2. Background

2.1 Legacy survey data

Archaeological surveys have been conducted around the Mediterranean for more than fifty years and 'topographical' surveys have an even longer history (Cambi and Terrenato 1994, 21-30). This great time-depth of antiquarian and archaeological investigation means that the Mediterranean provides an enormous archive of legacy data. By comparison with the sophisticated methodologies implemented today, many of these surveys are methodologically naïve. Issues such as site definition were frequently unconsidered and rarely documented (for an example of some of the problems, see Witcher and Craven, in press). As a result, many legacy data comprise little more than symbols on a map, perhaps associated with broad dates of occupation and/or a list of the most interesting finds. It should, of course, be emphasised that it would be anachronistic to judge such surveys by today's standards. Many were pioneering projects, responding to new opportunities with innovative techniques. Subsequent critique and methodological developments have been fundamental, but these should not be used to condemn the methods of earlier surveyors.

The legacy of Mediterranean field survey therefore comprises large quantities of relatively low-quality data that do not provide the kind of detail and methodological rigour which many recent approaches demand. For example, without full quantification, it is difficult to assess variation in the consumption of material culture across spatial and temporal scales (Witcher 2006, 51). In many cases, it seems impossible to do more than visualise a few basic characteristics. So why persist when legacy data are badly compromised and when sophisticated new surveys can produce better data?

The answer lies partly in the changing nature of the surface archaeological record. Scatters of archaeological material in fields are usually ploughed-out of buried archaeological stratigraphy (though material may also derive from manuring; Bintliff and Snodgrass 1985). In many cases, excavation demonstrates extensive or even complete destruction of associated buried deposits. The surface scatters may therefore be the only surviving record of past activity. However, the long-term survival of the surface archaeological record is threatened. Agricultural processes gradually abrade sherds of pottery and expose fragile materials to the elements; scatters as a whole may be 'smeared' and washed down hillsides. The combined effect is to reduce both the ability to recognise scatters and to recover any useful material from them. Other threats include urbanisation, road building and quarrying. Although survey has often been presented as a repeatable exercise (in contrast with excavation), in reality it is akin to 'rescue archaeology'; indeed, John Ward-Perkins (1961, 1) explicitly initiated the South Etruria survey in response to the destruction of sites by agricultural intensification. In a gloomy assessment, Cherry (2003, 157) notes that the surface archaeological record will have disappeared altogether by c. 2050.

A slightly different problem is created by agricultural abatement. In the northern Mediterranean, declining subsidies and new environmental priorities mean that much land that was taken into production during the post-war years, is now reverting back to pasture or scrub and eventually woodland. As a result, there is a decline in the amount of land under cultivation and therefore less land of good surface visibility available for field survey.

Combined, the long-term degradation of the surface record and the decline of arable cultivation mean that legacy survey data form a unique resource that is growing, not diminishing, in importance. In many cases, scatters have permanently disappeared and are preserved solely 'by record'. Other scatters have been significantly degraded since their original recognition and though resurvey may employ more systematic methodologies, the quality of material available for collection is invariably much reduced (e.g. Di Giuseppe et al. 2002).

A new generation of archaeologists has therefore begun to recognise the value of old data. In particular, many of these datasets have never been subject to any detailed analysis. The Forma Italiae surveys in Italy are a good example. This ongoing series of surveys aims to provide a record of the archaeological landsdcapes of Italy; each volume covers a single 1:25 000 c.10x10 km mapsheet (Cambi and Terrenato 1994, 27-32). The primary concern of the series is to catalogue and map the distribution of sites for Cultural Resource Management purposes; earlier volumes made little attempt to analyse these results (e.g. statistical tests of distribution, inter-visibility studies, etc.; recent volumes have adopted more rigorous field methodologies and analytical techniques). The unexploited potential of these earlier surveys invites renewed interest.

New research questions have also encouraged interest in old data. For example, ongoing debate about the demography of Roman Italy has found the older regional site-based surveys (e.g. South Etruria, Albegna valley) more suitable for population reconstruction than the recent small-scale, 'off-site' surveys (e.g. Rieti Basin) (Fentress forthcoming).

Hence, the accelerating degradation of the surface record, the limited analysis conducted by many original surveys and the evolution of archaeological ideas have all led to a growing trend in 'second generation analysis' of legacy datasets (for this term, see Diacopoulos 2004).

PREVIOUS NEXT CONTENTS HOME