IntroductionConstitutions present themselves as new beginnings, as documents in which a political order is invented or reinvented. Of course, no constitutional document makes a political world out of nothing, nor can one be written that ignores all existing power relations. But the images that surround constitution writingthe nation assembled (in a convention or constituent assembly) and, more recently voting (in a referendum)-and the language contained in the general and the detailed clauses (suggesting that it is the constitution and the sovereign people that grant authority and do not rely on existing institutions) both produce the image of the constitution as not merely a political effect but also a cause.At a minimum, then, constitutions are portrayed as steering politics in new directions even if they do not make completely new orders. Writing a constitution has been likened to revolution (Ackerman, 1992); the very word suggests that something new is being "constituted." Even a slightly more modest metaphor--"rebuilding the ship at sea"--while it allows that the process is carried out with whatever is already on hand, still sees the work as absolutely foundational; the post-1989 order was one in which "The question of the moment was not 'What is to be done?' but 'Is there anyone who might be able to do anything-including defining what needs to be done? '" (Elster et al, 1998, 25) The idea that constitutions have deep effects is not merely a conceit of their drafters; scholarship on the meaning and impact of different constitutional arrangements is voluminous indeed. But despite high expectations surrounding them "constitutional moments" and the fact that the constitutions to which they give birth often succeed in provoking deep debate they often fail to inaugurate any significant change in important aspects of state-society relations.2 Remarkably, for all their self-proclaimed status as a new start, the best indication of what is in a given constitution is generally what was in the one it is replacing (Elkins et al, 2009). This is even the case in the critical area of relations between state authorities, religious bodies, and religious traditions where attitudes are often polarized. Our paper illustrates this claim as it relates to the field of education by examining three cases -Norway, the Irish Republic, and Egypt -where, despite the fact that the relevant constitutions attempted to mark the achievement of legislative independence, the entrenched interests of religious bodies remained largely unaffected and have remained so for long periods of time. The stakes surrounding the writing of a constitution can be tremendously high for religious institutions; this is, after all, when their relationship to the political order is defined. But we find that the documents tend to reflect existing arrangements and once written down, they become difficult to change because of the kind of deep consensus required for any subsequent change. This is especially so when it comes to clauses protecting religious educatio...