It is hard not to be convinced by Akeel Bilgrami’s careful, patient, and generous exposition in “Secularism: Its Content and Context.” And indeed there is much with which I agree, especially the balance that Bilgrami strikes between a care for truth, on the one hand, and the idea of internal reasons, on the other. My remarks below are offered by way of exposition and clarification, but they are motivated by a spirit of interpretation: it seems to me that the paper operates in distinct tonal registers: a primary register of hope, a secondary register of tragedy, and an unacknowledged third register, which I will call prophetic.

First, the exposition. Most importantly, (S) is about religion; it is a “stance towards religion,” as Bilgrami puts it. He wants to narrow the concept in order to give it analytic purchase and clarity, and so he distinguishes it from “secularization” and “the secular.” Others have made similar distinctions, of course, but usually in order to identify a range of discourses and practices that are not obviously about religion but nevertheless central to its historical construction: Charles Taylor’s “secular age,” for example, or Talal Asad’s “anthropology of the secular.” Bilgrami goes in the other direction: he knowingly excludes from (S) a whole range of things that might be said to belong to the secular. Meditating, for example, on why some religious communities tend not to speak out against their more extreme fundamentalist elements, Bilgrami writes that:

In the case of Islam, this defensively uncritical psychology has been bred by years of colonial subjugation, by continuing quasi-colonial economic arrangements with American and European corporate exploitation of energy resources of countries with large Muslin populations, by immoral embargoes imposed on these countries that cause untold suffering to ordinary people, by recent invasions of some of these countries by Western powers, and finally by the racialist attitudes towards migrants from these countries in European nations.

Colonialism, neo-colonialism, economic neo-liberalism, and the presumed cultural superiority of the West do not seem unconnected to the history of secularism. Some might add mass incarceration and other forms of state-sponsored violence to the list, techniques of the body and new sensory repertoires, even capitalism itself.

(S) runs directly counter to this discursive expansion of secularism. We might think, Bilgrami writes, that the “rhetoric of ‘secularism’ … plays a role in the anti-Islamist drumbeat of propaganda that accompanies these other factors,” like neo-liberalism and the legacies of colonialism. But even if that is so, he argues, “the right thing to do is not to ask that secularism be redefined, but to demand that one should drop talk of secularism and focus instead on trying to improve matters on what is really at stake: the effects of a colonial past, a commercially exploitative present, unjust wars and embargoes, racial discrimination….” For Bilgrami, to discuss these things under the rubric of secularism is to make a category mistake.

I don’t intend to be making a point of which Bilgrami is unaware. Indeed, the whole goal of his paper is to produce a remarkably modest, minimal account of secularism (a goal not all of the commentators on the paper here on The Immanent Frame seem to have grasped). But it’s important to see what that minimalism entails. Here is Bilgrami’s description of (S):

In a religiously plural society, all religions should have the privilege of free exercise and evenhanded treatment except when a religion’s practices are inconsistent with the ideals that a polity seeks to achieve, in which case there is a lexical ordering in which the political ideals are placed first.

Importantly, (S) does not stipulate what the polity’s substantive ideals are or ought to be; it says only that those ideals have lexical priority over religious ideals when the two come into conflict. Modern democracies will likely tend toward familiar ideals—freedom of speech and association, for example—but other kinds of societies may well place other ideals first. The point is simply that those ideals, whatever they are, come first.

Bilgrami admits that this picture of secularism is “more adversarial” toward religion than Taylor’s multicultural ideal of neutrality. But it is adversarial in a limited sense: (S) only cares about religion “as it affects the polity. It is not dismayed by or concerned with the presence of religiosity in the society at large or in the personal beliefs of the individual citizens.” For (S), private religion is fine: even public religion is fine. Conflict only arises when religion tries to drive policies that run athwart the polity’s ideals. In that case, and only in that case, the lexical ordering kicks in.

Lest this sound imperious, Bilgrami emphasizes that the only reasons for holding (S) in the first place are “internal reasons” (a concept he adapts from Bernard Williams).

Internal reasons are “reasons we give to another that appeal to some of his own values in order to try and persuade him to change his mind.” We are all internally conflicted in some way. This doesn’t mean blatant contradiction, in the sense of believing both p and not-p; it just means that there are tensions among values that an interlocutor can help bring to the surface. In the same fashion, reasons for holding (S) must be “internal”—that is, those reasons cannot be separated from the values and commitments of the individuals or groups holding (S). They are not universal or context-independent. So internal reasons will persuade some people but not others. However, as with John Rawls’s notion of the “overlapping consensus,” Bilgrami suggests that there are plural reasons for holding (S). In a plural society, it is the consensus that overlaps, not the reasons. In the matter at hand, then, secularism should drop talk of universal rights in favor of seeking “local concepts and commitments within the [religious] community … that might put pressure on the community’s own practices.”

This is not only a matter of reasoning with someone in a cognitivist way. For even agents who hold tightly to an apparently unconflicted set of principles are subject to the changes of history: “internal conflicts may be injected by historical developments into moral psychological economies.” Indeed, most successful activist movements work in exactly this way, by bringing to light or making visible a historical change already underway, thereby forcing majorities to confront the historically-bound nature of their own commitments, which they might otherwise have continued to think of as timeless.

Yet the historical record certainly offers plenty of examples of unchanged minds, or of minds that change and then change back, or of minds that change for the worse rather than the better, becoming more entrenched, more dogmatic, and so on. These possibilities don’t register very strongly in Bilgrami’s paper. This is what I meant when I said above that one tonal register of this paper is hope. For Bilgrami has a humanist confidence that the movement of history will eventually force illiberalism to confront its own internal tensions. Here is where he takes an evaluative stand: he believes this not for metaphysical reasons (some grand Theory of History) but because to believe it is to care about the truth in a certain way. To want to argue with someone and convince them that their own deeply held principles are tension-filled and therefore ought to be modified is to care about truth as you see it in such a way that you want others to see things your way too. This is a sign of respect, and it also fosters an ethical project: generating “empathetic attitudes of engagement with the tradition and mentality of those one opposes.”

I like all this very much. It nicely sidesteps much of what is unpalatable or just plain shallow in some fashionable versions of relativism. It proposes a kind of dialogue that is respectful but also deeply committed to getting things right. And its picture of truth is dialectical and internal: we move toward truth through the hard work of examining internal points of tension within our own substantive commitments and moral/psychological principles. (This is why I think Justin Neuman rather misses the point when he writes in his post that Bilgrami is too ready to generalize about groups of people; as I understand it, one primary purpose of Bilgrami’s emphasis on moral psychology is to make it harder to engage in such generalizations.)

And yet it is clear that (S) depends upon a certain historically-specific definition of religion.  It builds on a picture of “religion” as ideally heading towards post-Westphalian Protestantism—a formation that, several historians have plausibly argued, helped to build the modern nation state as we know it. It seems likely that this hopeful trajectory is in some tension with a different theme that emerges in the middle of the paper. Here Bilgrami notes that secularism as a policy is the result and requirement of a post-Westphalian Western Europe, which strove to develop a “feeling for the nation” by identifying internal others (the Jews, the Irish) and thus “inventing” the problem of minorities. For Charles Taylor, those cases in which majorities and minorities are understood in terms of religion demand secularism in the form of neutrality. Bilgrami is skeptical that secularism as neutrality can actually handle the challenge of majoritarianism. When majorities and minorities are defined in terms of religion, he argues, “there inevitably arises a sense that religion itself is the problem, even though the historical source of the problem lies in majoritarianism.” At this point something stronger than neutrality is needed, namely the lexical ordering.

What I am calling the secondary tonality of tragedy enters here, because Bilgrami is clear that it didn’t have to turn out this way. Mohandas Gandhi, for example, tried hard to prevent the development of a situation in India (of nationalism and majoritarianism) that would in turn require secularism. (This argument isn’t really developed in the present paper, but Bilgrami has written of it elsewhere.) In this example, history is not a progressive force that gradually loosens the hold of illiberalism, but actually creates the conditions in which illiberalism can flourish, which in turn brings forth the need for (S) as our best hope in a situation that hasn’t turned out very well.

And, finally, the third tonality. Whether history is ultimately progressive or tragic, it is at least dynamic. Religion, by contrast, seems quite static. In Bilgrami’s schema, secularism points out to religion (or waits for history to do the pointing out) that its picture of things is full of internal tensions; it thereby hopes to convince religion to sign on to (S)’s lexical ordering for reasons that remain religious (that’s the overlapping consensus part of the argument) but have now been pluralized (that’s the moral psychological part of the argument). I don’t get any sense that for Bilgrami the arrow might sometimes point in the other direction. Like Lars Tønder and a few of the other commentators on the paper, I wonder whether this simply reintroduces the conceits of secularism (its confident sense that it is right, and that history is on its side) in more modest garb. Can we really envision, on Bilgrami’s grounds, an engagement that reveals that secularism, too, has internal tensions that should be confronted—a conversation, that is, that actually changes both participants? This is the possibility that William Connolly gestured toward in a brief reading of Søren Kierkegaard in Why I Am Not a Secularist, and that Tønder develops very nicely out of Maurice Merleau-Ponty’s notion of the “tolerance of the incomplete” in his piece: “rather than beginning with the issue of how to order political ideals,” Tønder writes, “Merleau-Ponty’s dialectic begins in the midst of lived experience, where perceptions, judgments, and ideals have not yet reached the threshold of conceptual clarity….” This would lead to epistemic modesty of a different kind—neither a liberal recommendation of tolerance because we might turn out to be wrong, nor the modesty of the overlapping consensus, but a modesty born of the sense that things are still in flux, still in process, and that all of us find ourselves suddenly in the middle of things and without fully secure footing. Elsewhere I’ve suggested that the romantic theory of the fragment offers a useful way to think about this kind of modesty.

I suspect that Bilgrami would say that I’m missing the point here. (S), he insists at the very beginning of his paper, isn’t a good in itself, and so it can’t be internally conflicted, nor can it be anthropomorphized. (S) merely seeks to promote other kinds of goods (to be established by the polity in question) and it is in order to protect those goods that the requirement of lexical ordering comes into play. Yet there is one good that (S) seems to be not simply protecting but also promoting, and that is the good of helping religions pluralize their own self-conceptions. This is where (S) becomes activist in a way that Bilgrami, I think, doesn’t fully admit. If I may introduce a new term here, we could say that (S) is prophetic in its relation to religion.

There are good reasons to think that religion is more dynamic than Bilgrami’s picture of it allows. Indeed, it seems to me that religion has often been prophetic in its relation to the state—even, perhaps, to the point of convincing it to change its lexical ordering. But that is a topic for another day.