participants in a network plays an active part in the effectiveness of
information transmitted.^29
In the north Aegean area, there is a wealth of data that can be exploited
to understand how social networks developed in the second half of the
first millenniumbc. Coin distributions indicate the non-random pattern
of economic networks, confirming literary and epigraphic references to
commercial agreements between specified partners (see further Chapters
4, 5, and 6). This allows a more coherent view to emerge of those
relationships that evolved as historic partnerships, with a marked degree
of longevity. Among the longest-lived of these partnerships was that
between the inhabitants of the island of Thasos and a wide network of
contacts, Greek and non-Greek, across the east Balkans.
The recent spate of research on social networks has drawn attention
away from ecological concerns, expressed most ambitiously by Horden
and Purcell’sThe Corrupting Sea. Whilst being much quoted, this sem-
inal study has been less influential on the historical analysis of individual
regions than might have been expected.^30 The canvas is far larger than
most historians and archaeologists are comfortable with, which has
discouraged imitators.^31 Yet few commentators have acknowledged
that Horden and Purcell’s work has changed the way that historians
conceive the ancient classical world as explicitly as Ian Morris has
done.^32 Not all Horden and Purcell’s concepts and arguments have
been equally convincing. Their rejection of cities as subjects of analysis,
and of concrete agglomerations in general, has been among the least
persuasive of their theses, despite the important arguments that they
have marshalled about the‘labile’urban fabric of much Mediterranean
city life. This is a topic to which I will return below.^33
Although they deny an‘interactionist’approach, there are aspects of
Horden and Purcell’s thesis that are entirely compatible with studies
inspired by Network theory. The idea of‘dispersed hinterlands’and the
notion of‘connectivity’itself belongs to the portmanteau of terms that
(^29) Ball 2004, 466, commenting on Duncan Watts’repetition of the letter-posting experi-
ment conducted by Stanley Milgram that demonstrated the‘six degrees of separation’
between human subjects (Watts 2003, 130–9, 154–61).
(^30) The most substantial engagement with Horden and Purcell’s conceptual framework is
by the contributors to Harris 2005. Among the many reviews I would single out are those of
Shaw (2001) and Algazi (2005); see also my remarks in Archibald 2005a, 6–8.
(^31) See now Broodbank (forthcoming).
(^32) Morris 2003; idem 2009b, 79–80.
(^33) Horden and Purcell 2000, 90–105, on the problems of defining settlement forms.
Societies and economies 97