This study has explored the problems and potential of using legacy survey data within a GIS environment. The emphasis has not been to promote a single analytical framework, but rather a more reflexive approach with which to supplement the existing diversity of applications.
Instead of using GIS as a convenient mechanism for simplifying and comparing data, this study has attempted to use GIS to explore the unevenness of legacy data and to generate contextual metadata as a substitute for more formal records. It has been argued that all datasets are artefacts of the methodologies used to collect them. If legacy data are to be used with any degree of sophistication, it is vital to consider how the conceptualisation and execution of fieldwork has influenced results; often these methodological decisions interact directly with the nature of the archaeological record. For example, a transect survey of a nucleated settlement pattern may achieve rather limited results. With legacy data, often questions arise which cannot be answered because of data collection methods. For example, is the lack of early imperial settlement in the Biferno valley due to rural poverty, settlement nucleation or population decline? The decision to 'grab' sample artefacts makes it difficult to assess which of these possibilities is the most plausible. However, by understanding the limitations of such data, it is possible to develop practicable field strategies for targeted resurvey to resolve such issues and bring about meaningful comparison (for an example, see Di Giuseppe et al. 2002).
Legacy survey data are not objective archives of facts and figures; they require careful and critical use. When the data from one survey are to be compared with another, these issues become even more significant. When the data come in a pre-digitised format, there is a real danger that these differences can be underestimated if unaccompanied by a wealth of formal metadata. If GIS is to become part of the solution, rather than potentially making the problem worse, then it is important to acknowledge that GIS is not a neutral tool but an affective approach (Gillings 2001, 109). It is an environment in which the data are transformed, interpreted and reinterpreted. The application of GIS should be an iterative process which focuses on enriching knowledge of data structures - with equal emphasis on the role of archaeologists and the behaviour of people in the past. Letting legacy data 'speak for themselves' risks failing to acknowledge the very real impact of the archaeologist in creating the data - whether the original surveyor or subsequent generations of archaeologists who will increasingly make use of them.
GIS is already established as an integral part of the analysis and interpretation of survey. Legacy data have been heavily used since the emergence of archaeological GIS in the early 1990s and their use continues to grow. This article argues that, alongside sophisticated modelling and analytical methods, the processes of digitisation and visualisation should be recognised as integral to the improved understanding of legacy data. Small-scale, flexible and reflexive GIS environments have much to offer the longer-term aim of large-scale GIS comparison of settlement, economy and population across the Mediterranean. An intermediate step is to explore the opportunities provided by overlapping and adjacent surveys for identification and refinement of basic techniques of data characterisation. The collation of formal metadata for legacy survey data is already underway; what is needed now is consideration of the other types of metadata that can be extracted, and further thought about how these can be documented, shared and used.
© Internet Archaeology/Author(s)
URL: http://intarch.ac.uk/journal/issue24/2/5.html
Last updated: Mon Jun 30 2008