I am very grateful to the many commentators on my essay “Secularism: It’s Content and Context” for their instructive and challenging responses and I am glad of this chance, in what follows, to try and make my essay clearer and better. It is a measure of the vibrancy of The Immanent Frame that it fetches such a high quality (not to mention, quantity) of commentary, and I hope I will be able to at least approximate some of this quality in my responses.

I’ll begin with some preliminary points which I will exploit in my responses, and then speak to each comment in turn, posting the responses one at a time over the next many days.


My essay began by distinguishing three cognate terms and their meanings as an effort to impose some clarity and distinctness to a somewhat confusing field of concepts. I did not deny that some of this sorting out was stipulative and proposed that the substantive discussion through the first four sections of the paper was intended partly to try and make this stipulative element non-arbitrary.

The three terms were “secular,” “secularization” and “secularism.” This family of terms, as is well known, grew out of a certain history and a certain intellectual history and my hope in these initial semantic explorations was to keep faith with that history as far as is possible.

I proposed that the term SECULAR simply be treated as a very generic marker of “mundiality” i.e., of all and any phenomena lying outside the concerns of “the cloister.” True, religions are sometimes supposed to be comprehensive doctrines in the lives of religious people, so it may seem that it would be contestable to those for whom it is such, that there is anything outside the concerns of the cloister. But I took it for granted that this extreme understanding of religion’s reach was implausible. There are several reasons for this. To begin with, the very word “cloister,” metaphor though it is, would lose its meaning if that were not so. More substantially, it is hard for anyone who understands elementary human social psychology and knows anything about the sociology of modern life to take seriously the idea that the players in social life are comprehensive doctrines. Subjects who enter the social arena are not doctrines but human beings and they have diversified psychologies—however important religious matters are in their lives and minds, they do not consume every moment and aspect of their lives. Finally, as Charles Taylor’s points out in his fine and much misunderstood book, if the reviews of it are any indication, A Secular Age, at some point in the history of Europe, however “comprehensive” religion may be, some of its doctrinal and its practical elements began to be seen as, “optional.” If nothing else, this surely opened things up for a domain properly describable as “the mundial” and it is elements in this domain that are properly describable with the term “secular.”

Of the trio of terms in this family, then, “secular” because of the generality of its coverage (anything outside the concerns of the cloister), is the most innocuous.

Following a familiar trend in common usage of the term by intellectual historians, I had said that the term SECULARIZATION described the process of decline in various forms of belief and practice that are loosely describable as “religious.” This process, as we all know, has had a very uneven development in different parts of the world. It is hardly disputable that secularization has been much more pervasive in Europe, where it first arose, and in Australia and in Canada, than it has, say, in many parts of the Middle East or South Asia or in the heartland of the United States, though this last may be a more distinctly modern form of religiosity, one that comes out from underneath of a process of secularization that has already taken place. This may also recently be true of certain modern, nationalist forms of religiosity to be found in other parts of the world, including South Asia and the Middle East, and even in some parts of Europe.

Taylor’s thematic focus in the book I have mentioned was on these two things that I have demarcated as the “secular” and “secularization,” and its gaze in particular was on Latin Christendom.  But Taylor, in other essays, for some years now, has been addressing what is a slightly different (though obviously not unrelated) and much more specific subject, which is the relation that religious doctrine and practice bear to a very specific domain, that of the polity and the state. The term “SECULARISM” is a good term to mark a particular position on this subject and Taylor himself uses the term in this way in these essays. I had said that the fact that the term summons for so many such slogans as “the separation of church and state” is something of a proof that the term has this restricted focus. It is not referring to a general mundiality that lies outside the cloister as the term “secular” does, it is not referring to a general process by which the domain of the mundial spreads and the domain of the cloister (i.e., the belief in and practice of religion) shrinks, as the term “secularization” does. It is referring to a position taken on the relation between religion and the specific domain of the polity (the “state” being a term that merely narrows somewhat the focus on the slightly wider domain of the polity).

As I say, in my essay, it is highly necessary to introduce this third term because the general phenomenon and process marked by the other two terms by themselves leave no space for the following possibility, which is in fact frequently realized. Many people are religiously devout in both belief and practice (which is to say that they are not yet fully given over to the process of secularization) but are nevertheless willing to leave aside some of their belief and practice that affects the polity in recognizable ways. I used policies and laws regarding free speech and gender-equality as examples of what is central to polities. I contrasted them with matters of dress and diet in the matter of religious practice, and in belief in God or in creationism in the matter of doctrine. These aspects of practice and doctrine have less directly to do with the polity than laws and policies regarding free speech or gender-equality.  Someone may protest: questions of dress and diet can be made part of the political domain by introducing policies and laws regarding them (as has been done in France for example by laws regarding what can and cannot be worn by females in schools). Well, there are all sorts of conversation-stopping ploys that could bring intellectual discussion to an end and it is certainly possible to do so by insisting that everything is or potentially is part of the polity and therefore secularism does not have such a restricted domain, as I am suggesting. But I would think that the very fact that it seems intuitively much more controversial to many that the hijab should be banned in schools than it does to disallow the banning of a book considered blasphemous by some religion, is some evidence that we intuitively do make a distinction between some things as being more central to defining the polity than others. Matters of dress and diet are only with some strain made to be matters that are directly related to the polity in a way that the matter of free speech is not. So there are good, substantial, and intellectually fruitful reasons to speak of secularism as distinct in meaning (because more restricted in scope) from “secular” and “secularization.”

My focus in the essay was entirely on “secularism” and I had characterized or defined it in my formula (S), which I won’t repeat here since it should be read and understood within the frameworking remarks that led up to its formulation in my essay.


Terminological distinctions apart, I had also made another crucial distinction between descriptive and normative aspects of the subject—that is, between, on the one hand, describing what secularism might sensibly be taken to be and, on the other, when and where it might be something that we should advocate or normatively advance.

The expression “when and where” in my last sentence conveys something that was central to my essay’s argument.

It was important for me to avoid confusion by insisting that what secularism means or is (the descriptive part of the subject) should be relatively fixed through different contexts but whether secularism, given this relatively stable meaning, is relevant or a good thing (the normative aspect of the subject) may vary from context to context. The essay tried to give both an historical and an analytical argument for why secularism only has normative relevance and should only be advocated in very specific contexts. In other words, different contexts should not make us want to redefine the term since that only gives rise to unnecessary and badly motivated theoretical confusion but, once defined and described in relatively clear and non-arbitrary ways, the ideal should be seen as having relevance and worthwhile application only in some historical and sociopolitical contexts, and not others.


Finally, I had made another distinction that I think is important in bringing clarity to the subject.  And that is to distinguish between matters of definition and matters of legitimation and implementation—that is, a distinction between what secularism means and how secularism should be justified and adopted by a polity. There has been much disagreement about how secularism should be justified: are there strictly rational (or what I had, following Bernard Williams, called “external) arguments for it that establish it as an objectively true doctrine or position in politics for all rational people? Or can it only be justified by appealing to more local (what I had, following Williams, called internal”) reasons that appeal to particular substantive values that may not be shared by all people? There has been a closely related disagreement as well about how secularism should be implemented: should the state be granted the right to impose it from on high or should it involve different religious groups in matters of crucial decision in its adoption, allowing them to come to secularism only if and when they found reasons within their own outlook and their own vernacular categories to adopt it? My own view, argued for at length in the opening and closing sections of the essay, was that on the matter of justification, we only had recourse to internal arguments, and in the matter of implementation, we ought to be inclusive of all religious groups in the process of the adoption of secularism. But none of these stances on how to justify and implement secularism, I had said, should affect the question of how we should define secularism.

The essay’s arguments and conclusions would be entirely unpersuasive unless these two distinctions (2 and 3) were kept well in mind.

The first was relevant in the following sense: Once we non-arbitrarily characterize secularism, we should realize that, so characterized, it was only relevant and necessary in a few contexts—contexts that I had said were first historically to be found in the history of European nation-building, and which were simply missing in many other parts of the world. There was no reason to advocate secularism in all parts of the world, therefore, as if the European context was the standard for all others. (Only if some other parts of the world had replicated relatively specific conditions that had loomed large in the history of European nation-construction does secularism, properly characterized, have relevance and application.) Thus it was not necessary to redefine secularism so that it had to fit all parts of the world or even to fit Europe in its changing contexts. It is theoretically sounder to stay with its stable meaning and simply deny its relevance to many of these contexts.

The second distinction’s relevance was this: Simply because one had different ideas about how to justify and implement secularism does not imply that we have different concepts or meanings of “secularism.” Once again, the meaning of the term should be considered to be relatively stable and fixed. And we could then either proceed to justify and implement it (in the contexts in which it was good to implement it) along lines that were internal and inclusive or external and state-imposed. Decisions as to how to justify it and implement it were relatively independent, therefore, from how to define it.