The title of this paper may seem too restrictive for an opening lecture in a colloquium concerned with the entire East of the Roman Empire, in which only half of the papers concern Egypt, and yet simultaneously far too ambitious in its scope. In the course of the colloquium, however, it became clear that the methodological issues that I was trying to confront were broadly relevant across the geographical span covered by the colloquium and to some extent raised fundamental questions about the very formulation of some of the organizers' questions to participants, questions which came back into focus in the lively concluding discussion. This in my view was to be anticipated, because part of my argument is that Egypt is not any more exceptional than anywhere else in the Roman Empire of late antiquity, and that the questions at stake there are broadly applicable, even if the answers vary.I should begin by recording two rather different debts that this paper owes to the work of other scholars. The greater one is to David Frankfurter's Religion in Roman Egypt.1 This book appeared four years prior to the colloquium in Münster, early in a semester when I was teaching a class on the social context of Egyptian Christianity to a seminar at Union Theological Seminary. It was a great source of stimulation for that seminar. As readers will find, there are many fundamental assumptions and arguments in that book with which I disagree; 2 but it was reading it that led me first to formulate some of the methodological points set out below and to force myself to clarify matters that I had left until then unanalyzed. This is a stimulus for which I remain grateful. The second obligation is to the brilliant conference paper given by [p. 24] Stephen Emmel at a colloquium in Leiden later that same fall entitled "Perspectives on Panopolis," the proceedings of which were published just before the Münster colloquium. 3 The analysis of Shenoute and Gessius -two central characters in a late antique real-life drama -given in that paper embodied and exemplified many of the precise points that I had been thinking about for the preceding months and helped move them from pure abstraction to exemplification. Gessius will come back a number of times in my remarks; but I should in fairness add that Professor Emmel is in no way responsible for what I have made of his analysis.