the donors’ status not only at the moment of the dedication but also at the time
when the vow was undertaken – as for example the libertuswho fulfilled a vow made
as a slave (ILS3491), or the senior clerk in the staff of a Roman procurator (cor-
nicularius procuratoris) who had made his vow as a recruit in the moment of being
sent to a war on the Crimea (quot (!)tiro proficiscens in bello Bosporano; AE1991,
1378). But it has to be emphasized that these people are only a part of those who
venerated a god and we cannot even approximately determine the percentage. Only
two things are certain: it was only a (very?) small percentage, and it was not a rep-
resentative sample, because dedicatory inscriptions required money and an interest
in this Greco-Roman form of cultural act. Even people who could read and write
did not all belong to this group. Votive inscriptions are unevenly distributed through-
out the Roman world geographically and especially chronologically (e.g. Derks
1998: 83). We can observe larger groups of adherents of a certain cult only in those
cases where a whole religious community paid for something (ILS3082, 3609, 3840,
5466, 5470).
Were all these hints of posts and honors motivated by the efforts of self-
representation? That this was not the only motive at least in some cases has been
shown by Eck (1989: 32ff.). He pointed out that Roman governors and other
dignitaries mentioned their priesthoods only in two contexts – if a whole cursus
was given, and in religious contexts (that is, if they dedicated a temple or erected
an altar). Apparently, these senators did not consider their priesthood just as an
honorary post, but acknowledged at least to a certain degree religious elements in
these functions.
Two problems arise with regard to the names of the gods. In a provincial context
- and 90 percent of the Roman empire was provinces – we can almost never exclude
the possibility that behind a god of a Greek or Roman pantheon an indigenous god
lurks who had been assimilated by the process of the so-called interpretatio Romana
(or Graeca) to a Roman or Greek one with some similar characteristics (for a com-
prehensive discussion of the problems connected with the interpretatio Romana
see Derks 1998: 94 –118). But it is also impossible to exclude the possibility that
the Roman deity was accepted by the local people, imitating the beliefs of Roman
dignitaries such as the governors.
It is a clear-cut case of a local god if his name is derived from a local language,
as for example the Dea Nehalennia (ILS4748ff.), Bindus (ILS4878), the Dii Magifae
(ILS4493), Bacax (CIL8. 18828), or the Deus Turmasgada (ILS4073– 4). It is also
possible to identify a local god where there was a standard equivalence – as for ex-
ample in the case of Mercurius and the Celtic god Teutates – or where the equivalent
chosen is a strange one – as for example in the case of a Roman veteran from the
Hawran who venerated [the]ô.Lykoúrgo.(IGRR 3. 1294, cf. Stoll 2001: 467). These
local gods, like all dei patrii, were without doubt for the majority of the population
of the empire much more important than the gods of the Greek or Roman pan-
theon (and the gods of the mysteria). However, in all these cases, we know almost
nothing about the rituals during which these inscriptions were erected. And even if
the so-called interpretatio Romanatook the form of an expressive paralleling by using
the indigenous name as an epiclesis, as for example when our examples combine Bindus
Inscriptions 181