Environmental History in Ostrogothic Italy 395
on the coast or in well-watered plains proved more susceptible to population
losses than inland places and hill settlements, which for once benefited from
their relative isolation and more sluggish communications and may have been
healthier.15 Similarly, an exception should be made for cities through which
Eastern Roman government functioned, such as Ravenna (never a crowded
place) or Otranto, whose importance to Aegean—Adriatic connectivity raised
its late antique profile.16
But if diversity is this period’s leitmotif, Italian cities share some common
demographic elements. Even though other cities were not, like Rome, depen-
dent on an artificially enhanced hinterland (virtually the entire Mediterranean,
not just the Italia Suburbicaria Constantine assigned to supplying Rome),
archaeologists find that during Late Antiquity Italian urban communities built
much less, occupied less space, and created consumption centres that were
reduced by comparison to the earlier centuries of the millennium.17 Deducing
from this, the shrinkage of urban populations is one general story we can tell
about late antique Italy.
For rural populations, the extraordinary results of decades of field surveys
provide a less unanimous verdict. Post-classical pottery shards, the usual indi-
cator of settlement (and hence of population) can be difficult to identify or
date. In addition, even without ceasing to exist people can resort to more
perishable containers (e.g. of wood) that elude archaeological surveyors.
Moreover, south Italy and particularly Sicily seem to have enjoyed more stabil-
ity than the rest of Italy.18 But 5th-century shards and more substantial proxies
suggest lessened number, size, and economic complexity of rural sites in much
of Italy, and this is generally thought to be a sign that fewer people inhabited
the countryside than when Roman villas had structured that space.19 All told,
archaeology reveals a rural Italy as depopulated in the 5th and 6th centuries
15 Christie, Constantine to Charlemagne, p. 426. Much depends on the significance assigned
to plague and malaria: for example, Harrison, “Plague”, pp. 30–2 provocatively argued that
cultural and economic choices, not demographic change, lay behind landscape and set-
tlement change. Meanwhile, Staffa, “I centri urbani,” pp. 449–51 gives examples of coastal
towns in Abbruzzo doing far better than isolated inland communities.
16 On late antique Otranto see Wilkinson, “Summary and Discussion”, pp. 41–58; Zanini, Le
Italie bizantine, pp. 116–17.
17 With late 20th-century Hong Kong as an example, it is perfectly possible for numerous
vacant lots to coexist with burgeoning urban life and population: Boyden et al., Ecology,
p. 113.
18 Volpe, “Villaggi”, pp. 426–35; Cacciaguerra, “Dinamiche”, pp. 441–6.
19 Lewit, “Pigs”, p. 79, warns against reading too much into archaeological invisibility. See
Chapter 10 in this volume for a more cautious evaluation of Ostrogothic demography.