First, as the entries by Noah Salomon, Mathijs Pelkmans, and Robert Hefner, among others, show, it is useful to step back from the US, and even from Western Europe, to consider alternative ways of organizing diversity. In northwest Madagascar, where I have conducted ethnographic fieldwork sporadically over a couple of decades, there has been religious freedom in the sense that the boundaries between practicing Christians and Muslims are fairly open and, even more, insofar as it has been perfectly acceptable to be neither Christian nor Muslim, without thereby being designated as immoral or ‘primitive’ or subjected to undue missionary activity. As I’ve written elsewhere, some families might gently direct one of their children toward Islam, another toward Christianity, and a third to ‘ancestral practices,’ which are simply referred to as “non-congregating” (tsy mivavaka) rather than by any substantive definition. Some people engage in combinations of each. Although I would not advocate a causal explanation, the pattern fits nicely with the logic of bilateral kinship and wide exogamy. Most people can recognize at least four grandparents and probably eight great-grandparents (and beyond), each of whom may have a distinctive identity with respect to social, political, religious, and geographical affiliation. From among these senior living or deceased relatives people make choices of stronger or weaker identification, influenced by such factors as which grandparent one is sent to stay with on vacations as a child and ending with in whose tomb and which mode of burial one finds oneself.

This enables an open society with a good deal of mutual understanding and respect, in which no single identification or institution behind it is absolutized. In some respects one could say the individual has a good deal of freedom of choice. However many Malagasy do not experience things in quite this way. In explaining why they live in one place rather than another or carry out a particular set of ‘religious’ or ‘ancestral’ practices they would say they had been called to it by a particular ancestor, who by showing them signs, notably manifest as illness or troubling dreams, subjects them to prohibitions which align them more firmly with that ancestor rather than others. Servants at the ancestral shrines were forced some generations ago to work there. Today those who remain as their successors cite the wrath of their own ancestors as the reasons for staying on.

In all this there is also a logic of the negative. People are defined and define themselves in the first instance by what they don’t practice, by the kinds of praying they don’t do, the foods they cannot eat, the days they cannot work, or the kinds of work or acts of deference they cannot perform, rather than by positive attributions. This is a kind of freedom by restriction; in clarifying the boundaries of what you cannot do, it leaves wide open what you can do.

My second general comment is that however we want to define religion (and perhaps we could take a leaf from northern Madagascar and leave it open, specifying only what it is not), one of the general features, as the Malagasy ethnography also suggests, is a kind of submission to something conceived as larger, higher, or more powerful than oneself. Durkheim called it society; Maurice Bloch calls it deference to authority or to other persons; Roy Rappaport describes it as one of the entailments of engaging in ritual performance. In participating in a ritual, whatever one’s state of mind or ‘belief’ at the time, and irrespective of the semiotic ideology that Webb Keane rightly and compellingly points to, one is accepting the outcome (assuming that the felicity conditions of the performative event are met) and moreover accepting the meta-performativity, i.e. that acts and utterances of this kind, felicitously produced, have the consequences that they do. To perform a ritual is, in the end, to accept a certain liturgical order of which it is part (irrespective of whether this also entails deference to specific officials, like priests). In other words, the freedom to carry out certain kinds of acts is premised on subjection to an order that defines what such acts are, that puts things under a definition and regulates the changes in definition. As I elaborate elsewhere, the process is one of the instauration of ethical criteria and it is intrinsic to human speech acts. Insofar as what we refer to as specifically ‘religious’ includes the most formal and consequential kinds of performative acts (baptized or not, etc.) one might say that what religion is not is freedom.

Hence the very idea of freedom of religion is paradoxical; it is the freedom to be unfree in a particular kind of way. Judicial and legislative bodies need to take this point, call it the relativity of freedom or unfreedom, or the deconstruction of freedom, into account. They need to notice Sullivan when she points to “the reinstatement of the rights of religious authority by political authority—in the name of religious freedom.” They then need to make informed decisions about which versions of unfreedom to support—and we should all, as Saba Mahmood emphasizes, pay attention to the politics and ideologies that underpin such decisions (a skepticism I share with Lori Beaman, concerning federal government initiatives at the present time in, of all places, Canada). If Muslims were the ones taking the lead in the US courts asking for certain rights and freedoms, surely the self-same justices would have argued another way.

This is certainly not to say let everyone be free to do as they please. Not only is such freedom impossible in the human condition, but there is the matter of whether my freedom impinges on yours. To emphasize a point in Mahmood’s account and mentioned in some of the other posts, the freedom of religion we demand elsewhere (though the point applies internally as well) too often means the freedom to missionize other people. The freedom to practice my religion impinges on the freedom to practice yours in peace.

We need to be careful here. I am not a historian but I imagine that religious freedom once meant freedom from oppression by the proponents of a stronger religion rather than freedom from interference by the state or the right given by the state for specific religions to interfere in other peoples’ business. Certain proponents of religious freedom in the US now seem to want to have it both ways: the state is criticized both for being secular and for promoting a ‘religion’ of its own. What is missing in such arguments is attention not to one’s own rights or freedoms but the obligation to enable the rights and freedoms of others.