from the elite, which retained religious authority on this basis. From this point of view, Greek and Roman city-states were identical. The most perfect realizations of this model were located in the city-states of archaic and classical Greece and at Rome. Later, the notion of polis-religion being diffused together with that of the city-state, the history of ancient religion was implicated in the fall of the classical city. After the decline of the ancient city, religion would no longer be a form of conduct tied to the city-state but became one choice among many groups who offered their own doctrines, experiences, and discrepant myths.23 Here, again, one needs to nuance these claims. For one thing, it seems to me overly fast to date the decline of the city-state from the battle of Chaeronea in 338 BCE. This theory, which dates from the nineteenth century, has been successfully contested and overcome, first through the work of Louis Robert and Philippe Gauthier24 and by many other historians of the second half of the twentieth century, to the point where one could now qualify the Hellenistic Age as a golden age of the classical city. One might likewise point to a remark of John North, who in the course of his argument refers to the third century CE: one cannot be said to prove much about broad changes in the history of religion in preceding periods through interpretive claims about evidence of that late date.
Finally, the chief reproach voiced against advocates of the model of polis-religion is that they ascribe everything to politics and overstate the historical salience of Romanization as well as Roman resistance to foreign influences.
Alongside these supposed characteristics of Roman cults, further weaknesses and failures of the model of civic religion are enumerated as follows:
• It is difficult on the basis of such a model to account for the complexity of ancient religion. According to this critique, the model cannot explain why successive layers of deities and cults were not syncretized by the religious authorities, so as to realize a certain order and harmony within the pantheon. For the moment, let it suffice to emphasize that in a polytheist regime the collective has no need to rationalize its pantheon, because diversity is its raison d’être.
• The model does not leave space for other aspects of religion that were important for some, such as myth or popular cults like those of Silvanus or the so-called Mothers in Germania. The problem here is that, in the Greco-Roman world, myth was not part of religion. Moreover, there existed domestic and private cults (like that of Silvanus) or local cults (like that of the Mothers). There was no reason to introduce them into the civic pantheon of Rome. Once again, these cults were tied to their social context and to the very structure of polytheism.
• The model is not able to explain change. It responded to changes that took place within some Mediterranean koinê. Thus, if the changes arose principally in the private sphere, every perspective that marginalized non-public cults was perforce incapable of taking those changes into account. The difficulty here is that, even if a single one of such private cults turns out to have been the start of a religious revolution, the mode of analysis of the other private cults proposed by critics of civic religion offers no better means to understand the reasons for this revolution.
• Treating private cult as a secondary religious phenomenon does not allow one to explain why paganism remained popular even when public cult had been abolished. This fact is paradoxical only in appearance and can be easily explained: for a long time, private cult was not subject to the same prohibitions, and so it could endure after the promulgation of laws that in the first instance forbade only public cult.
• Greek myths antedated the city-state and were panhellenic; they were thus available to be used by the city-states that inscribed themselves within a framework greater than themselves. The great sanctuaries with which the city-states contended prove that cult sites were independent from their function within a specific city. Also, the oracles prove that there existed a religion superior to that of the polis. Finally, the temples of Asclepios at Epidauros and Pergamon, like the mysteries at Eleusis, were chiefly concerned with cult as celebrated by individuals. On these varied points, it is necessary to observe that the existence of federal cults or oracles external to the city is a banal given in the world of city-states. Nonetheless, one should also note that in the framework of panhellenism, it was not the Greeks who existed in federal union, but the Greek city-states, and moreover the cults and oracles functioned in a fashion closely homologous with the model of civic religion.
• Pilgrimage is also invoked, as well as human mobility, which would have led to the existence within city-states of numerous non-citizen inhabitants or metics. Because polis-religion is based on citizenship, it would have been less and less able to integrate the totality of religious actors, since these did not have citizenship. We will return to this question.
• There would have been a gap between public and private cults, and it is this gap that would have been the reason for change. It is through this gap that foreign cults would have inserted themselves. The very idea of such a gap or lacuna itself poses a problem. It seems to me to amount to the importation of an anachronistic “religiosity” rather than to the detection of a historically verifiable dissatisfaction experienced by participants in ancestral religions.
Overall, this series of claims is thought to demonstrate that, whatever its strength, polis-religion never in fact fulfilled its function, neither in the age of the archaic and classical Greek city, nor, certainly, in the age of the Hellenistic and Roman cities. It was a concern of elites, who tried thereby to impose their domination on the lower classes, without ever successfully colonizing their private lives. Not only would there always be cults outside and beyond the city, but individuals would have possessed religious activities more dynamic and enduring than those of the city. In sum, one has the impression that, below political and social institutions, existed Religion-with-a-capital-R: not the religion of Zeus, Jupiter, Minerva, and their companions, but the Religion of Cybele, Mithra, Isis, Aesculapius, and of the God of the Jews and Christians, or the gods of the Gauls, who were thought to address themselves to individuals. There was thus an opposition between religion and Religion, in such a way that the latter spoke to the individual and was not the product of some manipulation of the crowd nor a means of social and political control.
“Religiosity”
Clearly, behind all these separate criticisms lies the notion of religiosity, whose status as a universal was refuted by Georg Wissowa. It appeared in the Protestant context of early nineteenth-century Germany.25 It refers to the subjective dimension of Christian religious experience, marked by its interiorization. This experience is opposed to objective, exteriorized religion, which (according to its theorists) finds expression in the institutions and dogmas of the Catholic Church.26 This insistence on the feelings and emotions of the individual, on individual perception of the infinite, is memorably expressed in the work of Friedrich Schleiermacher on Christian faith, published at the start of the nineteenth century.27 Effectively, he placed primary value on the private religious experience of the individual contemplating the universe. This contemplation triggers and shapes religious emotion, to the extent that by this means the individual recognizes and feels supreme order and his or her own absolute dependence in spiritual matters on the divine creator who animates everything. In other words, piety in itself is neither a knowing nor a doing, but a feeling and certainty of one’s dependence.28 This foundational principle eliminated as much as possible any role in religion for religious institutions. The individual could thus take himself to an existing church or just as easily to one he himself created. Indeed, in Schleiermacher’s theology there is a sense in which even the biblical tradition lost its central importance, since only pious feeling counted. This is a paradox in Schleiermacher’s thought that theologians have debated extensively, but it is not what interests us here.
Schleiermacher’s approach has shaped the study of antiquity since Hegel and his successors. One example among many is Richard Reitzenstein, a professor of philology at the University of Strasbourg who tried to interpret the Judeo-Christian tradition commencing from pagan antiquity, which is to say, commencing under the empire, as a monotheist religion of a savior originating in the Orient.29 Reitzenstein identified in Roman antiquity, in the Secular Games of Augustus, for example, the birth of a new form of religion, a “religiosity” that he imagined as an interiorization of religion. Observe right away that nothing connected to this