International Conference on Cultural Identity and Religion

Part II – Religion, Public Rationality and Moral Pluralism
Antwerp, 26-27 September 2002

Connie Aarsbergen - PhD candidate Philosophy of Religion
Free University of Amsterdam (Vrije Universiteit)

To realize the relative validity of one’s convictions
and yet stand for them unflinchingly,
is what distinguishes a civilized man
 from a barbarian.

(Isaiah Berlin,1969:172)

The role of religion in political debate has become the subject of renewed debate. On the one hand, religious people do not feel fully recognised as in political debate they are not (fully) allowed to speak about the world as they see it. In their view, the political justification of legislation is not neutral but biased in favour of liberalism. On the other hand, there is fear that when religious arguments are (fully) allowed, personal freedom is limited on only religious grounds. This fear is growing as Western countries are now confronted with immigrant orthodox or fundamentalist Muslims who may want to impose the Islamic shari’a on all citizens. In this paper I examine some recent proposals to deal with pluralist societies including the moral and epistemological attitudes connected with them. Among them are the attitudes of epistemological restraint, fallibilism and the willingness to use public and secular reasons. Can we do without these requirements? 

How can citizens live peacefully together when they have different religious and philosophical world-views? This question was one of the topic in John Rawls’ A Theory of Justice (1971). To answer this question, Rawls made use of some hypothetical devices, such as the notions of social contract, original position and the veil of ignorance. Suppose a social contract has to be made from scratch, how should it look like? In such an original position, adherents of different comprehensive doctrines are put behind a veil of ignorance, not knowing their personal background. As rational beings they have to make decisions about the basic rules. Not knowing if they are religious or in a minority or majority position, they will, according to Rawls, choose to make a distinction between private and public domain. In the 1970s and 1980s, especially the hypothetical nature of this theory was highly debated. Furthermore, Rawls realised that his theory is not stable as it is (like a modus vivendi) based on prudence and not on morality. Instead, in Political Liberalism (1993) Rawls makes use of the concept of an overlapping consensus. It is a consensus about the basic structure of a modern constitutional democracy and it includes certain moral attitudes and civic virtues (such as equal respect) to uphold this consensus. The justification of establishing laws is now on a moral grounds. This overlapping consensus has its roots in various religious, philosophical and moral doctrines, but once it is established it functions autonomously and can even be used as a criterion for rejecting comprehensive world-views that do not agree with these principles. This means that conceptions of the good outside the overlapping consensus can be excluded. For Rawls the overlapping consensus is not fully neutral, but only approximately neutral. It is only ‘neutral’ in the sense that there is no prior commitment to any wider doctrine. (No specific religion is privileged)

When in the political domain constitutional essentials are discussed, for Rawls only public reasons should be used (1999:485). Non public reasons, that consist of private, philosophical and religious reasons, should be excluded. However, public reasons should be used that contain:

virtues of reasonableness and fair-mindedness as shown in the adherence of the criteria and procedures of common-sense knowledge, and the methods and conclusion of science when not controversial, as well as respect for the precepts governing reasonable political discussion. (Rawls,1999:484)  

Most Christian political thinkers have responded negatively to this restriction, claiming that citizens who only have religious arguments, are now totally doomed to silence. In these Christian critiques, however, it is often overlooked that Rawls’ restriction regarding the use of religious (non-public) reasons only applies to constitutional essentials and not to discussions in civil society or in public or political debate about other topics. (Dombrowski:47,95). 

This discussion has taken place on academic level. In practice, however, religious people feel that in the entire public domain, they are restricted in using their arguments. According to the Christian political philosopher Nicholas Wolterstorff (1997), many religious people have difficulties with the public-private distinction. ‘Their religion is not, for them, about something other than their social and political existence; it is also about their social and political existence’ (Wolterstorff:105). They consequently feel limited in expressing their identity. He criticizes Rawls restriction to use public reason in the political domain, as it excludes religious people who have no other language. The political justification of legislation is not neutral but liberal. He wonders how the overlapping consensus can be improved when no religious arguments can be brought forward (Wolterstorff:101). Furthermore he denies the existence of a shared political culture in a plural society that underlies the overlapping consensus and is of the opinion that Rawls has an ‘extraordinarily idealized picture of the American political mind’ (Wolterstorff:97). There is for instance a distance between what many Americans think about homosexuals, and the civil rights homosexuals have. 

At this point I want to remark that by dismissing the public-private distinction, the use of public reason and denying the existence of an overlapping consensus, Wolterstorff, is faced with two major problems:

-                     How can he protect citizens of a pluralist society of being restricted in their liberty on religious grounds that they do not share?

-                     How can he avoid the negative consequence of moral relativism that is connected with particularism. (How can Wolterstorff condemn groups outside his own?)

These systematic questions will be a running thread in my account of the serious liberal response on the above Christian objections.

From an unexpected corner, Wolterstorff gets some support. Already in 1987, the liberal American philosopher Thomas Nagel acknowledged that in the political domain, religious people are forced to treat their beliefs as if they are not true and that could lead to doubts. For liberalism itself this is problematic as its ‘foundation’ would then be scepticism. For Nagel this need not be the case. For religious people it is not necessary to become sceptical since in a political argument it is not necessary to appeal to ‘the truth’ of one’s belief:

I believe that the demand for agreement, and its priority in these cases over a direct appeal to the truth, must be grounded in something more basic. Though it has to do with epistemology, it is not scepticism but a kind of epistemological restraint: the distinction between what is needed to justify belief and what is needed to justify the employment of political power depends on a higher stand of objectivity, which is ethically based. (Nagel:227) (Italics are mine) 

Nagel introduced a new type of epistemological attitude citizens can have in a pluralist society, namely epistemological restraint. It is a kind of reserve with regard to the own truth claims without sceptical consequences. The justification for the liberal conviction not to impose one’s own vision of the good life to others, is not based on knowing the truth or not being able to know the truth (scepticism), but on an ethical conviction. It is considered to be improper to impose one’s own vision of the good life on people who cannot share it, even if one is totally convinced that it is true. Thomas Nagel did not elaborate on this concept any further but  political philosopher Susan Mendus (University of York) has recently re-introduced it (Mendus:103-120). According to Mendus, liberalism should not be ‘grounded’ in scepticism, in an appeal to doubt. The moral conviction that it is improper to impose one’s own vision of the good, should be enough. When an attitude of epistemological restraint is observed, the connection between certainty (or uncertainty) in the mind of the agent and the legitimate use of power on the part of the state are separated. ‘So understood, liberal neutrality is not based on our own uncertainty about which conception of the good is best, but on the rejection of certainty as an appropriate deciding factor in the deployment of political power.’ (Mendus,118). Scepticism can be avoided by making a distinction between the existential level and justificatory level[1]. On the existential level, epistemological restraint does not require us to doubt or deny our own conception of the good. It only requires on the justificatory level we ‘abstract from’ our commitment to our conception of the good in deciding what can be legitimately imposed on others because it is morally improper.

Having ‘solved’ the problem of the ‘sceptic foundation’ of liberalism, a Christian thinker such as Wolterstorff could still argue that epistemological restraint is in fact based on an overlapping consensus about what is decent and morally improper in political debate. The existence of such an overlapping consensus is for him, however, doubtful. 

After a dialogue with Wolterstorff (1997), the American liberal epistemologist Robert Audi seems to be convinced of the importance of more religious input. In Religious Commitment and Secular Reason(2000) Audi argues that in political debate religious arguments should be used freely. However, he makes the restriction that when laws and public policy are discussed that limits the freedom of all citizens, the religious arguments used should be accompanied by at least one secular reason. For Audi, religious arguments are reasons based on scripture, non-scriptural religious authority (clergy), tradition, religious experience or natural theology (Audi,1997:10). Secular arguments are simply non-religious arguments; reasons that evidentially do not depend on the existence (or non-existence) of God or on theological considerations. In practice secular reasons refer to public health, public safety, educational need or national defence (Audi,2000:86). [2]  [3]  [4] In Audi’s proposal, religious people should try to seek for a theo-ethical equilibrium. The religious believer should ask herself the question if the legislation proposed or supported can be called ‘good’ independently from the scriptures or tradition. If the answer is positive, a secular reason can be found and a theo-ethical equilibrium[5] can be reached. For instance if a religious believer is inspired by the biblical notion of stewardship, this argument can be combined with secular concern about the alarming environmental state of the world. 

At first sight this may seem fair, but for religions there are still problems. The commitment to seek for a theo-ethical equilibrium, presupposes the belief in the autonomy of secular morality. Audi is aware that for religious people this can be difficult as in most religions morality is believed to have a divine origin. He therefore distinguishes between epistemic and ontological autonomy (Audi,2000:139). He makes no ontological claims about morality, but only argues on the epistemic level that we can understand moral concepts independently from theological ones. So it is not necessary for religious believers to deny that morality originates from God (Audi:1997,124). For most Christians an acknowledgement of the epistemic autonomy of morality should not be problematic since for instance Thomas Aquinas also took it as a starting point, namely by believing that God has provided humanity a route to moral truths along rational secular paths.

There is another difficult requirement for religious believers. Audi also wants religious believers to take a fallibilist attitude[6] in case they want to obey religious obligations but secular reasons cannot be found (Audi,1997:15). If for a specific religious obligation the only reason that can be found is reference to authority, then the religious believer should recognize the possibility that he or she is making an error.

For orthodox and fundamentalist religious groups, the above requirements can be very difficult to meet. The conclusion can be drawn that in Audi’s alternative religious groups are still denied access to the political domain. However, when Christians want no restrictions in the political domain, they are faced with the problem that they provide a ‘blank check’ to radical religious groups to promote their fundamentalist ideas. 

We still have Wolterstorff question to answer how the overlapping consensus can be improved when inspiration from religions is banned? Audi’s alternative allows room for (non fundamentalist) religious groups to use their arguments both public and political domain and also to be motivated by them. Because of the requirement of an additional secular argument, their religious motivation can also be understood by reasonable outsiders. This means that for society as a whole it is possible to enjoy the positive influence that religions can offer. For example, religions can shine their light on moral aspects that have been darkened in society. After illumination, it is possible that both secular and other religious people recognize these neglected or undiscovered aspects as good or bad and become motivated for (joint) moral action. 

Another unanswered question is the troublesome ontological status of the overlapping consensus. With his requirement to use secular reason, also Audi seems to presuppose the existence of an overlapping consensus. In Audi’s case, however, its nature is not fixed but can be changed by the input from various religions and world-views, but it still functions as protection against limitations of freedom on purely religious ground. This means that in this overlapping consensus, individual freedom is the highest value. The notion of overlapping consensus could therefore be regarded as a typical product of the West. In for instance a Middle East country, the majority of the population belongs to the Islam and their overlapping consensus will consequently have another nature and therefore less suitable as means to protect non-Muslims against intrusions in their individual liberty. Rejection of the notion of an overlapping consensus, however, has its price. It either leads to fundamentalism and monism (the belief that only the own vision is universally true and therefore rejection of pluralism) or to particularism (with its consequence of moral relativism and its impossibility to condemn outside the own group). I do not think that this is a price mainstream Christians, such as Wolterstorff, want to pay. There is, however, another option. In our world there is a growing international overlapping consensus as expressed in various UN organizations and the Declaration of Human Rights. Again, it could be argued that the UN and the human rights are typical Western products as individual freedom is their starting point. But do not all the struggles for freedom throughout history, also in non-Western countries, prove that there are some universal values, including the desire not to be forced to live under legislation that is only religiously motivated? 

My conclusion is that in a world that is threatened by both fundamentalism and moral relativism, the liberal political frameworks offered by Rawls and Audi are not perfect but still the best ways to deal with societies with different visions of the good life. They are neutral in the sense that no privileges are given to any of the existing world-views but they are only ‘approximately neutral’ in the sense that they are committed to individual freedom and peaceful co-existence in a pluralist society. The price for maintaining these values is the exclusion of radical religious groups with non-democratic intentions. For religions with good intentions the price for protecting these precious values is that an attitude of epistemological restraint and the use of public or secular reasons cannot be avoided. 

Connie Aarsbergen, 25th September 2002
 Print version


Audi, Robert, ‘The Place of Religious Argument in a Free and Democratic Society’, In: San Diego Law Review 30, 1993

Audi, Robert and Wolterstorff, Nicholas, Religion in the Public Square, academic debate between Robert Audi and Nicolas Wolterstorff, Rowman & Littlefield Publishers, 1997

Audi, Robert,  Religious Commitment and Secular Reason, Cambridge University Press, 2000

Berlin, Isaiah, Four Essays on Liberty, Oxford University Press, 1969

Dombrowski, Daniel A., Rawls and Religion, State University of New York Press, 2001

Mendus, Susan, ‘Pluralism and Scepticism in a Disenchanted World’, In: Pluralism. The Philosophy of Politics of Diversity, ed. M. Baghramian and A. Ingram, Routledge, London, 2000,

Nagel, Thomas, ‘Moral Conflict and Political Legitimacy’, in: Philosophy and Public Affairs, Princeton Univ.Press, vol. 16, 1987

Rawls, John, Political Liberalism, Columbia University Press, 1993

Rawls, John, Collected Papers, Harvard University Press, 1999

Wolterstorff, Nicholas, ‘The Role of Religion in Decision and Discussion of Political Issues’, in: Religion in the Public Square, Robert Audi and Nicolas Wolterstorff, Rowman & Littlefield Publishers, 1997                  

[1] The distinction between existential and justificatory level resembles the distinction between the context of discovery and the context of justification in the sciences. Some postmodern thinkers have argued that such a distinction is not possible. In Mendus’ case this distinction is made to protect religious believers from scepticism. 

[2] Audi avoids the term public reason as it  suggests intelligibility on the part of the general public, but a public can be ill educated or blinded by prejudice (Audi,2000:90).

[3] Audi does not claim that secular arguments are universally valid. By requiring a combination of both secular and religious arguments, he only wants to avoid limitations of freedom on purely religious grounds, without any motivation outside the authority of the religion

[4] Audi’s secular reasons should not be confused with anti-religious reasons. Atheist proposals, for instance to forbid schools to teach creationism, can be rejected on secular grounds as well in this case as it is at odds with the secular ‘do-onto-others-rule’ (Audi,2000:98).

[5] The theo-ethical equilibrium in fact consists of two requirements. The first requirement is the principle of secular rationale. This principle says that one has a prima facie obligation not to advocate or support any law or public policy that restricts human conduct, unless one has, and is willing to offer, adequate secular reason for this advocacy or support. (Audi,2000:86). This requirement is  followed by the principle of secular motivation, in which one has a (prima facie) obligation to abstain from advocacy or support of a law or public policy that restricts human conduct, unless in advocating or supporting it one is sufficiently motivated by (normatively) adequate secular reasons. (Audi,2000:96).

[6] The notion of fallibilism primarily refers to the mistakes in interpretation readers of scriptures or listeners or their leaders can make and it does not necessarily doubt the truth of the holy texts and the religion itself and therefore need not lead to scepticism.