From Habermas to Derrida: A Weak Form of Secular Universalism

In this article, I argue that a universalistic thrust of secularism should not be located in a Habermasian deontological liberal principle of the priority of universal morality over particularistic ethical doctrines. I show that Habermas cannot plausibly demonstrate that this principle can be invariably applied across different cases. However, in order not to succumb to parochialism, the failure of the deontological model should not prompt us to give up on the search for a universalistic drive behind secularism. To this end, I advocate a Derridean critique of religion and secularism as an alternative solution. By deconstructing the Kantian dichotomy of faith vs. knowledge, Jacques Derrida shows that secularism is, paradoxically, both a concrete socio-political regime and a possibility for a radical change.


Introduction
In political philosophy, most often, moral universalism has been advocated from the perspective of neo-Kantian deontological liberalism, which offers uniform deontic precepts applicable across different contexts. Such moral precepts stand higher than particular religious or ethical doctrines, and are supposedly neutral. From the practical point of view, deontological liberalism has two profound shortcomings when it comes to the issue of secularism. Firstly, despite the fact that the stipulation that the state should not give official status to any religious doctrine is considered to be relatively uncontroversial in many liberal democratic countries, it has been clearly violated in some mature democracies, like Sweden, Denmark, and Norway, and is still not meticulously followed in England. As a noted sociologist of religion José Casanova (2011: 70-71) observes, even in laicist France, 80 % of the budget of private Catholic schools is allocated from the state funds. Secondly, extending uniform secularist principles all across the globe will encounter considerable difficulties, given the variety of political, societal, or cultural factors influencing the relationship between the church and the state in different societies. Given this political difficulty, some influential commentators argue that secularism that is historically a Western phenomenon needs a reworking in a new reality in order to accommodate global diversity (see Assad 2003;Taylor 2007). However, these commentators can be accused of being too deferential to particular cultural contexts, completely giving up on the context-transcending potential of secularism. 1 So, can there be a middle ground between deontological liberalism as applied to the question of secularism and its contextualist counterparts?
In this article, I will first argue that the deontological view of secularism propounded by neo-Kantian deontological liberalism does not live up to its own standards and needs to be substituted in order to save its universalistic thrust. In particular, deontological secularism cannot be plausibly extended invariably across different contexts without compromising its own universalistic nature. I take Jürgen Habermas's conception of secularism as a paradigmatic example of the deontological view. Unlike Rawls, Habermas (1998: 85;2017: 190-209) does not make critical concessions to modus vivendi. 2 Instead, he argues 1 Even those authors who almost entirely reject the universalistic inclinations of secularism, largely fail to explain to what extent their conceptions are devoid of universalism. And if there is a universalistic element in their theories, then what is the ontological source of the phenomena? For example, a noted critic of liberal universalism, William Connolly (1999: 156) believes that an ethos that drives deep pluralism does not stem from "a few general premises", but it is rather akin to a "political formation" or a "cultural artifact". In this connection, he needs to clarify to what extent his version of secularism that is rooted in "deep pluralism" is not parochialist. Certainly, there should be some force in secularism that is capable of transcending its cultural limits. 2 Rawls's position on the universality of secularism is ambiguous, to say the least. He considerably departs from the neo-Kantian deontological approach. Nicholas Wolterstorff notices that the problem with the Rawlsian argument lies with the fact that it attempts to draw its principles from the cultural repository of a given liberal regime (Audi et al. 1997). Rawls (1996: 46) believes that one can locate such principles buried in the political culture of liberal democratic societies, waiting for a political philosopher to uncover them. Wolterstorff rightly concludes that even the existing liberal democracies, including the United States, do not have a homogenous political culture that would be suitable for deriving universal principles needed for settling the most serious normative conflicts regime (Audi et al. 1997). Such a conception is all the more problematic to apply to different societies. In contrast, a properly deontological position contends that the neutrality of the state can be propped up by appealing to moral principles that are independent of comprehensive doctrines or cultural practices. that the Kantian principle of the priority of universal morality over particularistic ethical doctrines needs to be maintained as a cornerstone of secularism (Habermas 2010: 78-79). Note that Habermas approaches secularism both from sociological and moral philosophical perspectives but for the purposes of this article, I will be mainly concerned with the philosophical point of view. The priority of universal morality over particularistic ethical doctrines implies state neutrality towards various worldviews, including both religious and non-religious systems of belief. It is argued that since such a higher-order principle is devoid of particularistic implications, it can be freely extended to various contexts. Furthermore, as the argument goes, since it is not a particularistic ethical doctrine, the imposition of such secularism cannot be viewed as an act of coercion. This is because coercion implies forcing substantive ethical claims upon others. In contrast, deontological liberals argue that what is "imposed" is only a principle that professes neutrality vis-à-vis any type of particularistic ethical worldview.
I show that the principle of the priority of morality over particularistic ethical doctrines as a higher-order form of reasoning is untenable. I examine to what extent the normative claims of the Habermasian deontological secularism can be viewed as universal in every particular context. Such a supposedly universal claim is the priority of morality over particularistic ethical doctrines that is supposed to be defended in every case. I argue that the universality of the priority of morality over ethics can be defended neither on the level of negative nor positive freedoms. For this reason, if we want to retain the universal thrust of secularism, we need to find a conception of secularism that does not ignore the necessity of being grounded in particular cultural contexts.
I show that Derrida's deconstruction of faith and secularism can provide us with the type of universalism that is not detached from the particular contexts. Deconstruction does so by providing an aporetic conception of ethics that describes the general structure of secularism without offering a rigid ethical norm applicable to different particular contexts. Derrida (1992: 23) argues that a context-transcending ethical act is a singular event that cannot be known beforehand. Moreover, such a singular universality arises out of a particular context, rather than being completely transcendent.
In this article, I will first argue that the Habermasian deontological liberalism cannot provide a plausible universalistic standard that is applicable across different cultural contexts. To drive home this point, first, I will take a look at Habermas's works The Future of Human Nature and The Religion in the Public Sphere. I chose to examine these two works, rather than more extensive ruminations on religion and secularism, such as Between Naturalism and Religion or This Too a History of Philosophy, because here, Habermas specifically focuses on overcoming the difficulties inherent to the neoKantian deontological approach to secularism. Second, I will propose the Derridean deconstructive critique of faith as an alternative form of context-transcending secularism.

The Failure of the Deontological Model of Secularism
Deontological secularism can be viewed as a convenient framework to address the questions that the state encounters in regulating the lives of people sharing different religious or ethical outlooks. It considers the priority of universal morality over religious or ethical conceptions to be a golden standard that is in principle attainable, even though it is not meticulously observed in some Western countries nowadays. On this view, deontological morality is inherently universal, affecting all possible contexts invariably, while religious and other sorts of ethical doctrines are particularistic and they do not transcend a particular community or an individual. By examining Habermas's solutions to the difficulties associated with the deontological liberal approach to secularism, I show that the priority of morality over particularistic ethical doctrines cannot sustain itself as a universalistic standard. 3 To shed light on this point, I will test the justifiability of this principle as applied to (a) negative and (b) positive freedoms. Following Isaiah Berlin's famous definition, I define negative freedoms as those liberties that protect us from unjustified interference from the state or other individuals. Positive freedoms are those that pertain to citizens' rights to realize their political or other sorts of aspirations. In both cases, deontological liberalism as applied to secularism requires from the state to stay neutral vis-à-vis various religious and other sorts of ethical doctrines.

A) Universal Morality as Applied to Negative Freedoms
The first type of argument against deontological secularism concerns the negative freedoms, which are supposed to be distributed equally among religious and non-religious persons. Theoretically, citizens have a right to pursue their versions of the good without interference from the government or other citizens. However, there are cases where different versions of the good clash in such a way that it becomes impossible for the state to stay neutral. One oft-cited example, which is also discussed by Habermas, is abortion.
In addressing the question of abortion, Habermas starts by arguing that resolving the issue by resorting to a scientific argument about the exact definition of human life is a violation of the principle of state neutrality. He rightly argues that science, in this case, would act as a metaphysical authority that would determine something far beyond its purview (Habermas 2003: 31-32). In order to stay faithful to the original ethos of deontological secularism, the state must decide the issue without any recourse to either scientific or nonscientific arguments. The same holds for genetic engineering since what is at stake in both cases is a definition of a human life that cannot be simply left undecided, especially, given the fact that citizens generally hold deeply-held convictions regarding it. In discussing such borderline cases, Habermas (ibid: 40) readily admits that state neutrality cannot always be strictly enforced, but instead of giving up on the whole principle, he tries to mend it to fit it into his original theoretical mold. Below, I will describe Habermas's solution and argue that it turns deontological liberalism on its head, instead of providing a much-needed scaffold to it.
Habermas's strategy for saving the universality of his neoKantian deontological conception of morality is to concede just as much ground as he thinks is needed to save his approach. So, what he proposes falls short of a communitarian endorsement of metaphysical, or ethical worldviews; instead, it remains faithful to his universalistic neo-Kantian approach. His ultimate aim is to show that comprehensive doctrines are only significant in the whole philosophical project inasmuch as they provide a much-needed scaffold for deontological secularism to operate.
In developing this watered-down version of deontological secularism, Habermas distinguishes between a "human dignity" and a "dignity of a human life" in the same way as he differentiates between supposedly universal and ethically neutral Kantian morality, on the one hand, and comprehensive doctrines, on the other. The "dignity of a human life," unlike the "human dignity," is an ethical, rather than a moral conception. For example, a fetus, according to Habermas, does not have human dignity, but rather, we may talk about the dignity of a (potentially) human life. To illustrate his point, Habermas recalls the decision by the Parliament of Bremen to regulate the procedures of funerals of stillborn and prematurely born children. He notes that despite the fact that the Parliament allowed burying fetuses anonymously in collective graves in a cemetery, the state was still keen to separate such funeral procedures from traditional ones (Habermas 2003: 36). In this way, the ethical value of human life was demarcated from the moral value of a human being as an end in itself. Ethical values, according to this wellknown classification, can be ranked, that is to say, they are commensurable, while moral norms cannot be measured; they are absolute by nature. The dignity of human life can be juxtaposed against other values,for example, against a woman's right to her body, while human dignity cannot be traded in any way.
Despite the fact the abortion debate cannot be solved from the perspective of human dignity, it is argued that an ethical conception that the state will adopt must still somehow stay faithful to the priority of morality over ethical doctrines. For this reason, it should be grounded in what Habermas calls an "ethical self-understanding of the species", i.e., certain essential common traits that human beings share across different cultures and societies (Habermas 2003: 38-41). These basic traits are directly linked to universal morality as long as they imply that human beings perceive themselves as "ethically free" and "morally equal beings guided by norms and reasons" (ibid: 40-41). The "ethical self-understanding of species" is not a moral norm, but rather its value is derived from a particular ethical doctrine. Habermas himself openly admits that this is so (ibid: 71). While a fetus does not have human dignity, it has a "dignity of a human life," which is a corollary of the former, albeit in ethical, non-moral terms.
Habermas attempts to address the question of genetic engineering from this vantage point as well. Here, he starts by drawing a distinction between negative and positive eugenics. Negative eugenics only prevents the occurrence of diseases and disabilities by not altering the ethical self-understanding of species, while the goal of positive eugenics is to cultivate "desirable" traits that undermine the ethical self-understanding of human beings as free moral persons capable of making autonomous decisions. Habermas claims that on the desirability of negative eugenics future subjects will agree counterfactually, while in the case of positive eugenics, future subjects will be unable to treat themselves as morally autonomous and hence, will be deprived of the capacity to give consent even theoretically (Habermas 2003: 50-74). So, as the argument goes, the question of eugenics, ultimately, cannot be solved without a resource to the ethical self-understanding of species. As we can see here, Habermas once more links particularistic ethical doctrines and morality to each other in such a way that the former provides a scaffold for the latter.
The fact that a particularistic ethical doctrine has to save morality, especially in such borderline cases like abortion or genetic engineering, does not prompt Habermas to reconsider his views on the priority of universal morality over particularistic ethical doctrines. In Faith and Knowledge Habermas (2003: 105-115) reiterates his support for the state neutrality towards various comprehensive doctrines and, at the same time, affirms his preference for particularistic ethical doctrines over morality in borderline cases like genetic engineering, without compromising the whole model of deontological secularism. For Habermas (2010: 13), despite all these inconsistencies plaguing his model of deontological secularism, "a direct link must not be permitted between the universal morality and the respective particular worldviews." However, I argue that once Habermas admits that an ethical doctrine needs to be introduced in moral theorizing in order to account for state neutrality, deontological secularism loses its universalistic thrust. Now, it is not clear how state neutrality is supposed to work as a quasi-transcendental, universal principle since it cannot sustain itself without the help of particularist doctrines. The argument becomes self-contradictory: the priority of morality over particularistic ethical doctrines is based on the universal human reason that is shared by the whole humanity and is not derived from particularistic worldviews; however, it cannot sustain itself without making recourse to the ethical self-understanding of species that is also based on the very idea of freedom and morality, but this time derived from the particularistic ethical point of view (see Habermas 2003: 38-39). In other words, I contend that if the ethical self-understanding of species is a particularistic ethical view, then the whole principle of the priority of morality over particularistic ethical doctrines must also belong to the same category. The seemingly cosmetic changes that Habermas makes to his deontological secularism jeopardize the ontological status of his whole project, by calling into question the universality of the neo-Kantian deontological morality.

B) Universal Morality and Positive Freedoms
The same conclusion can be reached when analyzing Habermas's take on positive freedoms,i.e. the freedom of realizing one's political or other sorts of visions. Habermas (2006: 10-11;2008: 130) famously argues that religious citizens are not supposed to employ religious arguments in the formal political public sphere without taking necessary steps to translate those arguments into a secular language. This is known as an institu-tional translation proviso, and it occupies a prominent place in Habermas's discussions on secularism. However, unlike Rawls, Habermas seeks to ease the burden of translation imposed on religious citizens, recognizing that it puts them on an unequal footing vis-àvis their secular compatriots. Habermas (2006: 10-11) promptly demands from secular citizens to assist religious citizens in the process of translation and to acknowledge the "possible truth content" buried in religious traditions.
By introducing the institutional translation proviso, Habermas seeks to accomplish two tasks. First, he aims to maintain a neutral, secular public sphere where religious doctrines and other particularistic ethical worldviews -including, certain forms of atheism -do not trump secular, universal reasons. This is done to prevent any doctrine from imposing their dogmas on unconvinced citizens in the political public sphere constituted by public institutions (Habermas 2006: 12). Translation undertaken by religious and secular citizens is supposed to soften the original metaphysical or ethical implications of religious arguments. As a result, secular citizens would be able to deliberate in the public sphere without being forced to acquiesce with religious or other sorts of ethical arguments. Religious citizens, on the other hand, would be able to express their religious attitudes even after the translation has taken place. By casting aside their particularistic views, "different communities can develop a more inclusive perspective by transcending their own universe of discourse" (Habermas and Taylor 2011). Secondly, Habermas recognizes that making translation mandatory for religious citizens may mean an imposition of unequal mental and psychological burdens on them. To preempt such an outcome, Habermas (2006: 11) turns translation into a cooperative process in which non-religious citizens also participate. In other words, religious citizens are allowed to express their reason in a religious language only if they respect the "precedence of secular reasons and the institutional translation requirement" (Habermas 2006: 15). In this way, Habermas seeks to ensure that the state is truly neutral in a sense of imposing comparable psychological and mental burdens on its religious and non-religious subjects. As a result, the deontological liberal approach to secularism is justified by adhering to a strict standard of state neutrality.
However, such a translation envisaged by Habermas will only work if we accept that it can neutrally convey religious attitudes, or feelings, without assigning entirely different meanings to them. If a translation is not neutral in relation to religious attitudes, this means that it cannot guarantee universal moral legitimacy of secular reasons. As Badredine Arfi (2015) aptly points out, this is exactly what happens. He notices that Habermas's is not a translation in a conventional sense; it does not translate words from one language into another, matching their meanings more or less accurately. Rather, this is what, following Derrida, can be called an "anasemic translation" which designifies truth contents and resignifies it later, losing its initial kernel in the course of such a transformation. 4 Religious contributions to the secular public discourse forfeit their unique identity in this process. According to Arfi (ibid: 498), "should Habermas' anasemic translation of religious discourse succeed, the outcome would be a transmutation of the truth contents of the religious discourse into something else which is alien to religious discourse and meaning." Thus, a secular state that is based on deontological principles cannot be considered to be a neutral domain as it forces religious persons to give up on their original beliefs. Now, as Maeve Cooke (2014) notices, this does not necessarily mean that religiously or culturally imbued practices or statements do not possess a context-transcending power. According to Cooke (ibid: 1112), we may, for example, experience the religious figures as exemplary, not necessarily out of religious reasons, but for the intrinsic value of their behaviour or rhetoric (11)(12). In this sense, Habermas is right in claiming that translation is not impossible. Instead, the trouble begins at the moment when one assumes that translation is "transparent"/neutral, and there is no power play involved in the process -no party exerts an uneven influence on it. As can be seen, in fact, translation is not a neutral effort that can possess universal legitimacy.
To sum up, neither negative freedoms nor positive freedoms can be defined solely in accordance with the principle of the priority of morality over particularistic ethical doctrines. For this reason, the latter principle does not sustain itself as a universalistic standard. Below, I will show that in contrast to Habermas's, Derrida's deconstruction is a weak form of universalism that does not attempt to impose rigid secularist principles on all cultural contexts. In view of this, it can be better suited to construct a universalistic vision of secularism.

Secularism beyond Faith and Knowledge
Deconstruction does not offer a rigid principle that would be smoothly applicable across various contexts. Instead, Derrida (1992: 24-25) admits that every particular context necessarily "contaminates" the universality of any given doctrine. This claim concerns not only the particularistic ethical worldviews that disrupt the universality of deontological precepts but also any type of preconceived rules or principles. According to Derrida, ethical particularism already establishes itself at the very moment something is calculated preliminarily as a program. For example,  argues that in order to define justice, we need to separate it from a simple rulefollowing. However, the difficulty is that the latter includes any type of law or a principle that was set before a particular decision took place. Moreover, it is not only that many acts of justice are necessarily tainted because of their grounding in some preliminary principle or doctrine, but even if someone delivers a "fresh judgement," it immediately ceases to be a just decision. The reason for this is that after it has been delivered, the decision "has again followed a rule or given itself a rule… it is no longer presently just, fully just" (ibid: 24).
The force of particularism here resides in the fact that any seemingly absolute standard of justice quickly becomes a particular rule. The latter is circumscribed by various factors, be it cultural, social, political, psychological, or some other. Universal morality that is completely detached from the determinate, particular context or is not predicated upon a preceding cause turns out to be an illusory promise.
The same logic applies to the issues of secularism and religion. According to Derrida, here as well, one is partially confined within the dogmatic core of the religious doctrine. This claim is related to what is called "political theology", but it also has much broader implications. This is because Schmitt's thesis about political theology mainly concerns the way how secular concepts are modified theological ones, whereas Derrida analyses the institutional and epistemological nature of religious doctrines, in general, and extends the whole argument to include secularist worldviews as well. The Derridean contention is that every sort of religion is premised upon the inescapable dogmatic core that cannot be easily eliminated in the name of a rational, nonparticularistic/universalistic faith. Kant's original separation between knowledge -i.e. dogmatic and institutional elements of religion -and a rational faith, is inverted in such a way that faith ceases altogether to be an independent source of religion. Instead, it is replaced by a different type of non-Kantian faith, which defies the original Kantian distinction. Below, I will clarify how the original Kantian thesis is deconstructed by Derrida and how it also applies to the Habermasian argument about secularism.
The distinction between the two sources of religion first arises in Kant's famous work Religion within the Boundaries of Mere Reason. Kant (1998: 11) argues that there is only one true moral religion, and there are many kinds of ecclesiastic faith. Moral religion is the one that follows Kant's understanding of morality as a product of practical reason which is unconditional and universal,i.e., without any grounding in the empirical world which can be experienced with our senses. Ecclesiastic faith is, on the other hand, only temporal, so to say, provisional means to advance the actual cause of moral religion. It exists only because human beings have the "natural need" to draw on their sensual experiences. ) broaches a possibility that this temporary buttress might vanish in the future when the true moral religion would reign over the whole world. In the meantime, the goal is to guide the followers of various kinds of ecclesiastic faith in the direction of reason and shield them from the influence of those servants of the church who only perform "counterfeit service", that is, only seek the divine in the domain of the particular, rather than universal truth (ibid: 162-166).
Habermas's secularism follows Kant in assuming that a universal morality unalloyed by the particular reality has an unconditional nature, while various religious doctrines only exist as non-universalizable, particularistic truths. The two sources of religion outlined by Kant follow this priority of universal morality over particularistic ethical doctrines. What Derrida calls into question is not the idea that religion has two contradictory sources but rather what we mean by "moral religion" and "ecclesiastic faith". While Kant and Habermas offer a "formula" to separate universal morality from particularistic truths, Derrida is reticent to delineate the two primary categories in strict terms. Instead, Derrida argues that everything that is taken as universally valid is suffused by a determinate, particular context. The same mechanism is at work in monotheistic religions as well (Derrida 2002: 45). To put it differently, pure universality is hard to sustain in autarkic terms.
In Faith and Knowledge, which is a reading of Kant's abovementioned treatise on religion, Derrida (2002, 70-93) makes his own distinction between the two sources of religion. The first source is what he calls "knowledge", which denotes any type of attachment to a determinate reality. Derrida (ibid: 78) believes that religion cannot help but attach itself to a particular place, institution, or ethos. It even uses "tele-technoscience" to advance its agenda on a global stage, as we see in the case of tech-savvy religious leaders, or the horrifying global reach of religiously-inspired terrorism (ibid: 82). The moment religion institutionalizes itself, it loses its appeal to the sacred which has to remain something inaccessible or as he calls it, 'unscathed" (ibid.).
In this way, "knowledge", as Derrida puts it, is different from what Kant and Habermas understand as a particularistic (as opposed to universalistic) truth. This difference is not sufficiently underscored by Habermas himself. For example, Habermas (2003:113) mentions Derrida as a "worthy winner of the Adorno prize" and lauds his way of conceiving "messianicity without messianism". He claims that reason without the aid of metaphysics will struggle in its mission to enlighten society as it lacks the necessary semantic resources to do so. According to Habermas (ibid: 114), by appealing to religious sources, reason makes up for its "scarce resource of meaning". Habermas here reiterates his past claim that ethical or metaphysical doctrines have no truth value, but they may have certain semantic efficacy. Derrida, from a Habermasian perspective, is seen as a philosopher who tries to combine the two -reason and metaphysics in order to fulfil precisely this pragmatic purpose. But as I argue in this article there is much more to Derrida than Habermas acknowledges. In fact, the deconstructive reading of secularism offers the way out of the impasse in which Habermas finds himself when confronting the aporias of deontological secularism.
From a Derridean perspective, deontological morality itself is a form of knowledge, as it presupposes following a specific rule that bars it from being an absolute standard. Recall that Derrida (1992) argues that every rule, no matter how abstract and detached it is from a particular reality, is necessarily "contaminated" and circumscribed by a determinate context. So, while Habermas's deontological secularism implies that only those positions that are based on the priority of morality over particularistic ethical doctrines are universal, from a Derridean perspective, morality falls short of providing a neutral standard, and it cannot completely sidestep particularistic ethical doctrines.
Does the foregoing mean that the possibility of context-transcendence is revoked in the Derridean philosophy? The failure of the deontological approach may force us in this direction. It is true that the priority of morality as a high-order form of reasoning cannot be universally applied across different particular contexts, but does this mean that there could be no "higher-order" moral force? Derrida believes otherwise. While he rightly points out the universalism of the Kantian moral faith is untenable, Derrida (1997: 21-22) still argues that without pretension to universal justice no political creed, including secularism, or religion can function. The reason for this is that faith is something that sustains every type of knowledge -no knowledge can exist without the possibility of doubting or believing in it. Derrida (ibid: 22) describes the role of faith in the following terms: You cannot address the other, speak to the other, without an act of faith, without testimony. What are you doing when you attest to something? You address the other and ask, "Believe me". Even if you are lying, even in a perjury, you are addressing the other and asking the other to trust you. This "trust me, I am speaking to you" is of the order of faith, a faith that cannot be reduced to a theoretical statement, to a determinative judgment; it is the opening of the address to the other. So this faith is not religious, strictly speaking; at least it cannot be totally determined by a given religion.
In order to get a better understanding of the Derridean argument, it is essential not to conflate two different meanings of faith: the first one is Kantian (and later, Habermasian) faith which is a regulative Idea. It is allegedly wholly detached from any particular context; the second, Derridean faith is something that is intrinsically tied to knowledge but in such a way that it still necessarily overflows it. As Michael Naas (2012: 67) notices in his scrupulous discussion of Derrida's Faith and Knowledge, the Derridean faith is directly related to another Derridean concept of the totally other (tout autre). The idea is that every determinate reality is ruptured from the inside in such a way that it is never selfsufficient; something foreign -a contexttranscending other always interferes in every contextually circumscribed particular act. It compels every particularity to "reinterpret" itself. Thus, no particular determinateness can sustain itself without appealing to such a heterogeneous force.
The Derridean faith is not the type of universal faith that can be determined a priori, i.e., detached from every particularistic ethical doctrine, as it is posited by deontological liberals. It is, on the contrary, indeterminable and arises out of every context anew in a singular manner. It deconstructs every self-enclosed system or a set of beliefs because each time something heterogeneous -a leap of faith is required to ground a seemingly selfsufficient, normatively entrenched worldview. For example, in every religion, there is a mechanism of undecidability at work that subjects it to various, sometimes conflicting interpretations. This mechanism makes sure that religious interpretations change depending on time and space. As Derrida (2002: 93) remarks in Faith and Knowledge, "one can always criticize, reject or combat this or that form of sacredness or of belief, even of religious authority, in the name of the most original possibility." From a Derridean perspective, such a faith only comes into being when the particularistic doctrines exhaust themselves, that is, are exposed to the impossibility of absolute determinateness. An uncertainty -or what Derrida sometimes calls "perhaps" (peut-être), or undecidability -is a universal precondition of any determinate context, since every positing implies the possibility of non-positing (Derrida 1994: 94;2005: 88).
The fact that one can avoid neither a necessity of context-transcending singular faith nor a necessary rootedness in a particular context leads us to the aporetic conception of secularism. (2002: 43). Such an aporetic conception cannot produce a definite conception of secularism applicable to every possible context. In contrast to the deontological view, there is no one clear-cut rule that can be seamlessly applied everywhere. So, this type of secularism is not a homogenous phenomenon but it depends on the particularities of space and time. A change in it only arises due to the fact that context-transcending singular faith always lurks behind the particularities of a given context.

Conclusion
In this article, I started from the problem of locating the source of universalistic thrust behind secularism. I argued that the Habermasian deontological liberal approach to secularism fails to meet its own standards, and cannot establish the universality of the priority of morality over particularistic ethical doctrines. However, this fact alone does not warrant a rejection of the universalistic claims of secularism as there is another path to take. The Derridean critique of faith can be mobilized as a revamped version of liberal universalism. On this view, secularism is necessarily grounded in a particular socio-cultural and political context, but despite its boundedness, a context-transcending "faith" always lurks behind any secularist regime. In this way, Derrida's deconstruction offers a weak form of secular universalism that can evade the mistakes that plague deontological liberalism.