Since our interest here is in the movement of gods and cults, and not in imperi-
alism or foreign gods or a specific rite, we can bracket for the moment issues of con-
text and simply adduce several further examples of gods who moved, or declined to
do so, in response to Roman requests. Perhaps the most famous such god is Cybele,
the Great Mother, an Anatolian goddess whom the Romans were told to bring from
her shrine in Asia to Rome in the final years of the Hannibalic War (Beard et al.
1998: 96 – 8). Two extended literary accounts of her transfer survive, by Livy and
by Ovid, and they share several features relevant to this argument. I single out two.
First, both authors identify the goddess with her baityl – that is to say, each slips
unproblematically between references to “the stone,” “the image,” and “the god-
dess” (Livy 29.11.7, 14.10 –14; Ov. Fast.4.317). Second, both authors imagine that
the adoption of a new cult requires the presence of the god, and each supposes that
in this case, at least, the goddess somehow is a particular cult object. As a con-
sequence, we might conclude that the location of the cult object fixes the location
of the goddess, and restricts her worship to that site. This would, I suspect, be too
narrow a conclusion. But that it is conceivable amounts to an urgent caution against
naturalizing the notions attendant upon “spreading”: that one can worship a god
wheresoever one chooses, regardless of wherever else that god is worshiped; and, for
that matter, that gods desire more worshipers, from more places, races, and walks
of life.
Comparable in its focus on the will of a god – in this case, on the god’s desire
not to move – is the story of the god Terminus (“Boundary”), who was asked to
leave his temple in order to clear ground for a new temple to Jupiter. The sources
differ whether the ritual used was evocatioor exauguratio, the latter a ritual by which
his temple would have been desacralized in respect of him, freeing the land to be
rededicated to another purpose. They agree, however, that Terminus declined, and
so remained where he was (see esp. Festus s.v. Nequitum 160 L, citing Cato’s Origines
frag. 24 Peter; Livy 1.55). Further examples might easily be adduced; but the point
is so far established.
The other substantial body of evidence on Roman religion’s commitment to a
particular sacred geography derives from Roman writings on religious law. That
literature, which must now be reconstructed from scattered allusions and quotations,
spans several centuries and embraces what were taken to be several distinct bodies
of law, each of which underwent internal change over time and yet differed, the one
from the others. Their specific content, and the content of those changes, are not
at issue here. I wish only to stress how many were the ways in which those bodies
of law figured the city of Rome as ritually and religiously distinct.
So, for example, according to Varro, augural law provided for the division of land
into five categories: Roman, Gabine, alien, hostile, and indeterminate (Varro, Ling.
5.33). In continuing to provide specifically for Gabine land, augural law appears to
have been ossified at an early stage in Rome’s development – Gabii was allied to
Rome in the early fifth century but more or less disappears from the historical record
in the fourth century bc. But what is important here is not so much the specific dis-
tinction drawn, but that it was drawn at all. Similarly, for the taking of auspices, it
was crucial that holders of imperiumdistinguish between civilian and non-civilian442 Clifford Ando