Internet Archaeol. 16. Phillips. Introduction

1.0 Introduction

[The European] approach to regional landscape analysis [in GIS] ... is concerned firstly with relationships between sites in terms of territories or spatial units representing social, political and economic interactions (Lock and Stančić 1995, xiii-xiv).

It is this landscape analysis of territories within a GIS platform that I wish to address in this article. In particular I will argue that the definitions of landscape (the horse) should inform the particular set of analytical models used in GIS (the cart).

Justification for this statement arises from the fact that spatial analysis, throughout its forty-year history, has been marked by a series of unresolved problems. Some of these problems will be highlighted through the examination of an ethnographic example: New Zealand Māori land-use systems were not consistent with many of the assumptions inherent in the GIS models. Particularly, the notion that the topography is inert will be contrasted with the contextual archaeology definition of landscape.

It will be maintained that creating diachronic environments and settlement histories from multi-disciplinary sources should be routine procedure in spatial analyses, from which to test the archaeological record. Moreover, where ethno-historical information is available the analyses should reflect cultural norms. Finally, it is proposed that the newer applications, such as dynamic modelling and simulation in conjunction with fuzzy logic within a landscape approach, hold better promise of resolving the existing problems in GIS analyses.