The Green River region of western Kentucky has played a significant role in the development of eastern North American archaeology since C. B. Moore's explorations there in the first decades of the twentieth century (Moore 2002). Throughout the 1930s and 1940s, William Webb and colleagues synthesized the results of massive Works Progress Administration (WPA) excavations at several western Kentucky middens into extensive trait lists that provided an empirical basis for the definition of an eastern Archaic pattern (e.g., Webb and Dejarnette 1942:319). Currently, Green River sites are an important focus of a debate concerning the origins of complexity among Middle to Late Archaic (ca. 6000 to 3000 B.P.) hunter-gatherers in eastern North America (e.g., Crothers and Bernbeck 2004; Sassaman 2004). This paper continues this debate by investigating interaction and social integration along the Green River through a multiscalar analysis of fishing technologies. Although focusing on the manufacture of fishhooks at three sites - Baker, Chiggerville, and Read- - variability in the technological choices made in the production of these fishhooks is placed within a macroregional context in order to infer the formation of distinct regional groups during the Middle to Late Archaic. The existence of such regional groups, suggested through earlier analyses of bone pin and atlatl weight styles (Burdin 2004, Jefferies 1997), supports the hypothesis that hunter-gatherer local groups were becoming more socially integrated and, as a result, more complex during this time.
As Sassaman (2004:231) points out in his recent review of complex hunter-gatherer research, the term "complexity" has no standard definition in archaeological discourse. To define a social group as complex in an absolute sense is impossible given the degree of variability found among social systems and the fact that political, economic, and other components of culture vary, to some degree, independently of one another. As a result, Sassaman defines complexity as "a relative measure of the number of parts in a system and number of interrelationships among those parts" (Sassaman 2004:231; emphasis added). This definition is similar to that adopted by Service (1971:173; see also Price and Brown 1985:7), who based his typology of forms of social organization on the manner in which increasing numbers of parts of social systems were integrated. The recognition that complexity is a relational concept is important in that it requires researchers to take a comparative approach to any evaluation of complexity and to specifically define which components of culture are to be addressed in any discussion of the concept.
Recently, much of the debate concerning huntergatherer complexity in the Southeast has focused on the political systems of Middle and Late Archaic moundbuilding societies in the lower Mississippi Valley and Atlantic Coastal Plain. For instance, Sassaman and Heckenberger (2004a, 2004b) cite geometric uniformity in the spatial configurations of many Archaic mounds in Louisiana as evidence for the existence of ranked kinship groups, while Russo (2004) interprets the spatial structure, composition, and distribution of shell refuse at Archaic shell rings along the Atlantic Coast as evidence these societies were transegalitarian huntergatherers characterized by status differentiation and unequal relations of economic influence. Alternatively, Gibson (2004) argues that mound building in the lower Mississippi Valley was a means of integrating society around a common ideology rather than a source of power for specific individuals, and Thompson (2007) argues that at least some coastal shell rings are the result of the gradual accumulation of household refuse in a circular pattern. These differences in interpretation among leading scholars illustrate the nascent character of Archaic hunter-gatherer complexity research in the Southeast. Much work remains to be done before a consensus as to the form and political-economic implications of hunter-gatherer social systems can be reached. …