Special Section Dynamic Landscapes and Socio-Political Process: The Topography of Anthropogenic Environments in Global Perspective

Article excerpt

Landscape archaeology: towards a definition

Sander Van Der Leeuw, in his recent plenary address at the 1998 Society for American Archaeology Meetings, suggested that archaeology as a discipline has moved its emphasis from site to settlement pattern, and now to the landscape. Though a landscape focus is not new, especially for the social sciences (Coones 1994; Cosgrove 1984; Glacken 1967; Jackson 1994), the landscape approach in archaeology (Wagstaff 1987) is still in its infancy. Landscape research varies widely from simple environmental reconstruction, to the systemic/scientific approach of Rossignol & Wandsnider (1992; see McGlade 1995), to historical ecology (Balee 1998; Crumley 1994; Crumley & Marquardt 1987; Kirch 1997; see Whitehead 1998 for a critique) to the phenomenological perspective of Tilley (1994) and Bender (1992; 1993) to the landscape archaeology of Ashmore & Barnard (1998), Bradley (1998a; 1998b) and Erickson (1993). Thus there is a wide variety of approaches that share certain key elements but lack a unifying metaphor. This is exacerbated by what Bender has termed the 'Atlantic void' between American and European landscape archaeology (Bender this volume). Thus one can ask, what exactly is landscape archaeology? This section was conceived as a way to begin to answer this question with global case-studies.

At the outset let us stress that landscape archaeology is not a paradigm shift that will replace processual archaeology. It is, instead, an outgrowth of regional-scale archaeological research focused on the human/environment dialectic (Crumley & Marquardt 1987), an area of inquiry long important in archaeology (Trigger 1989: 279-303). A landscape analysis is complementary to traditional forms of archaeological research. As with all archaeological investigation, the decision to apply a landscape approach is question dependent. Landscape archaeology is especially well suited for problems that elucidate our critical, shared connection to our physical and cognitive environment. By 'connection', we mean the manner in which human social, political and economic systems interact with, and are the result of, intentional strategies of landscape manipulation. By 'environment', we mean the humanly-built and conceived results of these strategies; the anthropogenic landscape, always in flux, never static. Thus the term landscape can be defined as 'a unit of human occupation', something akin to the Dutch progenitor of the term landschap (see Cosgrove 1984: 13-39; Schama 1995: 10). Landscape in this sense is a broad, inclusive, holistic concept created intentionally to include humans, their anthropogenic ecosystem and the manner in which these landscapes are conceptualized, experienced and symbolized.

By definition, landscape archaeology is a holistic, multi-disciplinary endeavour, and it is no accident that several of the contributors to this section are from academic disciplines outside archaeology. The best landscape research draws on recent advances from many complementary theoretical perspectives. Environmental history questions our notion of what is nature and the historicity of landscapes (Cronin 1996; see also Descola & Palsson 1996; Ingerson 1994; Thomas 1996: 20-29). New ecology calls into question notions of environmental equilibrium and systems ecology in favour of dis-equilibrium (Dotkin 1990; Zimmerer 1994; 1998); and political ecology yields a ready-made theoretical paradigm for recursively linking humans to their environment (Blakie & Brookfield 1987; Bryant 1992).

The papers that compose this section are a sample of the wide variety and theoretical orientation of those applying a landscape perspective. Though this research varies widely in time, space and focus, it is held together by three unifying themes. The first is the recognition of a dynamic, accretionary, humanly-constructed and maintained environment. The second is the conception of this landscape as a historically contingent entity. …