most inscriptions were not found in situ, that is, in the place where they were erected
in ancient times. For excavations to reveal a higher number of inscriptions in
their original place is the exception and not the rule (see e.g. Derks 1998: 83;
Spickermann 2003: 14ff.). Normally we find dedicatory inscriptions (as all other inscrip-
tions) in a secondary or even tertiary context – as part of a town wall, built into
medieval churches, etc. Thus, it is often impossible to combine archaeological and
epigraphical data. But in many ancient sanctuaries not only one god was venerated
but also theoí s ́ynnaoi (e.g. ILS3536). Therefore it is sometimes difficult to deter-
mine the principal god of a sanctuary known only by an archaeological excavation.
If neither the central inscription of the temple itself was found nor a greater num-
ber of dedications to a certain god, but only one or two altars, there is no certainty
that the god addressed was the principal one (see for example the so-called temple
of Aesculapius at Augusta Treverorum: Haensch 1997: 76; Scheid 1998d).
Finally, two points should be made. Despite the fact that probably more inscrip-
tions were lost through the ages than we will ever get knowledge of, cases are not
so rare where two or three dedicatory inscriptions were put up by one and the same
person. As for the beneficiarii, we find among the soldiers responsible for about six
hundred and fifty dedications more than thirty who erected two or more arae(Nelis-
Clément 2000: 31). We cannot determine whether this phenomenon was typical of
other people to the same degree because we cannot identify in their case groups
of persons and – consequently – individuals in the same precise manner. But there
are a number of examples – e.g. CIL 12 .4849 (Tusculum); ILLRP^2 64, 237 (Veii);
ILS3424 (Rome), 3561, 3767 (Verona), 3747a/b (Pola, in a fullonica), 3886, 4253
(Aquincum), 3268a/b; and CIL3.1096 (Apulum) – which seem to suggest that
our epigraphical tradition was influenced at least to some extent by a quite small
number of people who were especially impregnated with the epigraphical habit. But
on the other hand, one should not underestimate the number of people who were
sufficiently literate to be interested in such written forms of ritual acts. The people
who frequented the sanctuaries of Pesch and Morken-Harff, Monte de Focho, and
Thignica were apparently simple peasants living in a rural country and being only
marginally influenced by “city life.” What is more, the same can be observed in the
case of the users of the tabellae defixionum(see the next point) and the authors
of the “Beichtinschriften” (Petzl 1994). Apparently, much depended on the local
situation influenced by many factors which we will never know.
Curse Tablets (tabellae defixionum)
One of the most important sources for ancient magic is the so-called curse tablets.
These are inscribed pieces of lead, usually in the form of small, thin sheets, rolled
up into scrolls or folded into small packets (but in Roman times not pierced by a
nail), intended to influence, by supernatural means, the actions or welfare of per-
sons or animals against their will (Jordan 1985: 151; cf. Jordan 2001: 5). They were
left in water, in chthonic sanctuaries, and in tombs. In consequence of careful excava-
tions and the widespread use of metal detectors, the number of detected tablets isInscriptions 185