ABSTRACT
This chapter presents some of the issues that arise in constructing network models for archaeology that rely heavily on the basic attributes of physical geography: distance and topography. It argues that network models can provide the natural framework to transcend this dualism between physical and relational space. The arrival of sail in the Aegean in the Middle (MBA) makes the Aegean a 'Goldilocks' sea in the sense that the technology is just right for the appearance of an active trading network with a handful of key sites. Thus MBA networks live in the transition between social organization being strongly influenced by geography and it being much less relevant. Island archipelagos lend themselves to Proximal Point Analysis (PPA) network analysis because of the clear separation of sites. Broodbank suggests that the EBA Cyclades were agriculturally marginal, requiring social storage networks. The first optimization approach looks to identify the best network with the most likely network, all other things being equal.
