- The First Towns -
abandoned within a generation or so. The best documented cases of this are in the
Auvergne and in the Aisne valley. In the former case, at the time of the abandonment
of the open settlements such as Aulnat in the Grande Limagne, a plateau settlement
of some 40 ha was founded at Corent. This seems to have been occupied for about a
generation, and was replaced by a lowland site of about 35 ha on the river terrace of
the Allier at Gondole. With its massive earthen dump rampart of Fecamp type, this
site should date to about the time of the Roman conquest, and the finds from in and
around it belong to the decades just after 50 BC. It was a short-lived site, and around
30 BC a new defended site was established on the imposing spur of Gergovie. This
too was abandoned after a generation, with the foundation of the Roman town of
Augustonemetum on the site of modern Clermont-Ferrand.
In the Aisne valley the sequence is less complicated, and is most complete around
Soissons, centre of the tribe of the Suessiones. The earliest site in this case is a poorly
defended settlement of about 30 ha at Villeneuve St Germain (Audouze and
Biichsenschiitz 1992). This was replaced by the hilltop oppidum of Pommiers, with
Fecamp ramparts, itself replaced by the Roman town at Soissons. Villeneuve has
been extensively excavated, revealing a densely occupied planned settlement. One
peculiarity is that the interior was divided into four quadrants by cross-ditches
(Figure 10.3). In fact, these ditches may have been the substructure for a massive
roofed timber structure for which no parallels can be found. In part the division was
functional - one quadrant has palisaded enclosures (Figure 10.4), either high-status
dwellings or farm enclosures; another area seems to have been industrial. Limited
work on Pommiers suggests that it too had this internal division. The other major
excavation in the valley of the Aisne is the oppidum of Guignicourt. About 9 ha has
been excavated, revealing another planned layout on a rectilinear pattern, with a wide
range of different-sized houses and enclosures. With the exception of one or
two buildings which lie at an angle to the main orientation, and seem to belong to a
different period, the layout is unitary, and short-lived, as none of the houses was
repaired or replaced.
The sites in central and southern Gaul -Toulouse, Palais d'Essalois, Corent and
Mont Beuvray -are characterized by large quantities of wine amphorae. At Toulouse
they are found in large quantities in the puits funeraires, pits up to 10 m deep filled
with broken amphorae, and often containing other imported or prestige items such
as bronze vessels and helmets. Some contain cremation burials, but burial may not
be their only function. These quantities of amphorae have often been taken as
evidence that they were important trade centres, but rather we are seeing here
large-scale consumption of wine. Figures are not yet available, but the ratio of
wine amphorae to other pottery on these sites seems to be exceptional. In the case
of Mont Beuvray, the ancient Bibracte, it was an administrative centre where the
'senate' met, and though certainly it was a centre of industrial production, and
permanently occupied (Caesar himself stayed there over winter), the amount of wine
consumed is more than one would expect on an ordinary domestic site, suggesting
some sort of ceremonial or social/political activity which involved wine and food
consumption.
In many cases tribal territories can be approximately defined, so oppida can
be assigned to specific tribes. Some, like the Bituriges and the Helvetii, possessed