RELIGION 203but each in its own way: ‘civic’ religion emphasizes the socio- political dominance of
a specifi c group (Rives 1995), ‘polis’ religion the coherence of a local religious system
within a given cultural order. Doubts about the adequacy of such models are the
inevitable consequence of thinking of religion in terms such as discourse, pragmatics,
communication and media of representation (Bendlin 1997). Those who work on
provincial religion, especially in the north- west provinces, where cities and municipia
were thinly spread, and landscapes suffused with divine presences, are acutely aware
of the limits of either model (e.g. Woolf 1997, Derks 1992 and 1998, Dondin-Payre
and Raepsaet-Charlier 2006; cf. Schäfer 2007 on Sarmizegetusa in Dacia). Something
similar can be said about Anatolia (Dignas 2002, 223–78, Ricl 2003, Rostad 2006,
Schörner 2011), Syria (Kaizer 2008; cf. Belayche 2003) and Egypt (Frankfurter
1998). More ambitiously, it has been argued that civic rights in antiquity were not
co- terminous with religious participation, even in the case of Judaism (Krauter
2004). That claim has been harshly dismissed, at least for Rome (Scheid 2013, cf.
2005a), but it seems obvious that the power of elites over the legitimate religious
order, and their interests in sustaining it, produced considerable disparities in actual
religious practice in cities, to say nothing of rural areas, where perhaps eighty- fi ve
per cent of the total population lived and suffered (Gordon 1990, 2008; cf. Mellor
1992, Richardson and Santangelo 2011). The choice is basically between a
Durkheimian view of religion as a collective- normative enterprise and a Weberian
view of interests, power and information. Related to this issue is the differential
circulation of religious knowledge and attempts to defi ne, control or limit it, above
all in relation to divination and ‘magic’, i.e. overtly instrumental religious practice
(Phillips 1986, Potter 1994, Fögen 1993, Dickie 2001, Trzcionka 2007, Monaca
2009). A quite different aspect of divination, its possible value in processes of
individualization, is the theme of Rosenberger (2013).
We can summarize much of all this under the rubric of the pragmatics of religion.
A major advance here has been the emergence of a more sophisticated archaeological
practice, particularly in the provinces. Pride of place must here go to an innovative
study of everyday religious practice in Pompeii and Herculaneum, in which plans and
images play an essential role in communicating the argument (van Andringa 2009),
but we can also register notable advances in the presentation of sacred areas in Ostia
(Rieger 2004), and of Isiac temples (Kleibl 2009), and an interesting attempt to
present the archaeological evidence for the same cult in the form of a commented
atlas (Bricault 2001). After decades of neglect, the practice of votive- offering, central
to Greek and Roman religious practice, has begun to be re- conceptualized (Bodel and
Kajava 2009). Increasing awareness of the importance of landscape has stimulated
a project, not limited to the empire, of inventorizing and mapping all religious sites
in Italy (Scheid et al. 1997), parallel to, but more specifi c than, the series Carte
archéologique de la Gaule organized by the Académie des Inscriptions (URL:
http://www.aibl.fr/travaux/antiquite/article/la- carte-archeologique- de-la- gaule?lang=fr),
which was re- launched in 1988. Another result of improved archaeological method
is the increasing historicization of funerary practice, both in tracing mortuary ideals
(Riggs 2005, Brink and Green 2008, Rüpke and Scheid 2010, Ameling 2011) and in
providing quite extraordinarily detailed insight into social practice in relation to the
dead, in this case in part of the Porta Nocera necropolis outside Pompeii (van
Andringa et al. 2013). By contrast, the use of proximal point analysis as a means of
representing, if not yet explaining, the spread of religious ideas, runs up against
serious theoretical objections, even among network theorists (Collar 2013).