more demanding, more egalitarian democratic morality requiring equal socio-economic and fair cultural rights and opportunities -and corresponding policies of redistribution to achieve socio-economic and political 'equality of opportunities'and policies of cultural even-handedness.This demanding utopia, again, can and should be defended under the non-infringement proviso.It should be clearly distinguished from what philosophers have called 'comprehensive moral liberalism' which requires a specific way of a good life -leading an autonomous, self-chosen and transparent life free of illusions -and demanding the 'rational revisability' of all cultural and psychic 'conditions of choice'. 12econd, this brief outline allows me to specify what I mean by 'liberal-democratic constitutionalism' in legal and institutional terms.As we know, 'liberal democracy' has been a historically late and contested compromise in theory and politics and so is liberal-democratic constitutionalism.By 'liberal constitutionalism' I quite traditionally understand the core or rockbottom elements of any form of modern constitutionalism: the full list of civic rights (against violation by states and by all others), the rule of law, separation of powers, minority protection against simple majoritarianism, written or factual constitutions and constitutional courts and judicial review.Liberal constitutionalism contains inherent tensions, e.g. between life/security and civil freedoms.By 'democratic constitutionalism' I understand, again quite traditionally (Dahl, Beetham), free and equal political rights (freedoms of political communication, voting, representation), democratic elections of parliaments and governments, governmental responsibility, democratic majoritarianism and fairly limited powers for constitutions and constitutional review, if any.Liberal-democratic constitutionalism even in this minimalist understanding contains serious tensions, 13 most prominently between democratic majoritarianism and the protection of civic rights and the rights of minorities as well as between extensive powers of constitutional courts (liberal and neo-liberal judicial activism) and democratic distrust against Constitutional Courts, lacking democratic legitimacy, and judicial activism that drastically limits legitimate democratic political space for majority decisions.More specifically, I defend minimalist conceptions of liberal-democratic constitutionalism in opposition to thick international Human Rights maximalism (e.g. by Temperman), to more demanding concepts of 'thick' participatory, egalitarian democracy, and to both (neo-)liberal judicial activism by constitutional courts and 'thick' democratic egalitarian judicial activism.
Third, for these purposes it is important to distinguish liberal-democratic constitutionalism -articulating the core principles and institutional arrangements of all liberal-democratic constitutions deserving this name -from 'liberalism' as a political philosophy or political ideology in order to avoid the misunderstanding that it would obscure other perspectives and possible regimes than 'liberal democracy as currently known in the West.' 14 In my view priority for liberal-democratic constitutionalism covers minimal morality and minimally decent regimes as well as liberal-democratic ones.Decent states are not a 'Western' invention and my interpretation of liberal-democratic constitutionalism is -if the non-infringement proviso is fulfilled -open to more demanding interpretations of freedom, equality, fraternity or solidarity than the ones stressed by classical and neo-liberalism.Principles of liberal-democratic constitutionalism are in tension with each other, are inevitably underdetermined, and allow for a variety of institutional regimes (including non-Western ones such as India) and for alternatives (such as associative democracy).They allow criticism of predominant identifications of liberal-democratic constitutionalism with (specific varieties of) capitalist market economies, welfare regimes, representative political democracy, or 'Western' institutional traditions.
Fourth, it is common knowledge among constitutional lawyers that constitutions contain conflicting rights and principles even within liberal constitutionalism (e.g. the right to life/ security and civic liberties, individual versus associational autonomy or free speech versus nondiscrimination) and between liberal constitutionalism and democratic constitutionalism (e.g.'freedom' versus 'equality') -the stuff of all 'hard cases'.Yet most moral and political philosophers still think that the only strategy to overcome relativism is to find context-independent 'lexical hierarchies' of conflicting principle or one, deep, harmonizing foundational principle so that, eventually, there is just one right moral and legal solution/decision in all these hard cases.Moral pluralists, a tiny group of moral philosophers, myself amongst them, are deeply convinced that these two strategies of moral monists are doomed to fail not only practically but in principle: there is no and there cannot be a context-independent hierarchy and foundational strategies reproduce conflicts only on a deep 'meta level'.Yet we are not condemned to 'anything goes' or drowned in the abyss of relativism because we have, for the time being, good but revisable, contextual reasons and can find better or worse balances of conflicting rights and principles.
Fifth, one way of finding context-specific answers in cases of conflicting rights or firstorder principles is to resort to more general und unspecified second-order principles.If 'secularism' is not a constitutional first-order principle, as is conceded by Sajó, it might be a secondorder principle we can appeal to as a guideline in balancing rights.We do not only regularly appeal to 'public order', 'morals' or to 'proportionality' (further defined by case law) but also to principles (my favorite ones in brackets) such as (embedded) 'impartiality', (relational) 'neutrality' or 'fairness' (as even-handedness in cultural matters), and it seems that we cannot do without at least some of such abstract principles.Yet we should do so very carefully because they are characterized by higher degrees of under-determinacy (wide margins of discretion for Justices) compared to constitutional rights and principles and we should also respect a principle of parsimony: keep their number as small as possible.My claim is that 'secularism' either only repeats, in an unclear way, things that are articulated much more clearly by the other principles just mentioned or introduces at this meta-level an enormous amount of ambiguity, fuzziness, even complete indeterminacy, and incompatibility.It cannot guide our weighing but distorts and undermines all reasonable balances.In short, it is certainly the wrong second-order principle, or so I want to show.
Sixth, as a committed non-foundationalist in moral and political philosophy I think that our chances to minimize moral disagreement are at least as high in our everyday legal and constitutional debates as they are when it comes to deep, general or universal justifications of moral principles (in 'contract', 'nature', 'Reason', 'speech acts', 'needs' or what have you).So Sajó is indeed right that 'a constitutional doctrine is not about political theory; it is a jurisprudential concept' but his 'Constitutional Secularism' is clearly not a jurisprudential concept and he certainly does not stick to his proclaimed 'humble jurisprudential interests'. 15Instead he presents a rather demanding version of 'personal autonomy' and 'the use of the human faculty of reason' as an 'epistemological precondition' of secularism; 16 he speculates on 'Enlightenment' and engages, together with Zucca, in a superficial way in issues of sociology of religion. 17Instead of 'going meta' I rather trust the wisdom of ld-institutions and rights and the 'discipline' of law and jurisprudence, of our everyday legal and constitutional debates.
Finally, a few words on how I will proceed.
First, I will focus on rulings of constitutional courts in which the 'principle of secularism' played a prominent and highly contested role.My aim is to highlight tensions between the constitutional appeal to and the jurisprudential use of 'secularism', on the one hand, and basic principles and rights of liberal-democratic constitutionalism, more specifically between 'secularism' and democratic constitutionalism, on the other hand, because the tensions between 'secular-ism' and liberal constitutionalism have already been sufficiently analyzed (e.g. in the criticism of landmark cases of French High Courts, the Turkish Constitutional Court and the respective rulings of the ECtHR) 18 whereas the former has received much less attention.
Second, I will focus on two 'non-Western' countries, 19 Turkey and India, which have constitutionalized prominently different varieties of secularism but also share rhetorics and practices of militant democracy.Militant democracy is 'a form of constitutional democracy authorized to protect civil and political freedoms by pre-emptively restricting the exercise of such freedoms'.It is a particularly paradoxical way in which democratic states 'act in a militant and repressive manner to combat threats to its democratic future'. 20We know strong forms of militant democracy from the constitutions of many European countries such as Germany, Italy, France, Poland, Bulgaria, Spain, Hungary and the Ukraine directed against fascism, racism, right and leftwing and nationalist terrorist violence but they have not used 'secularism' as an argument.Increasingly, countries are faced with the perceived threats of religious violence and fundamentalist terrorism.Turkey and India have introduced militant democracy against presumed or real threats of religious violence much earlier and the jurisprudence of their constitutional courts is rife with intriguing cases from which we can draw important lessons.
My highly selective focus on only two countries in itself cannot plausibly demonstrate that we should replace secularism by a priority for liberal-democratic constitutionalism in general.Turkish laiklik, as French predominant laïcité, is not pluralist, inclusive or moderate and defenders of alternative secularisms might say it is a badly distorted (Kemalist) interpretation of a good principle.The Indian case, then, is more to the point in this regard because the Indian Constitution is often explicitly defended as a viable, even preferable alternative of secularism compared to many 'Western' ones. 21Already in 1999, I discussed 'American' secularism, arguably the best candidate for an inclusive, pluralist, passive and moderate secularism, in order to show why it would be better to drop 'secularism' as a second-order principle in constitutional debates.American 'passive' secularists, and also Supreme Court Justices, have for a long time been split into 'two different interpretations -accommodationism and separationism' and 'secularism' is not informative to decide the issues.
In Section 2, I briefly introduce the selected Court Rulings in the context of Turkey and India.In Section 3 'Constitutional secularism?What's in the name' I open the Pandora's box of 'secularism' and present an analytic taxonomy of the variety and overlaps of twelve different meanings derived from a decontextualized, symptomatic reading of the selected cases in order to discuss the normative uses and abuses, the internal tensions and the contradictions of 'secularism' by Justices of Supreme Courts.Already here, I insert a table of these twelve meanings of secularism to give the reader a first impression of the varieties of secularism.In Section 4 'How did secularism work?' I present a brief contextualized criticism of the decisions in the selected cases.In my conclusions 'Constitutionalizing secularism or contextual-izing liberal-democratic constitutionalism?' I defend my rather minimalist conception of liberaldemocratic constitutionalism and its consequences for the limited, minimalist but crucially important roles of Constitutional Courts and judicial activism.

Constitutional Court decisions on secularism in the Turkish and Indian context
As already indicated, Turkey and India show remarkable similarities: (i) an aggressive concept of militant democracy using a presumed violation of the principle of secularism as the core argument; (ii) a contested but prominent perception that the predominant religions have been fundamentalist or totalistic, threatening liberal-democratic constitutionalism; (iii) constitutionalizing the principle of 'secularism' and militant democracy has been and still is perceived to be the inevitable, necessary and adequate way of defending liberal-democratic constitutionalism.Yet the countries also show remarkable differences: (i) India was colonized and the British legal tradition had an important direct and indirect impact on the Indian legal system whereas Turkey has never been a colony.The impact of 'Western' legal traditions on the Ottoman Empire was indirect and self-chosen (though obviously under 'threat') and Kemalist 'modernization' has been explicitly and voluntarily oriented at 'the West', particularly at French conceptions and traditions of laïcité.
(ii) Turkey has, comparatively speaking, been a more mono-religious country from the 1920s onwards and still is today 22 whereas India has been one of the most religiously diverse countries at the time of Constitution-making and still is today.Islam could be more plausibly constructed as a 'fundamentalist' threat (or as a counter-revolutionary force associated with the 'ancient regime') in the early 20 th Century in Turkey (though this perception may turn out to be historically untenable) whereas Hinduism was ambivalently perceived as a source of tolerance and harmony as well as of extremist nationalism by the framers of the Indian Constitution.There are also remarkable changes since then: Kemalism, at least nowadays, is more and more a threat both to freedom of religion as well as to non-dirigist democracy: on three occasions authoritarian Kemalist military elites organized military coups against democratically elected governments and Kemalist judicial elites defended their version of 'secularist democracy' by banning democratic parties.The Kemalist political and constitutional order did not live up to a minimalist understanding both of liberal constitutionalism and democratic constitutionalism.Turkish 'political Islam', particularly the AK Party, on the other hand, is now defending -at least officially -a liberal understanding of freedom of religion and a non-elitist conception of democracy. 23In India we clearly see the opposite development: both Islam and Hinduism have considerably fundamentalized since the debates in the Congress Party and Constitution-making, particularly since the 1970s.
(iii) The constitutional status of 'secularism' is rather different: the Constitutions of the Turkish Republic since 1937 officially declare the state to be a 'secular' as well as a 'national, democratic and social' state (1961 Article 2; 1982 Article 2) forbidding any amendments. 24The words 'secular' or 'secularism' were absent from the Indian Constitution and have only been added, after two failed earlier attempts -without any specification -in the Preamble together with 'socialist', by the 42 nd Amendment in 1976.
(iv) We see two remarkably different varieties of 'secularism' not only politically and in public discourse but also in the Constitutions and the Jurisprudence of the Supreme Courts: the Kemalist-Turkish, like the French one, is aggressively secular in all regards when it comes to 'strict separation' of state and public institutions from religion (legitimate legal pluralism, education, headscarves, working hours in public establishments etc.) but does not respect the relative autonomy of religions (particularly Islam) from the state, so defending a rather myopic variety of the 'two autonomies', whereas the Indian variety is accommodative and ameliorative in all these respects.
I select Constitutional Court cases in which the interpretation and application of 'secularism' has been the core issue but neglect all cases on religious instruction and education, on secularist restrictions of freedoms of religious beliefs and practices so prominent in Turkey and so significantly absent in India.As already indicated, I focus on the tension between secularism and democratic constitutionalism, on the most massive consequences of the appeal to the principle of 'secularism' in the Turkish and Indian varieties of militant democracy which are comparatively among the most aggressive ones anyway.Since its establishment in 1962, the Turkish Constitutional Court (TCC) has banned 24 parties, 5 for violating the principle of secularism 25 whereas in European 'militant democracies' after the Second World War only 3 parties have been closed down.I select two cases of serious restrictions of freedoms for or the dissolution of political parties.On 21 May 1997 the Principal State Counsel at the Court of Cassation accused the Refah Partisi (the Welfare Party) of having become a 'centre of activities against the principle of secularism'.On January 16, 1998 the TCC ordered that the party be dissolved, that its assets be transferred to the treasury, that Erbakan, Kasan, and Tekdal be banned from sitting in Parliament or holding other forms of political office for a period of five years.The ECtHR, in a judgment delivered on 31 July 2001 by four votes to three, rejected the applicants' charge and held that there had been no violation of Article 11 (freedom of assembly and association) of the ECHR, a judgment unanimously upheld by the Grand Chamber of the same Court on 13 February 2003.On 14 March 2008 the Public Prosecutor filed his charge to close the governing AK (Justice and Development) Party, which had won the last general election on 22 June 2007 (46.6% of all votes and 341 of the 550 seats in Parliament), as a centre of anti-secular activities and requiring a five-year ban from all political activities for 71 senior politicians, including Prime Minister Erdogan and President Gul.The TCC immediately and unanimously accepted the indictment.The AK Party responded to the indictment in April and June 16 and the TCC ruled on 30 July 2008: a majority of 6 out of 11 Justices voted in favour of closure (short of 1 to reach the required 3/5 majority for closure) but the Court issued a 'serious warning' against anti-secular activities and a threat to halve the party's public funding.
Indian governance contains 'two controversial pillars'. 26First, the constitutional power of the Central Government of India to dismiss elected MPs and state-governments: President's Rule (Article 356 of the Constitution), a relic of its British Colonial Past (Government of India Act, 1935, vide Sections 93 and 45).Up to 1991, according to the Lok Sabha Secretariat, it was imposed on 82 occasions, and 13 times under Section 51.On 23 occasions the assemblies were dissolved on the advice of the Chief Ministers/or due to their resignations, and on 18 occasions the assemblies suspended were subsequently revived (Bommai §11).In the immediate aftermath of the violent dismantling of the Babri Masjid in Ayodhya in December 1992 and the following wave of violence the Central Government of India dismissed the elected governments of three states: Madhya Pradesh, Rajasthan and Himachal Pradesh.These dismissals were subsequently appealed to the Indian Supreme Court, whose nine-Justice Bench in the landmark decision in S.R.Bommai v Union of India, on 11 March 1994, upheld the actions of the Centre.Second, the statutory rules for conducting elections: India's election law contains corrupt practices provisions that are traceable to an 1883 British statute, which became the model for the Government of India Act of 1919.Section 123 (3 and 3A) of the Representation of the People Act (RPA, 1951) departs significantly from its legal predecessors only in one aspect that 'electoral corruption in India may involve abuses associated with religious speech'. 27On December 11, 1995, a three-judge panel of the Indian Supreme Court found this section to be constitutional (Prabhoo v Kunte & Others), upheld the decision of the Bombay High Court (1988) by which the election of Dr. Ramesh Prabhoo, a candidate representing Shiv Sena, an extreme Hindu nationalist party, to the Maharashtra State Legislative Assembly had been declared void (guilty of 'religious speech' as corrupt practices).

'Constitutional secularism?' What's in the name?
The status of 'secularism' in the Indian Constitution after the 42 nd Amendment has been hugely contested.The opinions of the Justices in my selected cases constitute, according to Panikar, 'by far the most significant interpretation of the secular character of the Indian Constitution to date'. 28ost justices agree that 'a secular state is not easy to define' (Setalvad 1969, quoted by Sawant §147) and that the expressions 'socialist' and 'secular' in the Preamble 'by themselves are not capable of precise definition' (Reddy §304).One conclusion has been that 'secularism is a vague concept, not defined in the Constitution and hence, cannot furnish a ground for taking action under Article 356 [President's Rule, VB]'. 29This vagueness is defended as such with a completely opposite intent: 'The term "secular" has advisedly not been defined presumably because it is a very elastic term not capable of a precise definition and perhaps best left undefined.By this amendment what was implicit was made explicit' (Ahmadi §29).'It is true that the Indian Constitution does not use the word 'secularism' in any of its provisions, but its material provisions are inspired by the concept of secularism' (Reddy §305).Is 'secularism' then 'a vacuous word or a phantom concept' and a fatal constitutional weakness or is it a necessary 'key to the understanding of the relevant provisions of the Constitution' (Reddy §308), 'part of the basic structure (…) and also the soul', or the basic 'creed' of the Constitution' (Sawant)?
So, 'what's in the name?'After extended debates in the humanities, social sciences and normative theorizing we could know that secularism denotes a family of concepts: 'secular', 'secularity', 'secularization', and 'secularism' with regard to institutions, social processes and politics, principles, and meta-narratives. 30For reasons of space I have to bracket any discussion on 'cultural secularization' and 'sociological secularization' as cultural and social 'preconditions' of normative constitutional secularism. 31Particularly for Justices and constitutional lawyers, nonexperts in these fields, it seems wise to disentangle debates on constitutional secularism from highly contested sociological statements.Hence I focus on the former only.Secularism is often indistinctively used as synonymous with (i) some of the core provisions, rights, and institutions of liberal-democratic constitutionalism (the 'content' or 'essence' of liberal-democratic constitutionalism) with regard to the relationship between society-politics-state and (organized) religions; (ii) with its implicit second-order principle(s) or basic creed/spirit/philosophy; (iii) as a specific way of the good life; (iv) as a specific foundation or justification or a 'precondition' of liberal-democratic constitutionalism; and (v) as a particular world view or symbolic universe.Secularism seems to be a catch-all phrase for all the goodies of liberal-democratic constitutionalism, either as the foundational principle of 'the fundamental and basic features of the Constitution' or on a par with them: 'secularism, socioeconomic and political justice, fraternity, unity and national integrity' (Ramashwami §189) 32 or 'socialism, secularism and democracy' ( §199).All of this is called, without further specification, 'the secular fabric of the country' (Sawant in Bommai §135).In order to understand these different meanings of secularism or these different 'pillars' or 'dimensions' of secularism (AK June, 18f) and their relationships and tensions, we have to disentangle this cluster and in doing so we have to use ordinary constitutional or legal language instead of treating it as a more or less 'poor' 'proxy' of secularism. 33In distinguishing twelve analytically different meanings of secularism, I move from rights and institutions to principles and to meta-talk starting with the content of liberal-democratic constitutionalism.
My analytical taxonomy, which could also have been derived from court rulings and constitutions in other countries, serves a dual aim.On the one hand, it helps to clarify our semantic understanding of the fuzzy, overlapping, and usually clustered terms used in legal, jurisprudential and constitutional texts.In this sense it may be of assistance if we can better understand what we mean when we disagree in our normative evaluations and judgments.On the basis of these distinctions I provide a general normative criticism of these varieties of 'constitutional secularism' from the standpoint of my minimalist understanding of liberal-democratic constitutionalism, in particular of the arguments and rulings of the Turkish and Indian apex courts and of the ECtHR.The advantage of the analytical distinctions in this regard is that we can discuss the respective normative disagreements point by point instead of mixing everything up.

Secularism 1: 'Secularity' of the state or the autonomy of the state from (organized) religions
Any decent, and a fortiori any liberal-democratic state has to be a 'secular' or a 'non-religious' state in a certain sense.This 'secularity' of the state seems to be the least contested, rock-bottom meaning of secularism. 34The public prosecutor in Turkey accused the Refah party and Erbakan of 'the abolition of secularism in Turkey', 'calling for the secular political regime to be replaced by a theocratic regime' mainly because of the 'intention to replace the Republic's statute law by shari'a' or the establishment of 'a plurality of legal systems'.These accusations have been fiercely rejected by Refah's applicants and, more convincingly, in the response by the AK Party to the indictments in April and June 2008.Justice Sawant concludes that the Indian Constitution 'by implication prohibits the establishment of a theocratic state' (Bommai §146) and Justice Reddy, responding to the challenge that secularism would be a vacuous or phantom concept, stresses that 'non-establishment' was an 'imperative in the Indian context' ( §308).Four clarifying remarks seem to be in order: First, in a historical and empirical perspective, it is obvious that prolonged 'church-state' battles have been a Western predicament (Robertson), that Western states in early modernity have not been secular states (cuius regio/dominio, eius religio 35 ) and that, even recently, none of the existing states (not even the US that serves as a model for many Indian Justices) is a strictly secular state.A secular state, then, if anything, is a utopia. 36econd, states that perceive themselves and are widely perceived as secular (e.g.Nazi Germany, the Soviet Union, Iraq under Saddam, Cambodia under Pol Pot) have violated the requirements of minimal morality as often as so-called 'religious states'.The 'secularity' of the state, if anything, is certainly not the same as a normatively praiseworthy 'secularism'. 37hird, from a minimalist liberal-democratic constitutionalism perspective, it seems quite unclear what the 'secularity' or the 'autonomy of the state from (organized) religions' exactly requires.In my view, it requires a minimal threshold of institutional, organizational, and role differentiation.Does it also require 'non-establishment' and 'disestablishment' or is it compatible with 'weak establishment' in countries with liberal-democratic constitutions (as in the UK, Norway, Denmark) or 'plural establishment' (as in Finland)?Does it require a 'strict separation of state from religions' (then no liberal-democratic country would qualify) or even a more demanding 'strict separation of politics from religions'?Does it require exclusive state law in every regard or is it compatible with private religious law under specific conditions?(see infra: 'Secularism 8').'Secularism 1' or 'secularity' is clearly not informative to decide these matters and, if so, it should certainly not be mistaken for liberal-democratic constitutionalism.
Fourth, liberal-democratic constitutionalism does not only require a certain threshold of differentiation of the state from religions -the famous 'first autonomy' -but also a 'second' autonomy, the relative autonomy of (organized) religions from the state.Self-declared 'secular states' (e.g.Turkey, France) extensively violate this second autonomy (state supervision and control of religions, see Turkish Constitution Articles 24, 136 and TCC). 38This version of 'secularism 1' is inimical to external freedoms to practise religion and to associational religious freedoms; it is clearly incompatible with liberal constitutionalism.This 'illiberal' character of the Turkish Constitution and the TCC rulings is so often and excellently criticized that I need not document or repeat it. 39The Indian Constitution and most Justices are also clearly opposed to these pitfalls of 'secularism 1'.

Secularism 2: Equal associational freedoms and collective toleration
Collective toleration of and associational freedoms for religions are certainly not a 'modern', 'Western', 'liberal' invention but have been guaranteed in the Ottoman millet system, 40 in preliberal arrangements in early modern Europe, 41 in Ashoka's Rock-edicts (infra) as well as by Akbar in 'premodern' India. 42They are as important as individual freedoms. 43Liberal constitutionalism, however, requires equal legal associational freedoms for religions and nondiscrimination (not fully guaranteed even in the late Ottoman Empire) as a basic principle.'The state is enjoined to accord equal treatment to all religions and religious sects and denominations' (Reddy in Bommai §146).Most Justices of the Supreme Court in India explicitly defend equal legal treatment for all as part and parcel of 'secularism', criticize its restrictions to 'individuals' in the 'private sphere', and also go beyond merely formal equality before and in the law in favour of an 'ameliorative' understanding of 'secularism'.In contrast, the Turkish Constitution and the TCC notoriously restrict associational freedoms but also do not guarantee equal treatment for minority religions, particularly Islamic minorities (see supra, note 22), in short predominant Turkish secularism violates the conditions of freedom and equality, combined in the liberal requirement of equal associational freedoms.

Secularism 3: Freedom of conscience and individual religious freedoms
The protection of freedom of conscience and of individual religious freedoms against the state, against all majorities -secular as well as religious majorities -and all minorities (and their organizations) is a specific 'modern' requirement of liberal constitutionalism.Yet, individual and associational religious freedoms are often in tension with each other and this 'Janus' face is an inherent characteristic of liberal constitutionalism (Levey, Rosenblum).In three prominent and characteristic ways, the conflicts between these principles are neglected or downplayed.
First, by declaring that religious freedom 'is primarily a matter of individual conscience' in 'the private sphere to which Turkish law confines religion' (TCC, backed by ECtHR 2003 §127; the Refah and AK parties explicitly affirm equal individual freedoms and criticize these restrictions).In the tradition of aggressive 'French' Enlightenment secularism, associative freedoms as well as external individual freedoms of religion -the 'freedom, either alone or in community with others and in public or private, to manifest his religion or belief, in worship, teaching, practice and observance' (ECHR Article 9) -are neglected or severely curtailed.Aggressive secularists are usually more interested in protecting individuals from any religion = superstition than in protecting individual religious freedoms.
Second, by massive and illegitimate state intervention in the internal affairs of religious associations even in core issues of belief and practice (as is routinely declared legal by French and Turkish Courts and recently also by the ECtHR).
Third, by the doctrine and practice of unconditional or 'absolute deference' of many American Courts that do not even protect the most basic rights of minors, women and dissenters within religious (majority or minority) groups and organizations.The Indian Constitution and Supreme Court are quite outspoken in this aim, but have a hard time in guaranteeing equal 44 The protection of exit rights (in opposition to all religious doctrines and practices penalizing apostasy) and also actual exit options are particularly important (see

21
individual religious freedoms for Hindus and Muslims. 44The comparative history of jurisprudence in hard and softer cases, in which individual and associational freedoms conflict, abundantly demonstrates that it is difficult to find reasonable, fair, context-sensitive balances 45 but an appeal to an unspecified principle of 'secularism' certainly does not contribute to doing so.

Secularism 4: Protection of religious minorities against unbound democratic majoritarianism
Liberal constitutionalism requires, among other things, the protection of individual and associational civil rights and of minority rights.As such, it is generally incompatible with unbound democratic majoritarianism (liberal-democratic constitutionalism is constitutional democracy, not just majority rule).This is from my perspective particularly important in two situations: First, in cases of aggressive secular majorities -and even more so in cases of aggressive secular ruling elites not forming majorities as in Turkey -threatening to seriously restrict equal religious freedoms of religious majorities and minorities, and, secondly, in countries with potential or actual totalistic, illiberal religious majorities which threaten -by a 'majority decision' -to restrict equal freedoms from religion (of 'secular' or better 'non-religious' people) thereby restricting the possibility not to believe in and practice any religion as well the freedoms of religious minorities.The second case is, according to Kemalist elites, already typical for Turkey under the Refah coalition Government (see ECtHR 2001 § § 59, 60) and particularly under the recent super-majority AK Party Government, though the AK Party rightly and fiercely rejects any such accusation of restricting freedoms or imposing some Sunni Islamic way of the good life.It may be more plausible for India to the degree that the Bharatiya Janata Party would follow radical 'Hindutva' policies and would gain majorities which, fortunately, seems not to be the case.In both countries, the principle of 'secularism' is seen to be the appropriate and necessary response. 46My concern with this proposal is that 'secularism' at the same time is appealed to as the main cause of the problem (as in the first case for Turkey) and as its ultimate remedy.Instead, we have much better concepts to exactly describe what is required: 'minority protection' is part and parcel of liberal constitutionalism. 47Liberal-democratic constitutionalism is different from 'ancient' and from other non-liberal types of democracy.As stated in the introduction, it is a historical and theoretical compromise full of internal tensions, and the tension between majority decisions and constitutional limitations is one of the most famous ones.Liberal constitutionalism, and not -as Ramashwami, under the misleading title 'Democracy and Socialism', thinks -'democracy stands for freedom of conscience and belief, tolerance and mutual respect' (Bommai §176).
Before turning to democratic constitutionalism I summarize the core elements of liberal constitutionalism with regard to religion.Liberal constitutionalism requires, first, a non-theocratic or non-religious state in the sense of a minimal threshold of institutional, organizational and role differentiation between state and (organized) religions.Second, it requires a minimal threshold of autonomy of (organized) religions from the state, equal associational freedoms of religions and collective toleration.Third, it requires freedom of conscience and individual religious freedoms (to belief and practice religions individually and collectively, in 'private' and in 'public') as well as freedom from religion.Fourth, it requires the protection of religious (and other) minorities against unbound democratic majoritarianism.These four pillars of liberal constitutionalism are phrased in appropriate constitutional terms; calling them 'secularism' 1-4 clarifies nothing and brushes over the potential and actual tensions and incompatibilities between liberal constitutionalism and secularism.

Secularism 5: popular sovereignty and protection against political discrimination:
'secular democracy' or 'political secularism' 'Modern' democracy, again different from ancient but also from classical liberal democracy, stands for 'vox populi vox dei', for the equal political status of all citizens and for equal political rights in opposition to all ascriptive discriminations, particularly also on religious grounds.This is often called 'political secularism' or 'democratic secularism'. 48As in the case of the secularity of the state, one can call any democracy that does not discriminate on religious grounds a 'secular democracy'.Then modern democracy is secular by definition: 'It is generally agreed that modern democracies have to be "secular"'. 49Secularism then means 'popular sovereignty' 50 or the condition that all defenders of 'absolute' truth claims, religious as well as secular ones, have to solve their fundamentalist dilemma that their 'absolute truths' are treated as 'opinions' when it comes to democratic decision-making. 51In Turkish and Indian constitutional debates and jurisprudence, secularism is either identified with or seen as a necessary precondition of political equality and democracy.Yet, this is neither self-evident nor clear in its meaning.It is, indeed, 'quite easy to conceptualize secularism and egalitarianism as distinct properties (as some justices did) but Reddy and Gajendragadhkar emphasize the connection between the two'. 52In interviews 'Justices Sawant and Verma confirmed that their understanding of secularism stands separate from the egalitarian political aspiration'. 53Identifying secularism and democracy or declaring their deep harmony 54 makes it impossible to see and investigate serious tensions.In my view, 'secularism' can be and actually is inimical to modern, democratic constitutionalism at least in four respects and, ironically, all these serious restrictions of democracy are legitimized as being necessary to 'defend' democracy.
First, Kemalist secularism as an aggressive, elitist, revolutionary, top-down strategy of 'modernizing' Turkey has been incompatible with both minimal institutional requirements of 55 The ECtHR, contrary to its assertion that the 'margin of appreciation left to states must be a narrow one', particularly 'where the dissolution of political parties is conceived, since the pluralism of ideas and parties is itself an inherent part of democracy democratic constitutionalism -a free and equal political process and multiparty competition.It has been criticized so convincingly (also by Refah and AK) that I need not repeat the main arguments against this variety of 'secularism' inimical to the first core element of democratic constitutionalism.
Second, more or less massive restrictions of the freedoms of political association, political parties and elected MPs, assemblies and (state) governments that survived earlier Kemalist regimes in Turkey and also characterize militant democracy in India.This variety of secularism is inimical to the second core element of democratic constitutionalism.
Third, secularist restrictions of free speech in both countries.This variety of secularism (infra: 'secularism 6') is inimical to the third core element of democratic constitutionalism.
Fourth, secularist restrictions of the 'pluralist', 'non-foundational' character of modern democracy.Varieties 10-12 of 'secularism' (infra) are inimical to the fourth core element of democratic constitutionalism.
The TCC, the ECtHR and the Indian Supreme Court all underwrite and praise the freedoms of political communication and association (ECHR Articles 10 and 11) as cornerstones of 'a pluralist, open, dynamic and lively democracy'.As in the case of all other rights, these freedoms are not 'absolute' but exceptions were to be 'construed strictly: only convincing and compelling reasons could justify restrictions' (ECtHR 2003 § § 57, 61, 99, 100; strictly circumscribed, closely scrutinized, compelling evidence, proportionality; see Sawant in Bommai §106).It is generally accepted among constitutional lawyers that serious threats to 'public order' and 'incitement to or use of violence' qualify as reasons for restrictions: 'protection of public safety, national security and the rights and freedoms of others and the prevention of crime' (ECtHR 2001 §40).Even prima facie, it is rather dubious what this all has to do with 'secularism' but both the TCC 55 and the Indian Supreme Court use assumed violations of the principle of secularism as a paramount reason to legitimize the constitutional dissolution and banning of religious political parties: 'If the Constitution requires the State to be secular in thought and action, the same requirement attaches to political parties as well.The Constitution does not recognize, it does not permit, mixing religion and State power.Both must be kept apart.That is the constitutional injunction.None can say otherwise so long as this Constitution governs this country.Introducing religion into politics is to introduce an impermissible element into body politic and an imbalance in our constitutional system.If a political party espousing a particular religion comes to power, that religion tends to become, in practice, the official religion.'(Justice Reddy in Bommai § 310, and conclusions point 10). 56ilitant democracy and 'secularism' form a rather aggressive anti-democratic mix prohibiting, without any further consideration, all religious political parties and allowing elected MPs, legislative assemblies and elected governments to be dismissed.

Secularism 6: Freedom of speech or 'exclusivist secularism'?
A crucial element of the freedoms of political communication are freedoms of speech (broadly understood) guaranteed by international covenants and liberal-democratic constitutions (also in Turkey and India (Article 19(2) or Section 153-A of the Indian Penal Code) against statist, majoritarian or elitist restrictions.The relevant, neatly circumscribed reasons for legitimate restrictions that are almost universally accepted by constitutional lawyers are incitement to violence ('clear and present danger') and serious, discriminatory 'hate speech'.Prima facie, the distinction between 'religious' and 'secular' speech does not seem to matter because secular speech can be as hateful and incendiary, promoting 'disharmony or feelings of enmity, hatred or ill-will' (Verma in Prabhoo) as religious speech.Still, it is a peculiarity of Indian (Constitutional) Law and jurisprudence to explicitly single out and forbid 'religious rhetoric' and 'religious corrupt practices' assuming, by implication, that all religious speech promotes enmity or violence, is opposed to equal citizenship, 'covers injustice', 57 arouses 'irrational', 'blind and disruptive passions' (Prabhoo 150) 'depriving people of their powers of rational thought and action'.
By contrast, then, 'reasons' are seen intrinsically as 'secular', as in 'exclusivist secularism' they are defended not only by many liberal philosophers but also by Judges such as Sajó and Verma. 58All varieties of exclusivist secularism, in my view, are clearly incompatible with constitutional and legal guarantees of freedoms of communication.Instead of presenting 'inclusive secularism' as an alternative and, by doing so prolonging the ambiguity of 'secularism'-talk, we should spell out the details of freedoms of communication and the closely circumscribed reasons for legitimate restrictions.

Secularism 7: Is a minimally social or a demanding 'socialist' state inevitably a 'secular'
state?Secularism-talk reproduces the under-determinacy of 'equality' as formal equality before the law, as more demanding equal political opportunities (particularly in caste-ridden and class-ridden societies), and as more substantive equality in general.Indian 'ameliorative' secularism would then be 'one facet of the right to equality woven as the central golden thread in the fabric' (1994, 6 SCC 360 (Ayodha case) at page 402, quoted in Prabhoo).A secular, democratic state apparently requires an 'orderly march towards an ideal egalitarian social order' (Ramashwami in Bommai §176).
The most minimalist understanding of substantive equality requires an active fight against inequalities of caste and of sex, legitimized by 'Hinduism'.There is broad agreement among most justices that this reform is an urgent requirement and that it is 'unimaginable without the direct intervention of the state in the spiritual domain' 59 though there remains some disagreement as to whether the 'amelioration of social injustices' requires 'secular teaching' 60 or whether the Hinduist tradition provides also internal resources to criticize religious legitimations of the 'status quo of religiously based hierarchy', as Gandhi, amongst others, forcefully argued.A more demanding interpretation requires that all citizens/residents equally, i.e. without any ascriptive discrimination, 61 have a basic subsistence right (not just charity) to be protected against 'malfare'. 62Still more demanding is Ramashwami's interpretation: The Constitution has chosen 'secularism as its vehicle to establish an egalitarian social order' and to 'secure to all its people socioeconomic needs essential for man's excellence and of moral wellbeing, fulfillment of material prosperity and political justice' (Bommai §186).Yet, even this would not require a fully egalitarian distribution of all important resources and rewards (usually meant by 'socialism' referred to in the 42 nd Amendment).
In the end, it remains completely indeterminate what 'a truly secular, or socially just society' 63 would require and what is meant by the 'ideal of the welfare state' (Reddy with Gajendragadkar, in Bommai §305).We have to live with Jacobsohn's quite disappointing conclusion that 'India will need to calibrate' 64 and find responses to the critics to 'explain exactly what this un-amendable feature actually means'.Indian ameliorative secularism is inherently ambiguous.

Secularism 8: 'Strict separation'? 'Strong' or 'moderate', 'positive', 'benevolent', 'ameliorative' secularism?
As we have seen above ('Secularism 1'), the secularity of the state rules out theocracy and exclusive, encompassing 'religious law' but the principle of secularism seems institutionally void or compatible both with 'strict separation' and with various forms of 'selective cooperation' between state and (organized) religions.Given all the attention paid to the principle of secularism by the Indian justices, it is astonishing that one does not find a serious discussion of the available institutional options although there is a clear preference for a combination of constitutional nonestablishment and some variety of selective cooperation, or a 'positive' secularism of 'principled distance' that is critical of purely 'negative' secularism, formal equality before the law and a 'wall of separation': the 'ideal of a Secular State in the sense of a State which treats all religions alike and displays benevolence towards them is in a way more suited to the Indian environment (…) than that of a truly Secular State'. 65Neither the ideal/myth of the 'American', nor that of the 'French' or the 'Turkish' model of secularism -in all cases, the gap between myth and reality is huge -is endorsed by the majority of justices. 66In order to understand what 'benevolent', 'passive', or 'ameliorative secularism' constitutionally require, we would have to spell out the advocated privileged, if any, legal status of (organized) religions; the relationship between civil marriage and divorce law (if any) and religious private law (if any); the rules and regulations regarding the autonomy of religious organizations, the financing of religions, of faith-based educational and social service institutions etc. 67 In all these regards, history matters, numbers matter, majority-minority relations and entrenched inequalities matter, at least for all critics of 'strict separation' secularism.The principle of 'secularism' dissipates into competing and incompatible 'secularisms'.In itself, it cannot be informative in all these respects.
'Strict-separation secularism' prescribes 'a uniform civil code' and proscribes the 'plurality of legal systems' in general, of religious personal law in particular, as is evident from the TCC's ruling in the Refah case.What does liberal-democratic constitutionalism require?The Grand Chamber of the ECtHR in 2003 not only 'missed the opportunity to analyze in more detail the concept of a plurality of legal systems, which is linked to that of legal pluralism and is wellestablished', particularly in private law (Justice Kovler in his concurring opinion), it also directly introduced into the European Convention System a highly controversial and contested French/Turkish understanding that any form of (religious) legal pluralism in private law is incompatible with 'liberalism and democracy'. 68The Court also missed the opportunity to clearly circumscribe the general limits of legal pluralism in constitutional democracies.I agree with Macklem's reconstruction of the 'three baseline conditions' in the ECtHR 2001 and 2003 rulings.First, the advocacy and introduction of a plural legal order 'must provide individuals with the freedom to choose whether to be bound by the norms of the distinctive and embedded communities to which they belong or by state law on the same topic' (freedom of entry/exit and choice of law).Second, the 'scope of lawmaking authority (…) must be limited in range and effect' (to private or personal law).Third, a plural legal order 'must respect and retain the state's role as a democratic guarantor of individual rights and freedoms' 69 or, more narrowly circumscribed, the state guarantee of -and court intervention in case of violations of -'basic interests' and the core of rights demanded by minimal morality. 70rt.44 of the Indian Constitution: 'The State shall endeavour to secure for the citizens a uniform civil code' is also seen as an expression of 'secularism' (Justice Sahai in Mudgal 1995).However, not only 'secularists' deplore the fact that India (just as Israel) has been unable to realize a civil code, because liberal-democratic constitutionalism requires one as an alternative to existing religious marriage and divorce laws and courts.It should also spell out the three baseline conditions for permissible private personal law.The latter has been partly achieved through important internal reforms in the case of Hindu, Christian, and Parsi, but not Muslim Personal Law. 71

Secularism 9: 'strict neutrality' or 'principled distance'?
Second-order principles, as stated above, are unavoidable yet highly abstract and indeterminate meta-rules which -together with legal or constitutional paradigms -may help interpret and balance conflicting legal principles and rights.One may call them the constitutional 'creed' or 'faith' (Sawant, Bommai §151) or 'constitutional philosophy' (Ahmadi, Bommai §29).My point is not to discard or denigrate them altogether.Rather, my point is that "secularism" is not the right kind of 'spirit', nor is it the right kind of second-order principle or legal paradigm.Elsewhere, I have extensively argued 72 this point and why 'secularism as strict neutrality' (so often resorted to by the TCC) should be replaced by three related second-order principles: (i) embedded impartiality vs. 'a view from nowhere'; (ii) relational neutrality vs. 'strict' or 'difference-blind' neutrality; and (iii) fairness as even-handedness in matters of culture vs. fairness as hands-off.Although the Indian Justices do not clearly distinguish between or discuss these principles, often identifying them with 'secularism' or lumping them together, my claim is that they would actually be quite sympathetic to such a re-conceptualization because they allow one to clarify much better what 'principled-distance secularism' (Bhargava), 'positive' or 'benevolent neutrality' (Reddy in Bommai §304) actually mean and require.Space prevents me from substantiating these claims.

Secularism 10: 'ethical' secularism?
Though modern, pluralist democracy cannot be strictly anti-perfectionist, it should be as 'antiperfectionist' as possible, 73 it should not explicitly or implicitly privilege a specific way of a good life, e.g. a comprehensive liberal ethics or a 'secular', 'humanist' ethics.The TCC's position that secularism in the Constitution 'is a civilized way of life' (Refah decision, p. 26) is not only rejected by all 'political secularists' in philosophy (such as Anne Philipps, Tariq Modood and Rajeev Bhargava), it is also adequately criticized in the AK party's response which clearly distinguishes secularism as a 'political and legal concept' from secularism as a lifestyle: a pluralist society is characterized by a 'wide diversity of lifestyles' that should be 'freely chosen': 'the right to the free determination of our way of life is one of our basic rights'; the role of the state is to guarantee that the 'different lifestyles -including different religious faiths and practices -do coexist peacefully', to 'prevent the hegemony of one lifestyle'; the imposition of secularism as a lifestyle by the state would result in a totalitarian society and an 'authoritarian state'. 74

Secularism 11: Foundational or Justificatory Secularism?
Justificatory or foundational secularism is here meant to indicate exclusive secularist attempts to justify liberal-democratic constitutions, institutions or politics/policies whether the latter are called 'secular' and are said to require 'principles of secularism' or not.This 'independent political ethics mode' 75 is defended in philosophy from Bayle and Spinoza to Audi and Habermas; it is also clearly endorsed by the TCC.It competes with what Taylor has called a 'common ground strategy' of justification (from Pufendorf to Mahatma Gandhi or Hans Küng) that 'leans on all existing religions'. 76The third alternative, 'Rawlsian' strategy, a nonfoundationalist wide and deep Overlapping Consensus -allowing people to subscribe to constitutional essentials for whatever reasons, religious or secular, they find compelling, 'only let them subscribe' 77 -which is in my view preferable for many reasons, 78 does not show up in any of the justices' opinions.
Liberal-democratic constitutionalism does not require any deeper, 'metaphysical', 'philosophical' or 'theological' foundation (this is what I call the 'priority of LD' over all competing theoretical foundations. 79Modern, pluralist liberal democracy is an open project helping to prevent the political form of a society from being seen as the realization of a transcendent vision or an 'ultimate philosophical foundation'.'Secularism' also in this regard is rather parochial (secularist or humanist 'Enlightenment exclusivism') and is potentially or actually inimical to liberal-democratic constitutionalism.

Secularism 12: a generalized 'meta-narrative', a 'symbolic universe'
The TCC states quite explicitly: 'the opinion that secularism is, in fact, the final phase of philosophical and organizational evolution of societies is also being shared in theory'.It can neither 'be narrowed to division of religion and state affairs' or 'separation of state and religious affairs in a narrow sense', nor should it be only understood as the particularist 'philosophical essence of life in Turkey'. 80It has to be understood quite universally and generally as 'the cornerstone of rationalism' and 'science' and also as the only possible normative foundation of general 'concept(s) of freedom and democracy', of 'independence, national sovereignty and the ideal of humanity' (TCC 26), of any 'civilized way of life'.In all these aspects, it replaces 'religion' and 'the dogmatism of the Middle Ages' by 'scientific data'; 'values based on reason and science replaced dogmatic values' (ECtHR 2003 §40).ECtHR 2001 §71 summarizes (and also seems to endorse) the TCC's view in its neat opposition between the mythical notion of shari'a as the 'antithesis of democracy in that it was based on dogmatic values and was the opposite of the supremacy of reason and the concepts of freedom, independence, and the ideal of humanity developed in the light of science' (my italics).In a quite similar vein, Ramashwami argues that secularism 'represents faiths born out of the exercise of rational faculties', 'the imperative requirements for human progress', it 'improves the material conditions of human life, but also liberates the human spirit from bondage of ignorance, superstition, irrationality, injustice, fraud, hypocrisy and oppressive exploitations (…) the whole course of human history discloses an increasing liberation of mankind, accomplished thought, all is covered by the term secularism' (Bommai §179).
'Secularism' here serves as a stand-in for the broad range of distinct but interconnected elements fairly well-known from 'modernization' theories/ideologies telling one linear, dichotomist, evolutionary story of 'progress, reason, rationality, science, civilization, humanity, freedom, equality, and democracy' which covers at the same time the 'epistemological' (cognitive and normative) as well as the social and political preconditions of liberal-democratic constitutionalism. 81This outdated conception of one exclusive foundational meta-narrative, incompatible with the open character of modern democracy sketched above, is also convincingly criticized by the AK Party: constitutional or 'legal and political secularism' is not a 'philosophical belief', the 'positivist secular ideology is outmoded'; not only Comte's 'scientific religion' but scientism generally, as an ideology 'replaces monotheistic religions with "mono-science" (…) it is easily converted into a convenient means for totalitarian ideologies to legitimize their antidemocratic objectives'. 82n short, constitutionalizing secularism as a meta-narrative seduces constitutional Courts both in Turkey and in India to declare 'rationalism' or 'scientism' as a general foundation and/or a universally accepted 'ideal', a competing 'Weltanschauung'; to officially pronounce 'ethical secularism' as the only acceptable or possible 'foundation' of liberal democracy; to construct all religion as dogmatic and inimical to humanity, civility, liberalism, democracy; and to speculate on the 'true', 'essential' meaning of religions declaring all contrary official statements of religious parties and representatives as a 'pretext'.

How did 'secularism' work in the selected cases?
In my brief contextual evaluation of the Turkish cases I focus on the Refah judgment because it provides the opportunity to take the rulings of the ECtHR into account and because the substantive arguments by the TCC in the AK Party case have been exactly the same although the outcome has been different, mainly for prudential reasons of state in order to prevent a huge constitutional and political crisis.Refah's applicants maintained that the party 'had consistently observed the principle of secularism'.This can be interpreted as an endorsement of 'secularism' for purely strategic reasons -because it is part of the Turkish Constitution and not endorsing it would mean quasi-automatic party dissolution -or as an alternative normative understanding of 'secularism'.The applicants clearly claimed the latter.In their interpretation the principle requires the relative autonomy of the state from (organized) religions and the relative autonomy of religions from the state together with full religious freedoms for all.They 'asserted that the principle of secularism implied respect for all beliefs and that Refah had shown such respect in its political activity' (EC 2001 §15).'Applicants accepted in principle that (legitimate aims) might depend on safeguarding the principle of secularism' ( §41).They criticized certain interpretations of the principle 'in the name of respect for freedom of conscience and freedom of expression, it had never advocated abandoning the principle itself or the Turkish constitutional order in general' ( §54; see EC 2003 §16, §19).This interpretation of religious freedoms seems prima facie much more in line with the European Convention System.It is not astonishing that 'the government did not believe that Refah was content to interpret the principle of secularism differently.In their submission, the party wished to do away with the principles altogether.This was evidenced by the submission made on Refah's behalf during the latest debates on amendment of the Constitution, since Refah had quite simply proposed deleting the reference in the Constitution to the principle of secularism.'The TCC clearly followed this interpretation, as could be expected.The main reasons for this have been that Refah would aim for theocracy and the installation of Shari'a and hence, knowing that this could not be done openly, had to 'hide' their 'real intentions' by a purely instrumental adoption of 'the principle of secularism', using 'the method know as "takiyye" which consisted in hiding its beliefs until it had attained that goal' (EC 2001 §59).Yet it is remarkable that the ECtHR fully endorsed this interpretation without even starting to discuss the possibility of such an alternative interpretation: 'on the pretext of giving a different meaning to the principle of secularism' (2001, Press Release; EC 2001 § § 59,  60, 77; 2003  § 105).
Both the TCC and the chambers of the ECtHR referred extensively to the historical and recent context in their arguments as to why the ban on Refah was prescribed by law, necessary, legitimate and proportional.The Turkish Government claimed that the principle of secularism is 'particularly important for Turkey in relation to other democracies.The Republic of Turkey had been founded as a result of a revolutionary process, which had changed a theocratic State into a secular state, and reactionary Islamic tendencies were still a danger in the present day (…) Political Islam.'The Government referred to the fact that 95% of Turkish citizens are Muslims: hence 'the abusive use of religious ideas by politicians was a threat to, and a potential danger for, Turkish democracy' ( §60), mentioning the danger of an export of neighbouring theocratic regimes (Iran, Saudi Arabia) to Turkey and their moral and financial support for fundamentalist movements, the rhetoric of waging holy war (jihad), and so on.Turkey is said to be the only Muslim country with a liberal democracy after the Western model.The TCC followed these arguments in all respects and the ECtHR 2001 and 2003 also endorsed this judgment, particularly with regard to the limits of freedom of speech and association -'any conduct which fails to respect the principle [of secularism -VB] cannot be accepted as being part of the freedom to manifest one's religion and is not protected by Art.9.' (EC 2001 §42) -and with regard to necessity and proportionality: the Court finds 'that the establishment of a theocratic regime (...) is not completely inconceivable [my italics -VB] in Turkey' ( §65) repeating all elements of the Government's claim: Muslim majority; the 'relatively recent history' of 'Ottoman theocracy'; the refah party did not exercise 'power alone' and hence had to hide or had not had 'the real chance to implement its political plans' (EC 2001 §77; 2003 §107); opinion polls made plausible that it would gain 67% of the votes in the next general election.
Obviously, the courts are right in taking 'account of the historical context' (EC 2003 §105) because, as stated in my Introduction, not balancing conflicting constitutional principles and rights is possible without doing so.The courts are also right that it would not only be politically, but also jurisprudentially, irresponsible not to take threats to liberal-democratic constitutionalism seriously.However, the courts failed blatantly in the following main respects.
First, in the way in which they took 'history into account': They repeated the neat dichotomy of 'theocratic Ottomanism' versus 'liberal and democratic' Kemalist secularism (EC 2003   § §125, 126).Both are historically indefensible constructs.They made history into destiny by neglecting or not considering the slow but remarkable development of 'political Islamism' in Turkey, in the history of Refah and the AK Party in particular, as any real historical contextualism would require.The 'progression over time' (EC 2003  §109) for the Court means that some old speeches by some members of the party are more important than recent explicit statements and acts by the party in (coalition) Government.
Second, the task of the courts is not to judge whether the establishment of a theocratic regime 'is not completely inconceivable' but to gather all compelling and convincing evidence that there was a 'real' or 'genuine threat', as Refah's applicants urged the Grand Chamber to do referring to the 'clear and present danger' test of the US Supreme Court (EC 2003  §14, 76), endorsed by the dissenting judges (EC 2001, 30).The arguments provided do not live up to any strict scrutiny test and do not provide 'any compelling and convincing evidence'. 83hird, in the light of these arguments, the measures (the party ban, assets transferred to the treasury, a five-year ban on RP leaders) seem to be clearly out of proportion.
Fourth, the weighing and balancing of the conflicting principles of the 'protection of public safety, national security and the rights and freedoms of others and the prevention of crime' (EC 2001 § 40) as legitimate aims to restrict freedoms of political communication (ECHR Articles 9 and 11) is absolutely one-sided and unreasonable given the fact that any compelling evidence of the imminent threat of violence is lacking and, hence, the 'sole ground' for the dissolution of Refah is the alleged violations of 'secularism' (EC 2001, dissenting judges) and not of rights and institutions of liberal-democratic constitutionalism.'Secularism' distracts from or prevents a considered contextual analysis and a convincing reasonable balance of conflicting rights in the Turkish cases: contextually incorrect judgments for the wrong reasons.
Calling Muslims 'snakes' and 'intemperate campaign rhetoric' 84 such as 'you will find Hindu temples underneath if all the mosques are dug out', or: 'we are standing for Hinduism.Whatever Masjids are there, if one starts digging the same, one will find Hindu temples under the same' by Thackeray, the leader of Shiv Sena at the time of the dismantling of the Babri Masjid in Ayodhya in December of 1992, is not only seriously discriminating ('hate speech') but also a direct incitement to violence, legitimate reasons to overrule free speech and other constitutional protections.The judgment of the Indian Supreme Court seems contextually right to meproviding 'compelling and convincing evidence' -but it is so for the wrong reasons because if 'secularists' would have incited communal violence at this time in India, it should and would not make any difference. 85n sum, in the Turkish and the Indian Supreme Court rulings, 'constitutional secularism' is not only rather indeterminate, it is often also rather inimical to liberal-democratic constitutionalism.In the Turkish constellation 'the laicist categorizations and dispositions available to European and Turkish judicial authorities' imply that there are only two available alternatives: 'benevolent secular democracy (in its Kemalist form) or menacing Islamic theocracy (overturning Kemalism).The judges dismissed the possibility of a reformulated or entirely different instantiation of secularism as a "pretext", arguing that the real intention of the party was to "institute Islamic law"'.86 Yet before endorsing a 'different form of secularism' or an 'alternative' normative secularism -as Hurd does -allowing the Refah and AK parties to 'oppose Kemalism' while supporting secularism, it may be a good idea to pause for a while for two reasons.First, one has to clearly see that religious parties are then forced to accept 'secularism' without good reasons instead of focusing on the really important questions of whether and how they endorse the minimal requirements of liberal-democratic constitutionalism.This does not promise stable doctrinal, attitudinal and institutional learning because such structural power asymmetries tend to create structural resistance.Second, the endorsement of an alternative or different principle and form of secularism would be plagued with exactly the same kinds of vagueness, ambiguities, inconsistencies and tensions as the predominant varieties.So, before praising the defensive attempts by the Refah and AK parties as 'a series of efforts to grant cultural and historical legitimacy to alternative models of religious separation and accommodation' or to 'refashion the secular' or 'to relocate the secularist settlement to a different position on the spectrum of theological politics' 87 it seems to be a better idea to clarify certain things: to get rid of the language of secularism in constitutional and jurisprudential matters.It is rather ironic that Refah's proposal to delete 'the reference in the Constitution to the principle of secularism' did not receive the attention and discussion it deserves, neither by the Courts, nor by radical critics such as Hurd.It is indicative of the performative power of words that none of the dissenting and concurrent opinions dares to even consider this.88 'Constitutionalizing secularism' also implies important strategic disadvantages.It distracts from the normatively relevant issues whether both 'secularists' and 'religious believers/ organizations' subscribe to minimal liberal-democratic constitutionalism.The generalized dichotomy 'secularists versus all religions' seduces one to choose the wrong friends (all 'secularists' instead of all defenders of liberal-democratic constitutionalism) and the wrong enemies (all 'religions' instead of all enemies of liberal-democratic constitutionalism, including illiberal and undemocratic Kemalist 'secularists').It forces religious political parties to defend (alternative versions of) 'secularism' 89 (accusing their political opponents of 'pseudosecularism') 90 and allows their opponents to describe this as a 'pretext' instead of requiring that both outspokenly subscribe to the detailed principles and rights of liberal-democratic constitutionalism.The ECtHR has clearly missed an important opportunity to allow for much less aggressively 'secularist' and/or 'alternative' interpretations of 'secularism' in all aspects (as defended by Hurd) 91 or, in my own words, to find in reformed political Islam an ally in fighting secularist authoritarianism of all kinds and in establishing constitutional states in Muslim majority countries without being forced to 'accept' any variety of 'secularism'.92 The Court could have contributed to stabilize and clarify processes of 'institutional' and 'doctrinal learning', already under way instead of reinforc-

33
ing securitized myths of Islam as being essentially inimical to liberal-democratic constitutionalism. 93

Conclusions: constitutionalizing secularism or contextualizing liberal-democratic constitutionalism?
In general, I agree with the view of the ECtRH that judgments have to be contextual and that national Constitutional Courts have a legitimate and necessary margin of appreciation.The task of the European Court is not to impose specific, nationally predominant readings of principles, let alone of specific institutional models -modes of governance of the economy, institutional varieties of welfare states, institutional patterns of democracy or, in this case, institutional models of state/politics and religion relationships.The ECtHR should indeed focus on 'constitutional essentials'.The same, in a different setting, is also true for countries' Supreme Courts. 94Indeed, the 'Court's task is not to take the place of the competent national authorities but rather to review whether Courts exercised its discretion reasonably, carefully or in good faith', whether the interference is justified and proportionate to the legitimate aim pursued, and whether reasons adduced by the national authorities to justify it are 'relevant and sufficient'.As already quoted, in cases of interference in freedoms of speech and association the 'margin of appreciation left to states must be a narrow one', particularly 'where the dissolution of political parties is conceived, since the pluralism of ideas and parties is itself an inherent part of democracy' (EC 2003 §81).Unfortunately, the Court did not make use of its possibilities to critically scrutinize the presented factual evidence in the Refah case.More important and far-reaching is the fact that EC 2001 and 2003 allowed the TCC, contrary to its own general opinion, a very wide and, in my view, completely illegitimate margin of appreciation in the restriction of two essential political freedoms of communication. 95y doing so, the Court itself endorses serious limitations of public debates and political initiatives, movements and parties that aim at deep and structural changes in the economic, societal, political and legal structure and institutions which are clearly compatible with the constitutional essentials or the core of liberal-democratic constitutionalism and which explicitly refrain from using force as a means of achieving these aims.Implicitly, by backing the TCC, the ECtHR declares as unconstitutional attempts to federalize the unitary republic, to institutionalize minority rights and protections for Muslim minorities and for national and ethno-linguistic minorities, to change the institutionalized relationships between the state and (organized) religions or to drop the contested 'principle of secularism' by constitutional amendment (see above for Refah's attempts). 96By declaring all these reforms unconstitutional, the Court itself contributes to the fact that these movements might consider 'unconstitutional' or 'revolutionary' changes, if need be also by force.
The TCC also freezes predominant interpretations and institutional settings that work usually in favour of the entrenched privileged powers.Constitutional courts and Justices -the supposedly impartial guardians of constitutional essentials -are then seen as biased and partial.Refah's applicants argued 'that the real reasons for Refah's dissolution' (EC 2001 §58) are not the protection of public order or liberalism or democracy but the interests of major Turkish companies, state borrowing, using state bureaucracy and the military, and, it should be added, the political interests of Kemalist elites (including judges) and obviously 'secularist' political parties having a massive interest in getting rid of successful political competitors.Indian provisions of President's Rule are always in danger of being used as a stick to beat competing political parties and state governments and this danger is clearly more serious if appeals are based on violations of 'secularism', as the counsel in Bommai stated: they are 'bluntly partisan acts'. 97ence, constitutional courts -under the guise of impartiality and neutrality -are tempted to use 'secularism' as a weapon in ideological and political battles and by doing so threaten the authority and legitimacy not only of the courts but also of the legal system as a whole.
In finishing, let me spell out some more general arguments on the tasks of constitutions and the limited but important role of judicial review and activism.
Constitutions strike a difficult balance of generality and specificity.Their provisions are by definition more general than those of laws and statutes.They even cannot avoid unspecified general clauses such as 'public order', 'morals' or 'proportionality' but these are defined by case law and should only serve to fill unavoidable under-determinacy gaps and certainly not take precedence over more specified principles.Constitutional courts may also have to resort to second-order principles and even legal paradigms in their interpretation of constitutionalized principles and rights and their context-specific balancing.Yet, as I have tried to show, 'secularism' is too malleable and indeterminate to serve as a valuable guide compared with embedded impartiality, relational neutrality and fairness as even-handedness in cultural matters, which also show a higher degree of under-determinacy than principles and rights of liberal-democratic constitutionalism, for me a reason why they too -as all other second-order principles -should not be 'constitutionalized' because they overstretch the tolerable margins of appreciation for courts and judges. 98However this may be, we should definitively not 'constitutionalize secularism'.What constitutions of liberal-democratic states should do, and most of them actually do, is to provide more specified guarantees of the requirements of 'minimal morality' and of the core of liberal-democratic constitutionalism.This core of liberal-democratic constitutionalism is 'thin' and 'minimalist' but 'strong'.
All actual constitutions, obviously, are much 'thicker'.We know from history and comparative experience that the core of liberal-democratic constitutionalism is abstracted from and does allow for a broad variety of institutional forms.The same basic principles, rights and minimalist institutional requirements -because they are, rightly, abstract and general -are in need of contextual specification and balancing.States and constitutions differ -necessarily and legitimately so for reasons of history and the context-dependency of institutional and cultural traditions -in the way they institutionalize the rule of law (e.g. common law versus codified law, the presence or absence of administrative and constitutional courts), democracy (e.g.majority or proportional representation systems; exclusively representative democracies versus democracies allowing for direct democracy; presidential versus parliamentary democracy), the require-ments of a 'social' or 'welfare' state (three or more varieties of welfare capitalism) or the requirements of economic orders compatible with 'freedom and democracy' minimally understood (varieties of capitalist and/or socialist market economies). 99These institutional varieties are all morally permissible as long as they do not violate the thin but hard core of liberaldemocratic constitutionalism. 100Exactly because they are much 'thicker', it is urgent that constitutions should explicitly allow for legitimate constitutional change that is compatible with the core of liberal-democratic constitutionalism instead of proscribing constitutional amendments (that require democratic supermajorities and lengthy and sophisticated procedures).If they do not -just as the Turkish Constitution and the TCC, but also some justices of the Indian Supreme Court -they proscribe democratically legitimate institutional and policy transformation and, by doing so, delegitimize both constitutions and Courts.
An implication of my argument is that the role of Constitutional Courts and Justices has to be specified.They should endorse judicial activism in all issues concerning the minimal core of liberal-democratic constitutionalism but should try to practice self-restraint with regard to all 'thicker' issues of legitimate varieties of the institutional structure of liberal-democratic polities and all transformations that are compatible with the core of liberal-democratic constitutionalism.These alternatives, as well as politics and policies trying to realize them, should be open to broad and free, contestational public debate and democratic decision making and courts should be wary of joining conservatives, libertarians, (neo-)liberals, but also radical democrats or socialists. 101ilitant democracy is one of the varieties of institutionalizing democracy.Quite in general, it has some serious difficulties regarding the core of liberal-democratic constitutionalism, which may be excused by specific contexts.In combination with constitutionalizing 'secularism', however, we get an explosive mix that tends to erode this core.

Table: 12 distinct secularisms varieties of secularism normative problems
Bader 2007, supra note 2, pp.218-222).45Bader 2007, supra note 2, chapters 4 and 5. 46 Bhargava, supra note 21; W. Menski, 'Indian Secular Pluralism and its Relevance for Europe', in R. Grillo et al. (eds.),LegalPractice and Cultural Diversity, 2009, pp.31-48; A. Alam,'The notion of secularism and the Supreme Court of India', unpublished conference paper, Utrecht, 25 May 2009:The core of the principle of secularism in the rulings of the Indian Apex Court and the worrying recent shifts.Reddy also argues that 'the dominant thinking appears to be that the majority community, Hindus, must be secular and thereby help the minorities to become secular'( §308).Instead of becoming secular Hindus have to be/become tolerant and learn to respect basic rights internally as well as with regard to other minorities, and this is what all religious minorities also have to learn.47 Jacobsohn, supra note 19, p.143, quotes Sushash Kashyap: 'If democracy leads you in one direction and secularism in the other, democracy should take precedence'.This statement is erroneous because 'democracy' is identified with the 'sanctity of majoritarian rule' or when democratic majoritarianism 'flouts constitutional principles' (p.138).However, if we would restate Kashyap's statement as: if liberal constitutionalism leads you in one direction and secularism in another, liberal constitutionalism should take precedence, then the 'conceptual intertwinement' of 'liberalism' and 'democracy' with secularism by Jacobsohn, Reddy and Gajendragadhkar loses its attractiveness.
93 See Bader, supra note 43, for criticism.94SeeEC 2003 §98: not to impute specific interpretations and specific institutional settings which themselves should be open to democratic change; and the concurrent opinion by Ress/Rozakis 2003: 'no exhaustive list of the rules of democracy' and 'very basic rights and freedoms' including campaigns to 'change rights and freedoms', so 'all depends on the specific rights and freedoms'.95 In its recent Lautsi v Italy ruling against the official display of crucifixes in Italian state schools, the Court imposed its emerging strict neutrality and strict separation doctrine, allowing Italy no margin of appreciation whatsoever (see Leigh, supra note 18, pp.15-19).96 The TCC decision in the 'headscarf case' in June 2008 is even more alarming.'Whereas the Constitution clearly restricts judicial review of constitutional amendments to procedural grounds, the Court ultimately reversed the properly ratified amendments on the substantive grounds that they violated the constitutional provisions on secularism.In its reasoning the Court not only expanded its own authority to review constitutional amendments on substantive grounds, it also placed new limits on the government's ability to amend the Constitution' (Bali, supra note 22, p. 4) and thereby 'effectively closed democratic channels for constitutional reform', e.g. for the draft proposal for a new civilian constitution by the committee headed by Ergun Ösbudun in the autumn of 2007 (pp.26 et seq.).